A DYNAMIC PROGRAMMING MODEL FOR NETWORK SERVICE SCHEDULING

Jesuk Ko

Video on Demand (VOD) is one of the most promising services in Broadband Integrated Services Digital Network (B-ISDN) for the next generation. VOD can be classified into two types of services: Near VOD (NVOD) and Interactive VOD (IVOD). For either service, some video servers should be installed at some nodes in the tree structured VOD network, so that each node with a video server stores video programs and distributes stored programs to customers. Given a tree-structured VOD network and the total number of programs being served in the network, the resource allocation problem in a VOD network providing a mixture of IVOD and NVOD services is to determine where to install video servers for IVOD service and both IVOD and NVOD services. In this paper we develop an efficient dynamic programming algorithm for solving the problem. We also implement the algorithm based on a service policy assumed in this paper.

A MULTI-AGENT HOME AUTOMATION SYSTEM FOR POWER MANAGEMENT

Shadi Abras, Stéphane Ploix, Sylvie Pesty and Mireille Jacomino

This paper presents the principles of a Home Automation system dedicated to power management that adapts power consumption to available power ressources according to user comfort and cost criteria. The system relies on a multi-agent paradigm. Each agent is embedded into a power resource or an equipment, which may be an environment (thermal-air, thermal-water, ventilation, luminous) or a service (washing, cooking), and cooperates and coordinates its action with others in order to find acceptable near-optimal solution. The control algorithm is decomposed into two complementary mechanisms: an emergency mechanism, which protects from constraint violations, and an anticipation mechanism, which computes the best future set-points according to predicted consumptions and productions and to user criteria. The paper details a negotiation protocol used by the both mechanisms and presents some preliminary simulation results.

CONSCIOUSNESS FOR MODELING INTELLIGENCE - Simulating the Evolution by Closure to the Inverse

Tudor Niculiu, Cristian Lupu and Sorin Cotofana

Intelligence = Consciousness x Adaptability x Intention and Faith = Intuition x Inspiration x Imagination, are the complementary parts of human mind. Conscience = Consciousness x Inspiration is the link between. The subtitle refers to an interpretation of Way, Truth, and Life. It is a strategy of development. Conscience simulation demands transcending from computability to simulability, by an intensive effort on extensive research to integrate essential mathematical and physical knowledge guided by philosophical goals. A way to begin is hierarchical simulation. Coexistent interdependent hierarchies structure the universe of models for complex systems, e.g., hardware - software ones. They belong to different hierarchy types, defined by simulation abstraction levels, autonomous modules, classes, symbols, and knowledge abstractions. Applying Divide et Impera et Intellige to hierarchy types reveals their importance for intelligent simulation. The power of abstraction is the real measure for the human mind. Turning the abstraction into comprehensive construction could be the aim of humanity, the unique God for different cultures of free humans. The way to freedom is by understanding necessity. We have to recall our conscience, to reintegrate our mind, and to remember that society has to assist humans to live among humans, not to consider that they only have to work for it. An operating system serves the autonomous programs, both for the function of the hard and for development of the soft. The society has to be reasonable, to assure health and education for every human, and to encourage search and research for every conscient human.

FAULT MAINTENANCE IN EMBEDDED SYSTEMS APPLICATIONS - Multiple Lift Control System as Safety Critical Embedded Application

Miroslav Sveda and Radimir Vrba

This paper describes principles of a designed multiple lift control system based on a dedicated embedded architecture. After reviewing dependable concepts and design method used, the main attention is focused on the hardware architecture, software, and communication services and protocols fitting the application requirements. The multiple lift control system presents in this case a real-world solution of a safety critical embedded system application. The design employs fail-stop safety model and dedicated distributed architecture to meet application constraints efficiently. The paper stresses those features that distin-guish the real project from a demonstration case study.

A NEURAL NETWORK-BASED SENSOR FOR ELDER FALLING DETECTION

Jiann-I Pan, Cheng-Jie Yung and Chung Chao Liang

Falling down is going to be a crucial problem to an elder today. In many countries, unintentional injury was being one of the leading causes of death in persons over age 65 years. As many of the elder are live alone on their own and because of the isolation, it is necessary to design an intelligent and sensitive falling detector for the elderly people. In this paper, we present an intelligent and portable fall detection device based on artificial neural network technology. This fall detector consisted of two main components: accelerometer and micro controller. The tri-axis accelerometer is used to continuously measure the variation of elder’s 3 ways acceleration. The micro controller reads the signals from the accelerometer and then through a back-propagation neural network model to perform the fall activity recognition. This device is integrated in a small box which can be holding on the belt for elder.

DATA FLOW FORMALIZATION

Thouraya Bouabana-Tebibel

The Object Constraint Language OCL is an extension of the UML notation for the expression of restrictions over the static and dynamic diagrams. We propose to take advantage of its formal capabilities for validating whether the UML model matches with the system properties. For this purpose, we develop an approach based on Petri nets and temporal logics. This approach allows the integration of the temporal logic properties translated from the OCL invariants with the Petri nets obtained from the UML modeling. A case study is given throughout the paper to illustrate the approach..

A MODEL PREDICTIVE CONTROLLER BASED ON SUPPORT VECTOR REGRESSION AND GENETIC OPTIMIZATION FOR AN SP-100 SPACE NUCLEAR REACTOR

Man Gyun Na and Belle R. Upadhyaya

In this work, a model predictive control method combined with support vector regression, is applied to the design of the thermoelectric (TE) power control in the SP-100 space reactor. The future TE power is predicted by using the support vector regression. The objectives of the proposed model predictive controller are to minimize both the difference between the predicted TE power and the desired power, and the variation of control drum angle that adjusts the control reactivity. Also, the objectives are subject to maximum and minimum control drum angle and maximum drum angle variation speed. The genetic algorithm is used to optimize the model predictive controller. A lumped parameter simulation model of the SP-100 nuclear space reactor is used to verify the proposed controller. The results of numerical simulations to check the performance of the proposed controller show that the TE generator power level controlled by the proposed controller could track the target power level effectively, satisfying all control constraints.

DETERMINING ELLIPSOIDAL BASINS OF ATTRACTION OF FUZZY SYSTEMS

Carlos Ariño and Antonio Sala

This paper discusses how to obtain local stability results from a fuzzy system for which global ones cannot be obtained, basically due to infeasibility of some associated LMI problems. Two different approaches are compared: modifying the consequent models vs. setting up some relaxed LMI conditions if bounds on the memberships are known. Some examples are used to illustrate the approaches.

KEY PERFORMANCE INDICATORS IN PLANT-WIDE CONTROL

Sebastjan Zorzut, Vladimir Jovan and Alenka Žnidaršič

Abstract: To improve production performance it is necessary to define production goals with a proper implementation strategy and a suitable closed-loop control for their achievement. Closed-loop control structures for simple systems such as temperature or velocity control are well defined, but a synthesis of plant-wide control structures is still recognized as the most important design problem in production management in process industries. A crucial issue to be resolved is the translation of implicit operating objectives, such as minimization of production costs, to a set of measurable variables that can then be used in a feedback control. A promising solution is the use of the Key Performance Indicators (KPIs) approach. To verify the idea of production feedback control using production KPIs as referenced controlled variables, a procedural model of a production process for a polymerisation plant has been developed. The model has been used during a number of simulation runs performed with the aim of developing and verifying the idea of KPI-based production control.

MODELLING ADAPTIVE CONTROLLERS WITH EVOLVING LOGIC PROGRAMS

Pierangelo Dell'Acqua, Anna Lombardi and Luís Moniz Pereira

The paper presents the use of Evolving Logic Programming (EVOLP) to model adaptive controllers. The resulting adaptive logic-based control system can be seen in the general framework of model reference adaptive systems. Two case studies are illustrated: in the first case the controller is implemented by using EVOLP, while in the second case EVOLP is used for the supervisor module with a generic controller. The advantage of using well-defined, self-evolving logic-based controllers is that it is possible to model dynamic environments, and to formally prove systems' requirements.

ENCODING FUZZY DIAGNOSIS RULES AS OPTIMISATION PROBLEMS

Antonio Sala, Alicia Esparza, Carlos Ariño and Jose V. Roig

This paper discusses how to encode fuzzy knowledge bases for diagnostic tasks (i.e., list of symptoms produced by each fault, in linguistic terms described by fuzzy sets) as constrained optimisation problems. The proposed setting allows more flexibility than some fuzzy-logic inference rulebases in the specification of the diagnostic rules in a transparent, user-understandable way (in a first approximation, rules map to zeros and ones in a matrix), using widely-known techniques such as linear and quadratic programming.

ENHANCEMENT OF MANEUVERABILITY OF A POWER ASSIST OMNI-DIRECTIONAL WHEELCHAIR BY APPLICATION OF NEURO-FUZZY CONTROL

Kazuhiko Terashima, Juan Urbano and Hideo Kitagawa

For helping attendants of handicapped people and elderly people, a power assist system has been added to an Omni-directional Wheelchair (OMW). With this addition it is possible for the attendants to deal with heavy loads, but there is a problem of operability when the attendants want to easily move OMW laterally or rotate around OMW's Gravity Center (CG). To solve the present problem, this paper provides a fuzzy reasoning method for estimating the navigation direction according to the force added by the attendants to the handgrips of the handle of OMW. A neuro-fuzzy system (ANFIS) is used for auto-tuning of the membership functions of the fuzzy system according to each attendant's characteristics, by using input data of attendants.

AUTOMATIC GENERATION OF OPTIMAL CONTROLLERS THROUGH MODEL CHECKING TECHNIQUES

Giuseppe Della Penna, Daniele Magazzeni, Alberto Tofani, Benedetto Intrigila, Igor Melatti and Enrico Tronci

We present a methodology for the synthesis of controllers, which exploits \emph{(explicit) model checking techniques}. That is, we can cope with the systematic exploration of a very large state space.\\ This methodology can be applied to systems where other approaches fail. In particular, we can consider systems with an \emph{highly non-linear dynamics} and \emph{lacking a uniform mathematical description (model)}.\\ We can also consider situations where the required control action cannot be specified as a local action, and rather a kind of \emph{planning} is required.\\ Our methodology individuates first a raw optimal controller, then extends it to obtain a more robust one.\\ A case study is presented which considers the well known \emph{truck-trailer obstacle avoidance parking problem}, in a parking lot with \emph{obstacles} on it. The complex non-linear dynamic of the truck-trailer system, within the presence of obstacles, makes the parking problem extremely hard.\\ We show how, by our methodology, we can obtain optimal controllers with different degrees of robustness.

INFLUENCE OF NBMAX AND TABU LIST IN THE SCHEDULING PROBLEM

Antonio Gabriel Rodrigues and Arthur Tórgo Gómez

In this paper is proposed a computational model that considers the Part Selection Problem and the Job Shop Scheduling Problem in a Flexible Manufacturing Cell. The objective of the proposed model is generate a schedule which reflects the management of three components in a objective function: (i) tardiness time, (ii) number of setups and (iii) number of tool switches. Previous experiments identify the conflict between tardiness time and setup and tool switches numbers. To manage this conflict, new experiments were made, in which the Tabu Search parameters (nbmax and tabu list size) are variated. This variation influences the objective function results, allowing better results in minimization of tardiness time.

COMPUTATIONAL FRAMEWORK FOR POWER ECONOMIC DISPATCH USING GENETIC ALGORITHM

Tahir Nadeem Malik, Abdul Qudus Abbasi and Aftab Ahmad

Economic Dispatch Problem (EDP) is the important step in Power System operation and is non-convex optimization problem. It has been solved comprehensively with mathematical programming approaches. However, these approaches handle non-convexity with assumption and resulting in an inaccurate dispatch. Genetic algorithms are potential tools for Economic dispatch and can handle it effectively. Computational framework “PED_Frame” has been developed which can handle economic dispatch using classical optimization and Genetic Algorithm based approaches independently and in hybrid form. It has been tested on standard three machine and twenty machine test systems.

FUZZY LOGIC BASED UAV ALLOCATION AND COORDINATION

James F. Smith III and ThanhVu H. Nguyen

A fuzzy logic resource allocation algorithm that enables a collection of unmanned air vehicles (UAVs) to automatically cooperate will be discussed. The goal of the UAVs’ coordinated effort is to measure the atmospheric index of refraction. Once in flight no human intervention is required. A fuzzy logic based planning algorithm determines the optimal trajectory and points each UAV will sample, while taking into account the UAVs’ risk, risk tolerance, reliability, and mission priority for sampling in certain regions. It also considers fuel limitations, mission cost, and related uncertainties. The real-time fuzzy control algorithm running on each UAV renders the UAVs autonomous allowing them to change course immediately without consulting with any commander, requests other UAVs to help, and change the points that will be sampled when observing interesting phenomena. Simulations show the ability of the control algorithm to allow UAVs to effectively cooperate to increase the UAV team’s likelihood of success.

AN OPTIMIZATION ALGORITHM TO IMPROVE SECURITY OF ELECTRICAL ENERGY SYSTEMS - An Hybrid Approach Based on Linear Programming and Load Flow Calculations

José V. Canto dos Santos, Arthur T. Gómez and Antônio G. Rodriguez

Power system restoration is one of the main problems in the electrical engineering area, due to the improving dependency of electricity of the modern industrial society. The restoration of large electrical power systems after the occurrence of serious blackouts is a complex problem where the basic goal is to obtain the system configuration in order to supply loads with different priorities. The restoration is done through stages and in each stage the service is restored to a predetermined set of loads. A method to solve the basic problem in a real power system restoration process is presented in this work. The solution takes into account the nonlinear electric network model (AC model) as well as its constraints and operational limits. The fictitious network concept is extended to the reactive model. Linear programming, a new model for the linearized power flow and conventional load flow calculation are also used. Results obtained with a test system and with a large realistic system are presented.

A SOLUTION TO THE VEHICLE ROUTING PROBLEM USING TABU SEARCH

Etiene Pozzobom Lazzeris Simas and Arthur Tórgo Gómez

This paper presents a model to the Vehicle Routing Problem using Tabu Search. The Vehicle Routing Problem aims to serve a set of clients by a fleet of vehicle through the creation of least-cost routes that satisfy some constraints. In this paper, only the vehicle capacity is considered. The objective of this model is to design least-cost routes to serve a set of clients with know demands in such way that some constraints that were defined are satisfied. An application to construct this model is proposed using Tabu Search. There were generated experiments that confirm that an increase in diversification on search space politics can generate results that are more qualified.

ON THE USE OF OPTIMIZATION METHODS FOR THE MINIMIZATION OF FERTILIZER APPLICATION ERROR WITH CENTRIFUGAL SPREADERS

Teddy Virin, Jonas Koko, Emmanuel Piron and Philippe Martinet

Fertilizer application is one of the most important operations in agricultural production. Thanks to their low cost and robustness, centrifugal spreaders are widely used to carry out this task. However, when distances between successive paths followed by the tractor in the field are not constant, application errors occur. These ones generally consist in over and under-application of nutrients. In the first case, over dosages can result in waters pollution. In the contrary case, economic issues occur with important yield losses. In this paper, to limit harmful environmental effects and disastrous drop in production due to centrifugal spreading, we propose an approach based on optimization techniques to improve the fertilization quality. An optimization criterion relying on a spatial distribution model, obtained in previous works, is considered. To compute optimal parameters which should be used as reference variables for the control of the spreader in the future, mechanical constraints are introduced. Faced with a large scale problem, we use an augmented lagrangian algorithm combined with a l-bfgs technique. Simulations results show low application error values comparing to fertilization inaccuracies found without optimization.

STABILITY OF TAKAGI-SUGENO FUZZY SYSTEMS

İlker Üstoğlu

Takagi-Sugeno (T-S) fuzzy models are usually used to describe nonlinear systems by a set of IF-THEN rules that gives local linear representations of subsystems. The overall model of the system is then formed as a fuzzy blending of these subsystems. It is important to study their stability or the synthesis of stabilizing controllers. The stability of TS models has been derived by means of several methods: Lyapunov approach, switching systems theory, linear system with modeling uncertainties, etc. In this study, the uniform stability, and uniform exponential stability of a discrete time T-S model is examined, a pointwise-in-time eigenvalue condition for exponential stability based on Rayleigh-Ritz inequality is presented. Moreover, a perturbation result and an instability condition is given. The subsystems of T-S models that is studied here are time varying and a new exponential stability theorem is given for these types of TS models by examining the existence of a common matrix sequence.

CONSIDERATIONS FOR SELECTING FUNCTIONS AND TERMINALS IN GENETIC PROGRAMMING FOR FAULT-DETECTION IN EMBEDDED SYSTEMS

Matej Šprogar, Domen Verber and Matjaž Colnarič

The article describes the terminals and functions used by genetic programming to discover specific parameters for fault-detection in embedded control systems design. Choice of different functions and terminals affects the convergence speed. The state of embedded controller is mapped into a space of valid/invalid points and genetic programming is used to divide the space into hypercubes that can be used to trivially recognize faults during system operation. The fault-detection logic operates by monitoring the input and output variables of the embedded controller. It is based on acquired and built-in knowledge about the normal behaviour in order to detect abnormalities. The fault-detection problem is approched by the use of monitoring cells, which implement the system supervising logic.

FEATURE SELECTION FOR IDENTIFICATION OF SPOT WELDING PROCESSES

Eija Haapalainen, Perttu Laurinen, Heli Junno, Lauri Tuovinen and Juha Röning

Process identification in the field of resistance spot welding can be used to improve welding quality and to speed up the set-up of a new welding process. Previously, good classification results of welding processes have been obtained using a feature set consisting of $54$ features extracted from current and voltage signals recorded during welding. In this study, the usability of the individual features is evaluated and various feature selection methods are tested to find an optimal feature subset to be used in classification. Ways are sought to further improve classification accuracy by discarding features containing less classification-relevant information. The use of a small feature set is profitable in that it facilitates both feature extraction and classification. It is discovered that the classification of welding processes can be performed using a substantially reduced feature set. In addition, careful selection of the features used also improves classification accuracy. In conclusion, selection of the feature subset to be used in classification notably improves the performance of the spot welding process identification system.

DYNAMIC GOAL COORDINATION IN PHYSICAL AGENTS

Jose Antonio Martin H and Javier de Lope

A general framework for the problem of coordination of multiple competing goals in dynamic environments for physical agents is presented. This approach to goal coordination is a novel tool to incorporate a deep coordination ability to pure reactive agents. The framework is based on the notion of multi-objective optimization. We propose a kind of “aggregating functions” formulation with the particularity that the aggregation is weighted by means of a dynamic weighting unitary vector w(S) which is dependant from the system dynamic state allowing the agent to dynamically coordinate the priorities of it’s single goals. This dynamic weighting unitary vector is represented as a n ? 1 set of angles. The dynamic coordination must be established by means of a mapping between the state of the agent’s environment S to the set of angles ?i (S) by means of any sort of machine learning tool. In this work we investigate the use of Reinforcement Learning as a first approach to learn that mapping.

NEURAL NETWORK MODEL BASED ON FUZZY ARTMAP FOR FORECASTING OF HIGHWAY TRAFFIC DATA

D. Boto-Giralda, M. Antón-Rodríguez, F. J. Díaz -Pernas and J. F. Díez Higuera

In this article, a neural network model is presented for forecasting the average speed values at highway traffic detectors locations using the Fuzzy ARTMAP theory. The performance of the model is measured by the deviation between the speed values provided by the loop detectors and the predicted speed values. Different Fuzzy ARTMAP configuration cases are analysed in their training and testing phases. Some ad-hoc mechanisms added to the basic Fuzzy ARTMAP structure are also described to improve the entire model performance. The achieved results make this model suitable for being implemented on advanced traffic management systems (ATMS) and advanced traveller information system (ATIS).

A FUZZY APPROACH FOR FAULT DETECTION AND ISOLATION OF UNCERTAIN PARAMETER SYSTEMS AND COMPARISON TO BINARY LOGIC

Salma Bouslama Bouabdallah and Moncef Tagina

This paper deals with fault detection and isolation off-line affecting sensors and actuators of uncertain parameter systems modelled by bond graph. A fuzzy approach for fault detection based on residual fuzzification is proposed. Besides, an isolation method based on fuzzy processing of the detection results is proposed. Finally binary approach and fuzzy one are compared through an illustrative example.

MATRIX-BASED HIERARCHICAL FUZZY SYSTEMS

Santiago Aja-Fernández and Carlos Alberola-López

A matrix inference method for fuzzy systems is used to deal with hierarchical fuzzy systems (HFSs). A method to decompose a multiple input fuzzy system into a HFS is presented. This method is based in representing the structure of a fuzzy system using matrices. An example of such a conversion for a three-input system is included.

FINDING A COMMON QUADRATIC LYAPUNOV FUNCTION USING CONICAL HULLS

Rianto Adhy Sasongko and J. C. Allwright

Consider a set of linear time-invariant continuous-time systems that is a convex hull with vertices formed by a given set of systems. The problem of finding a common Lyapunov function $v$, specified in terms of a symmetric positive definite matrix, for the convex hull of systems is tackled by searching for a symmetric positive definite (PD) matrix $P$ which causes $d\mrm{v} \slash dt$ to be negative definite for each vertex system. The approach involves an extension of an existing method for solving optimization problems for positive semidefinite (PSD) matrices that is based on a representation of the cone of PSD matrices as a conical hull. The condition that the derivative of the Lyapunov function for each vertex system is negative definite is converted naturally into the condition that the matrix $P$ belongs to the interior of the intersection of several conical hulls: one for each vertex system to ensure $d\rm{v}\slash dt$ for it is negative definite. The determination of a $P$ in the intersection is viewed as the solution of a quadratic programme on the product space of the cones. Then the existing theory and algorithms for conical hull problems are adapted to the solution of the quadratic programme. Initial numerical results indicate that the new approach appears to be about $26\%$ slower than the projective method used by MATLAB however the times for the new approach have been obtained using MATLAB scripts whereas MATLAB's projective algorithm was used as a executable file obtained from C. One would expect, therefore, that the algorithms should be competitive after the scripts have been converted into such executable files.

HYBRID LEARNING METHOD FOR DISCRETE MANUFACTURING CONTROL USING KNOWLEDGE BASED MODEL

Ewa Dudek and Tadeusz Dyduch

A conception of a hybrid learning method for discrete manufacturing processes control is presented. The method is based on a special form of a knowledge based model of discrete manufacturing process, named here hybrid knowledge based model (HKBM). The model consists of two parts, each of a different type of model: algebraic-logical model in a state spacethat is created on a basis of process technology description and set of expert rules referring to control. A general scheme of HKBM of a vast class of discrete manufacturing processes (DMP) is given in the paper. Then the method of synthesis of intelligent, learning algorithms that use information on the process gained in previous iterations as well as an expert knowledge is described. To illustrate the presented ideas, the scheduling algorithm for a special NP-hard problem is given.

DISTRIBUTED CONTROL SYSTEMS BASED ON COTS COMMUNICATION DATA BUS

Martin Švéda, István Szabó and Vladimír Opluštil

This paper deals with the distributed commercial off the shelf (COTS) databus based on Controller Area Network (CAN) used as a communication databus for Autonomous Locomotion Robot (ALR) and a System of Avionics Modules (SAM) that is used in civil aircraft Ae270. This article describes main characteristics of CAN communication databus and its higher layer protocol CANaerospace that are used in communication system of ALR and SAM. The basic idea of distributed control systems are described and their main characteristics are presented. Developed control systems proved that the CAN with HLP CANaerospace is eficient and reliable communication databus that can be used in safety critical applications like mobile robots, automotive, and avionics systems.

PREEMPTIVE SCHEDULING IN A TWO-STAGE MULTIPROCESSOR FLOWSHOP WITH RESOURCE CONSTRAINTS

Ewa Figielska

A heuristic combining the column generation technique and a genetic algorithm is proposed for solving the problem of preemptive scheduling in a two-stage flowshop with parallel unrelated machines and renewable resources at the first stage and a single machine at the second stage. The objective is to minimize the makespan. The lower bound on the optimal makespan is derived to be used in the performance analysis of the heuristic. The performance of the heuristic is analyzed by a computational experiment. The results show that the heuristic is able to find near-optimal solutions in reasonable computation time.

DISTRIBUTED EMERGENCY MANAGEMENT WITH SPATIAL SCENARIOS

Peter Sapaty, Robert Finkelstein and Joaquim Filipe

A radically new approach will be described for the fully distributed and dynamic management of advanced crisis relief operations and missions. It is based on the installation of a universal “social” module in many existing and massively used data processing and control devices, including (but not limited to) internet hosts, laptops, mobile robots and mobile phones. These modules can collectively interpret a special scenario language while exchanging higher-level program code with accompanying data and control in parallel. This can dynamically integrate any scattered post-disaster human and technical resources into an operable distributed system which, from one side, is effectively supervised externally, and from the other side, is capable of solving complex self-analysis, coordination, survivability, relief, and reconstruction problems autonomously.

NEURAL NETWORK SYSTEM FOR WASTE-WATER RECOGNITION

Radek Kuchta and Radimir Vrba

This paper presents modern method of using neural network for waste-water recognition by using sensor array. Each sensor in sensor array detects chemicals in waste-water with different sensitivity. Set of measured data is digitized and recognized by a neural network. Measuring process doesn’t need any human operator. The result gives the only information: contaminated or not contaminated

HEAT-AND-POWER PROCESSES OPTIMIZATION BY MEANS OF MODEL-BASED SIMULATION

Dmitry Antropov, Rosica Ivanova, Renat Sadykov, Sergey Yeryomin and Rauf Kafiatullin

A firmware was developed for simulation of heat-mass transfer processes in power equipment such as steam or water boilers and dryers. Hardware of this pilot plant is based on modern microprocessors control devices. Software rests on specially developed mathematical models. The functions and structure of the model of fully automated boiler and dryer control system (B-DCS) are described in detail. One of the variants of implementation of B-DCS on the example of the dryer unit for drying bioactive products is considered. The analysis of the optimality criterion problem and selection of the optimal control structure are reviewed using Pontryagin’s maximum principle. The objective of optimization is to reduce expenditures of operational process.

HYBRID EVOLUTIONARY COMPUTATIONS - Application for Industry Investment Problem

Tadeusz Dyduch

The aim of the paper is two-fold. Firstly, it is to present in more details a special type of hybrid evolutionary computation, named by the authors Two-Level Adaptive Evolutionary Computation (TLAEC). The method consists in combination of evolutionary computation with deterministic optimization algorithms in a hierarchical system. Novelty of the method consists also in a new type of adaptation mechanism. Post optimal analysis of the lower level optimization task is utilized in order to modify probability distributions for new genotype generating. The second aim of the paper is to present an algorithm based on TLAEC method, solving a difficult optimization problem. A mathematical model of this problem assumes the form of mixed discrete-continuous programming. A concept of the algorithm is described in the paper and the proposed, new adaptation mechanism that is implemented in the algorithm is described in detail. The results of computation experiments as well as their analysis are also given.

THE HIERARCHICAL MAP FORMING MODEL

Luis Eduardo Rodriguez Soto and Cheng-Yuan Liou

In the present paper we propose a motor control model inspired by organizational priciples of the cerebral cortex. Speci…cally the model is based on cortical maps and functional hierarchy in sensory and motor areas of the brain. Self-Organizing Maps (SOM) have proven to be useful in modeling cortical topological maps (Palakal et al., 1995). A hierarchical SOM provides a natural way to extract hierarchical informa- tion from the environment, which we propose may in turn be used to select actions hierarchically. We use a neighborhood update version of the Q-learning algorithm, so the …nal model maps a continuous input space to a continuous action space in a hierarchical, topology preserving manner. The model is called the Hierarchical Map Forming model (HMF) due to the way in which it forms maps in both the input and output spaces in a hierarchical manner.

DETECTING LICENSE PLATE USING CLUSTER RUN LENGTH SMOOTHING ALGORITHM

Siti Norul Huda Sheikh Abdullah, Marzuki Khalid, Rubiyah Yusof and Khairuddin Omar

Vehicle license plat recognition has been a much studied research area in many countries. Due to the different types of license plates being used, the requirement of an automatic license plate recognition system is rather different for each country. In this paper, an automatic license plate recognition system is proposed for Malaysian vehicles with standard license plates based on image processing, feature extraction and neural networks. The image processing library is developed in-house which we referred to as Vision System Development Platform (VSDP). The Kirsch Edge feature extraction technique is used to extract features from the license plates characters which are then used as inputs to the neural network classifier. The neural network model is the standard multilayered perceptron trained using the back-propagation algorithm. The prototyped system has an accuracy of about 91%, however, suggestions to further improve the system are discussed in this paper based on the analysis of the error.

N-ARY TREES CLASSIFIER

Duarte Duque, Henrique Santos and Paulo Cortez

This paper addresses the problem of automatic detection and prediction of abnormal human behaviours in public spaces. For this propose a novel classifier, called N-ary trees, is presented. The classifier processes time series of attributes like the object position, velocity, perimeter and area, to infer the type of action performed. This innovative classifier can detect three types of events: normal; unusual; or abnormal events. In order to evaluate the performance of the N-ary trees classifier, we carry out a preliminary study with 180 synthetic tracks and one restricted area. The results revealed a great level of accuracy and that the proposed method can be used in surveillance systems.

VISUAL TOPOLOGICAL MAP BUILDING IN SELF-SIMILAR ENVIRONMENTS

Toon Goedemé, Tinne Tuytelaars and Luc Van Gool

This paper describes a method to automatically build topological maps for robot navigation out of a sequence of visual observations taken from a camera mounted on the robot. This direct non-metrical approach relies completely on the detection of loop closings, i.e. repeated visitations of one particular place. In natural environments, visual loop closing can be very hard, for two reasons. Firstly, the environment at one place can look differently at different time instances due to illumination changes and viewpoint differences. Secondly, there can be different places that look alike, i.e. the environment is self-similar. Here we propose a method that combines state-of-the-art visual comparison techniques and evidence collection based on Dempster-Shafer probability theory to tackle this problem.

LOCALIZATION WITH DYNAMIC MOTION MODELS - Determining Motion Model Parameters Dynamically in Monte Carlo Localization

Adam Milstein and Tao Wang

Localization is the problem of determining a robot’s location in an environment. Monte Carlo Localization (MCL) is a method of solving this problem by using a partially observable Markov decision process to find the robot’s state based on its sensor readings, given a static map of the environment. MCL requires a model of each sensor in order to work properly. One of the most important sensors involved is the estimation of the robot’s motion, based on its encoders that report what motion the robot has performed. Since these encoders are inaccurate, MCL involves using other sensors to correct the robot’s location. Usually, a motion model is created that predicts the robot’s actual motion, given a reported motion. The parameters of this model must be determined manually using exhaustive tests. Although an accurate motion model can be determined in advance, a single model cannot optimally represent a robot’s motion in all cases. With a terrestrial robot the ground surface, slope, motor wear, and possibly tire inflation level will all alter the characteristics of the motion model. Thus, it is necessary to have a generalized model with enough error to compensate for all possible situations. However, if the localization algorithm is working properly, the result is a series of predicted motions, together with the corrections determined by the algorithm that alter the motions to the correct location. In this case, we demonstrate a technique to process these motions and corrections and dynamically determine revised motion parameters that more accurately reflect the robot’s motion. We also link these parameters to different locations so that area dependent conditions, such as surface changes, can be taken into account. These parameters might even be used to identify surface changes by examining the various parameters. By using the fact that MCL is working, we have improved the algorithm to adapt to changing conditions so as to handle even more complex situations.

HIERARCHICAL MULTI-ROBOT COORDINATION - Aggregation Strategies Using Hybrid Communication

Yan Meng, Jeffrey V. Nickerson and Jing Gan

Multi-robot coordination is important for searching tasks. Usually discussions of this coordination presuppose a reliable explicit communication infrastructure. However, limited power, low radio range, and an ever changing environment all hinder communication. Maintaining weakened connections will cause robots to cluster during searching, which may be suboptimal with respect to the searching time. In this paper, several integration strategies with a hierarchical networked architecture are proposed to coordinate a team of robots which have lost explicit communication. To speed up the reconnection procedure for the proposed aggregate strategies, implicit communication through vision sensors is proposed in this paper to establish a movement plan to recover the explicit communication. Simulation results are presented and discussed. Experiments with 3 Pioneer robots have been conducted, and the experimental results show that our proposed strategies using a hybrid communication mechanism are feasible and efficient in a searching task. The proposed strategies can be extended to a large-scale searching environment as well as to a combination of humans and robots.

STUDIES ON VISUAL PERCEPTION FOR PERCEPTUAL ROBOTICS

Özer Ciftcioglu, Michael S. Bittermann and I. Sevil Sariyildiz

Studies on human visual perception measurement for perceptual robotics are described. The visual perception is mathematically modelled as a probabilistic process obtaining and interpreting visual data from an environment. The measurement involves visual openness perception in the virtual reality which has direct implications for navigation issues of actual autonomous robotics. The perception is quantified by means of a mapping function which converts a distance to an elemental perception estimate. The measurement is carried out with the averaging of the elemental perceptions in real time. This is accomplished by means of exponential averaging. The mapping function parameters are optimized uniquely by means of genetic algorithm approach where the data set for model development consists of a number of perception data samples. These are obtained from individuals who are confronted with a number of scenes and asked for their perceptual openness statements. Based on this data, a perception model is developed for a robot where the simulated vision interaction of the robot with the environment is converted to visual openness estimation through the model output. The model outcome is essential visual information for the navigation of an autonomous perceptual robot.

VISUAL SPEECH RECOGNITION USING WAVELET TRANSFORM AND MOMENT BASED FEATURES

Wai C. Yau, Dinesh K. Kumar, Sridhar P. Arjunan and Sanjay Kumar

This paper presents a novel approach using feature extraction that combines Discrete StationaryWavelet Transform (SWT) and image moments to classify utterances consisting of consonants. A view based method is adopted to represent the 3-D image sequence of the mouth movement in a 2-D space using grayscale images named as motion history image (MHI). MHI is produced by applying accumulative image differencing technique on the sequence of images to implicitly capture the temporal information of the mouth movement. A 2-D SWT at level 1 is applied to decompose MHI to produce one approximate and three detail sub images. Three different moment-based features, namely Zernike moments, geometric moments and Hu moments are computed from the approximate representation of MHI to form the feature vectors. Supervised feed forward multilayer perceptron (MLP) artificial neural network (ANN) with back propagation learning algorithm is used to classify the moment-based features. The performance and image representation ability of these features are compared in this paper. The preliminary results show that this method can achieve high recognition rate in classification of 3 consonants.

LOCALITY AND GLOBALITY: ESTIMATIONS OF THE ENCRYPTION COLLECTIVITIES

Cristian Lupu, Tudor Niculiu and Eduard Franţi

In this paper we try to define a collectivity, to model and to measure it. Because N. Bourbaki names "collectivizing relation" the relation defining a set, we name collectivities only the sets selected or built by the help of the relations. The orthogonal interconnections model very well the collectivities. The behavior (structural self-organization) around the origin is different for homogenous and non-homogenous interconnections. How can we measure this behavior? A way is by locality and globality. The locality measures analytically by neighborhoods, neighborhood reserves, {\it Moore} reserves and synthetically by diameters, degrees, average distances. The globality is the behavior of an interconnection around a property. The globality vs. symmetry measures by the compactity, efficiency and interconnecting filling. The locality and the globality are among primary manifestations of the self-organization. In this way, collectivities modeled by self-organizing interconnections can contribute to changing our fundamental view of computers by trying to bring them nearer to the nature.

ONLINE HIERACHICAL CONTROL FOR LEGGED SYSTEMS BASED ON THE INTERACTION FORCES

José R. Puga, Filipe M. Silva and Boaventura R. da Cunha

This paper presents a motion planning and control method with application in the field of legged robots. The general aim is to explore a set of simple underlying principles that govern balance of posture and gait of biped robots, and to develop control methodologies for such a highly unstable and no linear plants. The proposed controller reflects a hierarchical structure based on the interaction forces between the foot and ground and simple feedback rules used online. The algorithms are applied to a simulated 3-D leg model with five degrees of freedom (DOF). The simulation analyses demonstrate the capability of the control system to keep balance when the leg executes different tasks. To validate the proposed method several aspects are investigated, such as the posture robustness on the level ground when subject to external perturbations, the adaptation when standing in a moving platform and the improvements introduced by the compensation of the tangential reaction forces.

DIFFERENT CLASSIFIERS FOR THE PROBLEM OF EVALUATING CORK QUALITY IN AN INDUSTRIAL SYSTEM

Beatriz Paniagua-Paniagua, Miguel A. Vega-Rodríguez, Juan A. Gómez-Pulido and Juan M. Sánchez-Pérez

In this paper we study the use of different classifiers to solve a classification problem existing in the cork industry: the cork stopper/disk classification according to their quality using a visual inspection system. Cork is a natural and heterogeneous material, therefore, its automatic classification (usually, seven different quality classes exist) is very difficult. The classifiers, which we present in this paper, work with several quality discriminators (features), that we think could influence cork quality. These discriminators (features) have been checked and evaluated before being used by the different classifiers that will be exposed here. In this paper we attempt to evaluate the performance of a total of 4 different cork quality-based classifiers in order to conclude which of them is the most appropriate for this industry, and therefore, obtains the best cork classification results. In conclusion, our experiments show that the Euclidean classifier is the one which obtains the best results in this application field.

IMPROVING TRACKING TRAJECTORIES WITH MOTION ESTIMATION

Jorge Pomares, Gabriel J. García and Fernando Torres

Up to now, different methods have been proposed to track trajectories using visual servoing systems. However, when these approaches are employed to track trajectories specified with respect to moving objects, different considerations must be included in the visual servoing formulation to progressively decrease the tracking error. This paper shows the main properties of a non-time dependent visual servoing system to track image trajectories. The control action obtained integrates the motion estimation of the object from which the features are extracted. The proposed motion estimator employs information from the measures of the extracted features and from the variation of the camera locations. These variations are obtained determining the Homography matrix between consecutive camera frames.

USING THE TRANSFERABLE BELIEF MODEL TO VEHICLE NAVIGATION SYSTEM

Touil Khalid, Zribi Mourad and Benjelloun Mohammed

In general, navigation systems estimating a vehicle position is done either by using the Global Positioning System (GPS) or the Dead Reckoning (DR) systems. Other modern estimations are based on the combination of the two systems (GPS/DR). However, the position of a vehicle determined by GPS/DR is far from being perfect since it produces many errors. To solve this problem, a map-matching method is proposed in order to reduce the errors of localization caused by GPS/DR. This algorithm, which uses a digital road map, allows the detection of the correct road where a vehicle moves. In this paper, we introduce a new map-matching algorithm that employs the Transferable Belief Model (TBM). The TBM presents a general justification of belief theory and provides a flexible and adapted representation for the measured beliefs. Experimental results show the effectiveness of the utilization of the TBM to the vehicle navigation system.

SIMULTANEOUS LOCALIZATION AND MAPPING IN UNMODIFIED ENVIRONMENTS USING STEREO VISION

A. Gil, O. Reinoso, C. Fernández, M. A. Vicente, A. Rottmann and O. Martínez Mozos

In this paper we describe an approach that builds three dimensional maps using visual landmarks extracted from images of an unmodified environment. We propose a solution to the Simultaneous Localization and Mapping (SLAM) problem for autonomous mobile robots using visual landmarks. Our map is represented by a set of three dimensional landmarks referred to a global reference frame, each landmark containing a visual descriptor that partially differentiates it from others. Significant points extracted from stereo images are used as natural landmarks, in particular we employ SIFT features found in the environment. We estimate both the map and the path of the robot using a Rao-Blackwellized particle filter, thus the problem is decomposed into two parts: one estimation over robot paths using a particle filter, and N independent estimations over landmark positions, each one conditioned on the path estimate. We actively track visual landmarks at a local neighbourhood and select only those that are more stable. When a visual feature has been observed from a significant number of frames it is then integrated in the filter. By this procedure, the total number of landmarks in the map is reduced, compared to prior approaches. Due to the tracking of each landmark, we obtain different examples that represent the same natural landmark. We use this fact to improve data association. Finally, efficient resampling techniques have been applied, which reduces the number of particles needed and avoids the particle depletion problem.

A SOLUTION FOR EVALUATING THE STOPPER QUALITY IN THE CORK INDUSTRY

Beatriz Paniagua-Paniagua, Miguel A. Vega-Rodríguez, Juan A. Gómez-Pulido and Juan M. Sánchez-Pérez

In this paper we study a possible solution to a problem existing in the cork industry: the cork stopper/disk classification according to their quality using a visual inspection system. Cork is a natural and heterogeneous material, therefore, its automatic classification (usually, seven different quality classes exist) is very difficult. The solution proposed in this paper shows all the stages made in our study: quality discriminatory features selection and extraction, texture analysis, analysis of different (global and local) automatic thresholding techniques and possible classifiers. In each stage we have given more importance to the study of those aspects that we think could influence the cork quality. In this paper we attempt to evaluate each of the stages in our solution to the problem of the cork classification in an industrial environment, and therefore, finding a way to justify the design of our final classification system. In conclusion, our experiments show that the best results are obtained by a system that works with the following features: total cork area occupied by defects (thresholding with heuristic fixed value 69), textural contrast, textural entropy and size of the biggest defect in the cork, all of them working in an Euclidean classifier. The obtained results have been very encouraging.

DYNAMIC PARAMETERS IDENTIFICATION OF AN OMNI-DIRECTIONAL MOBILE ROBOT

André Scolari Conceição, A. Paulo Moreira and Paulo J. Costa

This paper presents the experimental dynamic parameters identification of an omni-directional mobile robot with four wheels. Three methods of parameters identification related to dynamic equations are described, the parameters are the viscous frictions, the coulomb frictions and the inertia moment of the robot. A simulation environment, simulation results and real results are presented.

TRAJECTORY CONTROL AND MODELLING OF AN OMNI-DIRECTIONAL MOBILE ROBOT

André Scolari Conceição, A. Paulo Moreira and Paulo J. Costa

This paper presents a dynamic and kinematic model and a trajectory controller for an omni-directional mobile robot. The parameters of the controller are optimizated based on trajectory following simulations, with the mobile robot model, take into account aspects like time and errors of position and orientation of the robot. Simulation and real results of trajectory following are presented.

FAULT DETECTION OF THE ACTUATOR BLOCKING - Experimental Results in Robot Control Structures

Matei Vinatoru and Eugen Iancu

In this paper is presented an algorithm, which allows for certain robotic structure, under the terms of an actuator blocking occurrence during the operation, either a correct positioning (if it is possible) or a positioning in an acceptable proximity of the desired co-ordinates by minimising an optimal criteria (through the adequate commands to the functional elements). The paper is proposing the synthesis of the commands to a poly-articulated robotic arm (3 segments). First, a workspace analysis is made, then is presented the algorithm for the actuators, first in the terms of a normal operation (finding the optimal motions) and second in terms of the blocking of some robotic segments.

TWO LAYER CONTROL STRATEGY APPLIED TO BUILDING AUTOMATION

João Figueiredo and José Sá da Costa

In this paper a control and monitoring platform for an intelligent building is developed using a SCADA system (Supervisory Control And Data Acquisition). The control strategy develops a two-level architecture where inner-loops are performed by local PLCs (Programmable Logic Controller), and the outer-loop is managed by the centralized SCADA system that interacts with the entire local PLC network. The outer loop performs an intelligent Generalized Predictive Control strategy (GPC). Tests on a prototype are shown, where all the instrumentation in place is controlled by an industrial PLC master/slave network. The master PLC is connected, in real-time acquisitions, with the SCADA system via MultiPoint Interface (MPI). The experimental work shows the potential of instrumentation, control and monitoring in the future buildings to prevent accidents and improve the quality of living.

A GAIN-SCHEDULING APPROACH FOR AIRSHIP PATH-TRACKING

Alexandra Moutinho and José Raul Azinheira

In this paper a gain scheduled optimal controller is designed to solve the path-tracking problem of an airship. The control law is obtained from a coupled linear model of the airship that allows to control the longitudinal and lateral motions simultaneously. Due to the importance of taking into account wind effects, which are rather important due to the airship large volume, the wind is included in the kinematics, and the dynamics is expressed as function of the air velocity. Two examples are presented with the inclusion of wind, one considering a constant wind input and the other considering in addition a 3D turbulent gust, demonstrating the effectiveness of this single controller tracking a reference path over the entire flight envelope.

DEPTH GRADIENT IMAGE BASED ON SILHOUETTE - A Solution for Reconstruction of Scenes in 3D Environments

Pilar Merchán, Antonio Adán and Santiago Salamanca

Greatest difficulties arise in 3D environments when we have to deal with a scene with dissimilar objects without pose restrictions and where contacts and occlusions are allowed. This work tackles the problem of correspondence and alignment of surfaces in such a kind of scenes. The method presented in this paper is based on a new representation model called Depth Gradient Image Based on Silhouette (DGI-BS) which synthesizes object surface information (through depth) and object shape information (through contour). Recognition and pose problems are efficiently solved for all objects of the scene by using a simple matching algorithm in the DGI-BS space. As a result of this the scene can be virtually reconstructed. This work is part of a robot intelligent manipulation project. The method has been successfully tested in real experimentation environments using range sensors.

A NEW METHOD FOR REJECTION OF UNCERTAINTIES IN THE TRACKING PROBLEM FOR ROBOT MANIPULATORS

Juan A. Méndez, S. Torres, L. Acosta, E. González and V. M. Becerra

This paper presents a new strategy for robust tracking in robot manipulators. The aim of the strategy is to reject parametric uncertainties due to model or load disturbances. The basic controller acting on the manipulator is a robust controller designed by Lyapunov’s direct method. Acting on this controller there is an adaptive system responsible for the adaptation of the basic parameter of the robust feedforward term. The paper describes in detail the theoretical setup of the proposed method. The performance of the strategy is tested in a Puma-560 manipulator. A comparison with existing techniques is done to verify the efficiency of the presented controller

AUTONOMOUS BEHAVIOR-BASED EXPLORATION OF OFFICE ENVIRONMENTS

Daniel Schmidt, Tobias Luksch, Jens Wettach and Karsten Berns

Besides safe motion control the gain of environmental knowledge is a key for a reliable home or office service robot. When being set into a completely unknown environment the robot has to be able to derive a certain abstract internal representation of this world without any user interaction. This knowledge enables the robot to known how to get from its actual place in one room to a target position in another room as a prerequisite transportation tasks for example. In this context, the combination of a behavior-based motion control system and an abstract topological map based on geometric representations of rooms seems promising. As the concept of motion and exploration behaviors facilitates to compete with noisy sensor information and geometrically imprecise maps, it has been used to develop exploration strategies for deriving topological representations of common indoor environments completely autonomously. The only prescribed world knowledge is the fact that these environments are composed of rectangular entities (rooms) which are connected by openings (doors). The developed system has successfully been tested in simulation and reality. Next steps concern the integration of furniture objects into the map as well as increasing the reliability of the mapping strategy in highly cluttered areas.

AUTONOMOUS GAIT PATTERN FOR A DYNAMIC BIPED WALKING

Christophe Sabourin, Kurosh Madani and Olivier Bruneau

In this paper, we propose an autonomous gait pattern for a dynamic biped walking. Our approach takes simultaneously advantage from a Fuzzy-CMAC based computation of robot's swing leg's desired trajectory and a high level control strategy allowing regulating the robot's average velocity. The main interest of this approach is to proffer to the walking robot autonomy and adaptability involving only one parameter: the average velocity. Furthermore, this approach allows increasing the robustness of the walking robot regarding the forwards pushed force.

DISTRIBUTED CONTROL SYSTEM OF AN EXPERIMENTAL ROBOTIC CELL WITH 3D VISION

Andrés S. Vázquez, Antonio Adán, Roberto Torres and Carlos Cerrada

We present a distributed control architecture for the integration of an experimental robotic cell with 3D visual servoing. This architecture allows us to control a 6 DOF robot in hard Real-Time and the global experimental system in soft Real-Time. We have developed distributed applications, based on this architecture, for the robot control (whose characteristics permit us to teleoperate the robot), the 3D data acquisition and for an advanced simulation and visualization. These applications, together with the algorithms developed by our computer vision research group, allow a full intelligent robotic manipulation in complex scenes to be made. This can be useful in manufacturing environments where an automated piece manipulation is necessary.

STATIC FACE DETECTION AND EMOTION RECOGNITION WITH FPGA SUPPORT

Paul Santi-Jones and Dongbing Gu

Throughout history, spoken language and face-to-face communication have been the primary mechanics of interaction between two or more people. While speech processing, it is often advantageous to determine the emotion of the speaker in order to better understand the context of the meaning. This paper looks at our current effort at creating a static based emotion detection system, using previously used techniques along with an FPGA neural network (FFP) to speed up recognition rates.

DESIGN OF A PROTOTYPE ROBOT VACUUM CLEANER - From Virtual Prototyping to Real Development

Leire Maruri, Ana Martinez-Esnaola, Joseba Landaluze, Sergio Casas and Marcos Fernandez

This paper presents the prototype of a robot vacuum cleaner designed and constructed by IKERLAN. It details, above all, the hardware and software components used, as well as the navigation algorithm, designed using fuzzy logic. In conjunction to this an existing virtual prototype of the robot and the domestic environment was updated with a view to fine-tuning and testing the real controller of the autonomous robot by means of SIL (Software-in-the-Loop) simulations. Finally, some of the position estimation problems that arose in the experimental tests are described.

MULTIPLE MOBILE ROBOTS MOTION-PLANNING: AN APPROACH WITH SPACE-TIME MCA

Fabio M. Marchese

In this paper is described a fast Path-Planner for Multi-robot composed by mobile robots having generic shapes and sizes (user defined) and different kinematics. We have developed an algorithm that computes the shortest collision-free path for each robot, from the starting pose to the goal pose, while considering their real shapes, avoiding the collisions with the static obstacles and the other robots. It is based on a directional (anisotropic) propagation of attracting potential values in a 4D Space-Time, using a Multilayered Cellular Automata (MCA) architecture. This algorithm makes a search for all the optimal collision-free trajectories following the minimum valley of a potential hypersurface embedded in a 5D space.

ELECTRONIC SOLUTION BASED ON MICRO-CONTROLLER AT91SAM7S256 FOR PLATOONING MULTI-AGENT SYSTEM IMPLEMENTATION

José M. Rodríguez, AbdelBaset M.H. Awawdeh, Felipe Espinosa, Julio Pastor, Fernando Valdés, Miguel A. Ruiz and Antonio Gil

In this work a low cost electronic solution adapted for control and communication of a convoy of electrical vehicle prototypes based on multi-agent system -MAS- is presented. From the obtained results in previous works, focused on mobile platforms with PC architecture and Bluetooth communication module, a new electronic system has been designed based on the 32 bits microcontroller AT91SAM7S256 and the communication module nRF2401A. With which, it obtains a greater integration of the final mobile prototype and a greater communication capability between the devices connected in wireless network.

NEURO-ADAPTIVE DYNAMIC CONTROL FOR TRAJECTORY TRACKING OF MOBILE ROBOTS

Marvin K. Bugeja and Simon G. Fabri

This paper presents a novel functional-adaptive dynamic controller for trajectory tracking of nonholonomic wheeled mobile robots. The controller is developed in discrete-time and employs a Gaussian radial basis function neural network for the estimation of the robot's nonlinear dynamic functions, which are assumed to be completely unknown. Optimal on-line weight tuning is achieved by employing the Kalman filter algorithm, based on a specifically formulated stochastic inverse dynamic identification model of the mobile base. A discrete-time dynamic control law employing the estimated functions is proposed and cascaded with a trajectory tracking kinematic controller. The performance of the complete system is analysed and compared by realistic simulations.

Dynamic and Distributed Allocation of Resource Constrained Project Tasks to Robots

Sanem Sariel, Tucker Balch and Nadia Erdogan

In this paper we present the design and implementation of a multi-robot cooperation framework to collectively execute inter-dependent tasks of an overall complex mission requiring diverse capabilities. Given a heterogeneous team of robots and task dependencies, the proposed framework provides a distributed mechanism for assigning tasks to robots in an order that efficiently completes the mission. The approach is robust to unreliable communication and robot failures. It is a distributed auction-based approach, and therefore scalable, but it does not provide guarantees of optimality. In order to obtain optimal allocations effective bid evaluations are needed. Additionally to maintain optimality in noisy environments dynamic re-allocations of tasks are needed as implemented in dynamic task selection and coalition maintenance scheme that we propose. Real-time contingencies are handled by recovery routines, called Plan B precautions in our framework. Here, in this paper, we present performance results of our framework for robustness in simulations that include variable message loss rates and robot failures. Experiments illustrate robustness of our approach against several contingencies.

A STUDY ON ASR/TTS SERVER ARCHITECTURE FOR NETWORK ROBOT SYSTEM

In-Ho Choi and Tae-Hoon Kim

“The URC(Ubiquitous Robotic Companion, Server computer-based networked robotic)” systems exploiting Internet-related technologies and Server computer require effective techniques for timely delivery of requested data to remote clients. In these systems, there is a need to process real-time data in server computer from/to robots and clients during system operation. In this paper, we describe and evaluate ASR, TTS server systems in the context of a real-time environment for the URC applications. Experimental results show that the server-based ASR, TTS support timely delivery of data to a potentially large number of robots during system operation.

STEREO DISPARITY ESTIMATION USING DISCRETE ORTHOGONAL MOMENTS

Tomasz Andrysiak and Michał Choraś

In the article we present various theoretical and experimental approaches to the problem of stereo matching and disparity estimation. We propose to calculate stereo disparity in the moments space, but we also present numerical and correlation based methods. In order to calculate disparity vector we decided to use discrete orthogonal moments of Tchebichef, Zernike and Legendre. In our research of stereo disparity estimation all of these moments were tested and compared. In the article we also propose the original method of determining the global displacement vector between the stereopair images in order to find the common part of these images (adequate for matching) and the margins of these stereo images. Experimental results confirm effectiveness of the presented methods of determining stereo disparity and stereo matching for robotics and machine vision applications.

THE VISIBILITY PROBLEM IN VISUAL SERVOING

C. Pérez, R. Morales, N. García-Aracil, J. M. Azorín and J. M. Sabater

This paper deals with the visibility problems occurring during the execution of a visual servoing task. First, a review of the scientific works related with the visibility are recalled and then the solution proposed by the authors is presented and extended to the case of the sudden disappearance of features on the center of the image. Experimental results demonstrate the improvements (stability and continuity) that can be obtained in the performance of the vision-based control task when the weighted features formulation is used.

IMPROVING THE RESULTS OF THE CONTENT-BASED IMAGE QUERY ON MEDICAL IMAGERY

Liana Stanescu, Dan Dumitru Burdescu, Anca Ion and Marius Brezovan

The article presents a solution for raising the quality of the content-based image query process, namely of the number of the relevant images retrieved from the database for a query image, in the case of the color medical images. The solution combines the content-based image query on color feature with color texture feature. There have been effectuated and presented studies of content-based image query on color images from the field of the digestive apparatus gathered with an endoscope. The color information is represented by the color histograms computed on HSV color space quantized at 166 colors. In order to represent the color texture the co-occurrence matrices are used. To compute the dissimilitude between the images, the histogram intersection has used for the color and the Euclidian distance for the color texture. The union of the results obtained with the two content-based image query methods on color and color texture, performed in parallel, leads to a greater number of retrieved relevant images. The reason is, that, generally, in the case of the considered diseases there are changes in the color and the texture of the sick tissue.

HYBRID IMPEDANCE CONTROL FOR MULTI-SEGMENTED INSPECTION ROBOT Kairo-II

C. Birkenhofer, S. Studer, J. M. Zöllner and R. Dillmann

The huge redundancy of multi-segmented robot KairoII can be utilized to add to a general robot configuration any inspection subtask . For doing so, an extensive control scheme has to be installed that is able to handle both, contact scenarios with the environment and ambiguous robot configurations. A method for implementing an appropriate scheme using transposed Jacobians based on Hybrid Impedance Control (TJ-HIC) is described and validated for multi-segmented robots. Crucial parts of this model are identified and implemented. Those parts are a dynamic robot model that is realized in Recursive Newton-Euler equations (RNE) and a sensory system for apropriate force feedback information.

ROBOTIC ARCHITECTURE BASED ON ELECTRONIC BUSINESS MODELS - From Physics Components to Smart Services

José Vicente Berná-Martínez, Francisco Maciá-Pérez, Virgilio Gilart-Iglesias and Diego Marcos-Jorequera

In this article we presented a view for the robots and robotic systems design based on applying models, architectures, techniques and tools that have allowed contributing valid solutions in other dominions of application, like the electronic business. Before being able to apply these solutions, it is essential to subjugate to the physical elements that compose a robotic system to a process of normalization that allows characterizing them from the point of view of its functional contribution. At this point we showed to the conceptual model and the technical architecture of the robotic system based on services oriented architectures. The work also gathers the implementation of a normalized robotic element according to the exposed techniques that allow verifying the validity of the proposal

PDPT FRAMEWORK - Building Information System with Wireless Connected Mobile Devices

Ondrej Krejcar

The proliferation of mobile computing devices and local-area wireless networks has fostered a growing interest in location-aware systems and services. Additionally, the ability to let a mobile device determine its location in an indoor environment at a fine-grained level supports the creation of a new range of mobile control system applications. Main area of interest is in model of radio-frequency (RF) based system enhancement for locating and tracking users of our control system inside buildings. The framework described here joins the concepts of location and user tracking in an extended existing control system. The experimental framework prototype uses a WiFi network infra-structure to let a mobile device determine its indoor position as well as to de-liver IP connectivity. User location is used to data pre-buffering and pushing information from server to user’s PDA. Experiments show that location deter-mination can be realized with a room level granularity.

MONTE CARLO LOCALIZATION IN HIGHLY SYMMETRIC ENVIRONMENTS

Stephan Sehestedt and Frank E. Schneider

The localization problem is a central issue in mobile robotics. Monte Carlo Localization (MCL) is a popular method to solve the localization problem for mobile robots. Unfortunately, usual MCL has some shortcomings in terms of computational complexity, robustness and the handling of highly symmetric environments. These three issues are adressed in this work. We present three Monte Carlo localization algorithms. The focus lies on two of these, which are especially suitable for highly symmetric environments. These algorithms aim on the efficient use of the samples and the usage of variable sample sets. That makes it feasible to use the presented algorithms in real-time applications. Furthermore, We use a procedure called Two-Stage Sampling as our resampling scheme, which allows us to keep track of multiple hypotheses over extended periods of time.

TRACKING MULTIPLE OBJECTS USING THE VITERBI ALGORITHM

Andreas Kräußling, Frank E. Schneider and Stephan Sehestedt

Tracking multiple targets is a great challenge for most tracking algorithms, since these algorithms tend to loose some of the targets when they get close to each other. Hence, several algorithms like the MHT, the JPDAF and the PMHT have been developed for this task. However, these algorithms are specialized on punctiform targets, whereas in mobile robotics one has to deal with extended targets. Therefore, in this paper an algorithm is proposed that can solve this problem. It uses the Viterbi algorithm and some geometrical characteristics of the problem. The proposed algorithm was tested with real world data.

HUMAN ARM-LIKE MECHANICAL MANIPULATOR - The Design and Development of a Multi -Arm Mobile Robot for Nuclear Decommissioning

Mohamed J. Bakari and Derek W. Seward

This paper reviews the design and development of a human arm-like mechanical manipulator, which is the basis of research currently being undertaken at Lancaster University, in order to address the complex tasks found in the rapidly expanding field of nuclear decommissioning. The requirements of multi-arm robot architecture for use in decommissioning tasks are discussed. The manipulators are integrated to work cooperatively and perform similar functions to humans in both scale and dexterity. The role that automation and robotics can play in enabling quicker demolition and at the same time reducing the exposure of workers to harmful radiation is examined. The key issues surrounding radioactive materials and safe dose levels are explained. The different stages of a particular system engineering process are outlined together with the essential physical steps. The paper will conclude by identifying the compliance of the system engineering used here with the requirements of designing a multi-arm robot.

A NOVEL HAPTIC INTERFACE FOR FREE LOCOMOTION IN EXTENDED RANGE TELEPRESENCE SCENARIOS

Patrick Rößler, Timothy Armstrong, Oliver Hessel, Michael Mende and Uwe D. Hanebeck

Telepresence gives a user the impression of actually being present in a distant environment. A mobile teleoperator acts as a proxy in this target environment, replicates the user’s motion, and records sensory information, which is transferred to the user and displayed in real-time. As a result the user is immersed in the target environment. The user can then control the the teleoperator by walking naturally. Motion Compression, a nonlinear mapping between the user’s and the robot’s motion, allows exploration of large target environments even from small user environments. For manipulation tasks haptic feedback is important. However, current haptic displays do not allow wide-area motion. In this work we present our design of a novel haptic display for simultaneous wide area motion and haptic interaction.

PARTICLE-FILTER APPROACH AND MOTION STRATEGY FOR COOPERATIVE LOCALIZATION

Fernando Gomez Bravo, Alberto Vale and Maria Isabel Ribeiro

This paper proposes a Particle-Filter approach and a set of motion strategies to cooperatively localize a team of three robots. The allocated mission consists on the path following of a closed trajectory and obstacle avoidance in isolated and unstructured scenarios. The localization methodology required for the correct path following relies on distance and orientation measurements among the robots and the robots and a fixed active beacon. Simulation results are presented.

ELLIPTIC NET - A PATH PLANNING ALGORITHM FOR DYNAMIC ENVIRONMENTS

Martin Saska, Miroslav Kulich and Libor Přeučil

Robot path planning and obstacle avoidance problems play an important role in mobile robotics. The standard algorithms assume that a working environment is static or changing slowly. Moreover, computation time and time needed for realization of the planned path is usually not crucial. The paper describes a novel algorithm that is focused especially to deal with these two issues: the presented algorithm - Elliptic Net is fast and robust and therefore usable in highly dynamic environments. The main idea of the algorithm is to cover an interesting part of the working environment by a set of nodes and to construct a graph where the nodes are connected by edges. Weights of the edges are then determined according to their lengths and distance to obstacles. This allows to choose whether a generated path will be safe (far from obstacles), short, or weigh these two criterions. The Elliptic Net approach was implemented, experimentally verified, and compared with standard path planning algorithms.

ROBUST AUGMENTED REALITY TRACKING BASED VISUAL POSE ESTIMATION

Madjid Maidi, Fakhr-Eddine Ababsa and Malik Mallem

In this paper, we present a robust fiducials tracking method for real time Augmented Reality systems. Our approach identifies the target object with an internal barecode of the fiducial and extracts its 2D features points. Given the 2D feature points and a 3D object model, object pose consists in recovering the position and the orientation of the object with respect to the camera. For pose estimation, we presented two methods for recovering pose using the Extended Kalman Filter and the Orthogonal Iteration algorithm. The first algorithm is a sequential estimator that predicts and corrects the state vector. While the later uses the object space collinearity error and derives an iterative algorithm to compute orthogonal rotation matrices. Due to lighting or contrast conditions or occlusion of the target object by an other object, the tracking may fail. Therefore, we extend our tracking method using a RANSAC algorithm to deal with occlusions. The algorithm is tested with different camera viewpoints under various image conditions and shows to be accurate and robust.

A TECHNIQUE FOR IMPERCEPTIBLE EMBEDDING OF DATA IN A COLOR IMAGE

Kaliappan Gopalan

A method of embedding data in a color image for applications such as authentication of an employee carrying a picture identification card is described. By converting the color image to a one-dimensional signal in red, green, or blue, audibly masked frequencies in the 1-D signal are determined for each segment or block. Embedding of data, such as key biometric information or other unique identification of the person, is carried out by modifying the spectral power at a pair of commonly occurring masked frequencies. Preliminary results show that the spectrum modification technique is simple to process and causes barely noticeable distortion in the embedded image. Using an oblivious technique and a key consisting of the frequencies where spectrum is modified, successful data retrieval with no bit errors has been achieved. Embedded image corrupted by low level noise still retained the hidden data with low bit errors. Higher payload of hidden data can be obtained at a cost of perceptibility of embedding.

COMBINING REINFORCEMENT LEARNING AND GENETIC ALGORITHMS TO LEARN BEHAVIOURS IN MOBILE ROBOTICS

R. Iglesias, M. Rodríguez, C. V. Regueiro, J. Correa and S. Barro

Reinforcement learning is an extremely useful paradigm which is able to solve problems in those domains where it is difficult to get a set of examples of how the system should work. Nevertheless, there are important problems associated with this paradigm which make the learning process more unstable and its convergence slower. In our case, to overcome one of the main problems (exploration versus exploitation trade off), we propose a combination of reinforcement learning with genetic algorithms, where both paradigms influence each other in such a way that the drawbacks of each paradigm are balanced with the benefits of the other. The application of our proposal to solve a problem in mobile robotics shows its usefulness and high performance, as it is able to find a stable solution in a short period of time. The usefulness of our approach is highlighted through the application of the system learnt through our proposal to control the real robot.

MISSION PLANNING, SIMULATION AND SUPERVISION OF UNMANNED AERIAL VEHICLE WITH A GIS-BASED FRAMEWORK

Pedro Gutierrez, Antonio Barrientos, Jaime del Cerro and Rodrigo San Martin

A framework for mission planning, simulation and supervision of unmanned aerial vehicles (UAV) has been developed. To provide a rich context for mission planning an Enhanced Reality is created from Geographic Information System (GIS) sources and dynamic aggregation of available geo-referenced data. The mission is expressed as statements and expressions of the Aerial Vehicle Control Language (AVCL), the abstraction mechanism needed to bridge the gap between a strategic mission planner and a heterogenous group of vehicles and active payloads. The framework is extendable by design and its aimed at the integration of diverse vehicles with existing systems. It has been tested as a Mission Planning and Simulation tool with our real-time small helicopter model.

TOWARDS A SLAM SOLUTION FOR A ROBOTIC AIRSHIP

Cesar Castro, Samuel Bueno and Alessandro Victorino

This article presents the authors ongoing work towards a six degrees of freedom simultaneous localization and mapping (SLAM) solution for the Project AURORA autonomous robotic airship. While the vehicle's mission is being executed in an unknown environment, where neither predefined maps nor satellite help are available, the airship has to use nothing but its own onboard sensors to capture information from its surroundings and from itself, locating itself and building a map of the environment it navigates. To achieve this goal, the airship sensorial input is provided by an inertial measurement unit (IMU), whereas a single onboard camera detects features of interest in the environment, such as landmark information. The data from both sensors are then fused using an architecture based on an extended Kalman filter, which acts as an estimator of the robot pose and the map. The proposed methodology is validated in a simulation environment, composed of virtual sensors and the aerial platform simulator of the AURORA project based on a realistic dynamic model. The results are hereby reported.

A HYBRID FEEDBACK CONTROLLER FOR CAR-LIKE ROBOTS - Combining Reactive Obstacle Avoidance and Global Replanning

Matthias Hentschel, Oliver Wulf and Bernardo Wagner

This paper presents a hybrid feedback controller for path control of autonomous mobile robots. The controller combines reactive obstacle avoidance with global path replanning, enabling collision-free navigation along a preplanned path. Avoidance of local obstacles is accomplished by adjusting the vehicle’s lateral deviation from the path trajectory reactively. Global path replanning is performed to circumvent obstacles which cannot be avoided locally. In contrast to common approaches, this is done by searching an optimal path returning to the initial trajectory beyond the obstacle. Following the description of the hybrid feedback controller, experimental results will demonstrate the effectiveness of this approach.

ESTIMATION OF PERFORMANCE OF HEAVY VEHICLES BY SLIDING MODES OBSERVERS

N. K. M’Sirdi, A. Boubezoul, A. Rabhi and L. Fridman

The main objective of this work, is performance handling and maneuverability, by means of the observation of vehicle dynamics in order to obtain safer and an easier driving. A model is proposed to describe the dynamic of a tractor and semi-trailer. The model is developed for cornering manoeuvre at constant speed. First and second order sliding mode observers are developed to estimate the vehicle state. Lateral forces are estimated in a last step. Simulation results are compared to validate the approach.

ESTIMATION OF ROAD PROFILE USING SECOND ORDER SLIDING MODE OBSERVER

A. Rabhi, N. K. M’Sirdi, M. Ouladsine and L. Fridman

This paper presents an algorithm to estimate the road profile. This method is based on a robust observer designed with a nominal dynamic model of vehicle. The estimation has been validated experimentally using a trailer equipped with position sensors and accelerometers.

TOWARD 3D FREE FORM OBJECT TRACKING USING SKELETON

Djamel Merad and Jean-Yves Didier

In this paper we describe an original method for the 3D free form object tracking in monocular vision. The main contribution of this article is the use of the skeleton of an object in order to recognize, locate and track this object in real time. Indeed, the use of this kind of representation made it possible to avoid difficulties related to the absence of prominent elements in free form objects (which makes the matching process easier). The skeleton is a lower dimension representation of the object, it is homotopic and it has a graph structure. This allowed us to use powerful tools of the graph theory in order to perform matching between scene objects and models (recognition step). Thereafter, we used skeleton extremities as interest points for the tracking.

AN ENERGY-BASED BACKGROUND MODELLING ALGORITHM FOR MOTION DETECTION

Paolo Spagnolo, Marco Leo, Tiziana D’Orazio, Andrea Caroppo and Tommaso Martiriggiano

Detecting moving objects is very important in many application contexts such as people detection, visual surveillance, automatic generation of video effects, and so on. The first and fundamental step of all motion detection algorithms is the background modeling. The goal of the methodology here proposed is to create a background model substantially independent from each hypothesis about the training phase, as the presence of moving persons, moving background objects, and changing (sudden or gradual) light conditions. We propose an unsupervised approach that combines the results of temporal analysis of pixel intensity with a sliding window procedure to preserve the model from the presence of foreground moving objects during the building phase. Moreover, a multilayered approach has been implemented to handle small movements in background objects. The algorithm has been tested in many different contexts, such as a soccer stadium, a parking area, a street, a beach. Finally, it has been tested even on the CAVIAR 2005 dataset

EYE AND GAZE TRACKING ALGORITHM FOR COLLABORATIVE LEARNING SYSTEM

Djamel Merad, Stephanie Metz and Serge Miguet

Our work focuses on the interdisciplinary field of detailed analysis of behaviors exhibited by individuals during sessions of distributed collaboration. With a particular focus on ergonomics, we propose new mechanisms to be integrated into existing tools to enable increased productivity in distributed learning and working. Our technique is to record ocular movements (eye tracking) to analyze various scenarios of distributed collaboration in the context of computer-based training. In this article, we present a low-cost oculometric device that is capable of making ocular measurements without interfering with the natural behavior of the subject. We expect that this device could be employed anywhere that a natural, non-intrusive method of observation is required, and its low-cost permits it to be readily integrated into existing popular tools, particularly E-learning campus.

A MULTI-AGENT COLLABORATIVE CONTROL ARCHITECTURE WITH FUZZY ADJUSTMENT FOR A MOBILE ROBOT

Bianca Innocenti, Beatriz López and Joaquim Salvi

One of the current challenges of control research is to make systems capable of showing intelligent responses to changing circumstances. To address this task, more complex systems are being developed. However, it is technologically difficult and potentially dangerous to build complex systems that are controlled in a completely centralized way. One approach to building decentralization systems is using multi-agent technology for building control architectures. But it seems risky to recursively extend using multi-agent systems to develop part of the system, such as a single behaviour, when it becomes complex. One alternative approach is to use collaborative control to deploy specific (low level) behaviours, so that several controllers are combined in a single agent of the multi-agent architecture in order to achieve the behaviour wanted. This paper is related to this idea: a multi-agent architecture of collaborative control is defined. Each agent represents a behaviour which in turn is implemented by means of collaborative control. The experiments were carried out using a Pioneer mobile robot.

REDUCING ACCUMULATED ERRORS IN EGO-MOTION ESTIMATION USING LOCAL BUNDLE ADJUSTMENT

Akihiro Sugimoto and Tomohiko Ikeda

The binocular independent fixation control is a camera control of two mounted active cameras where each of them independently and automatically fixates its optical axis to its own fixation point. A method using this camera control was proposed to incrementally estimate ego-motion from two time-series images. The method, however, has a problem that estimation accuracy gradually becomes worse as the motion trajectory becomes longer and longer. This is due to accumulation of estimation errors incurred in each estimation step. To keep estimation accuracy stable even for a long trajectory, we propose to locally apply the bundle adjustment to each estimated motion so that the modified estimation becomes geometrically consistent with time-series images captured so far. This modification realizes stable estimation of a long motion trajectory.

SEMIOTICS AND HUMAN-ROBOT INTERACTION

João Silva Sequeira and Maria Isabel Ribeiro

This paper describes a robot control architecture supported on a human-robot interaction model obtained directly from semiotics concepts. The architecture is composed of a set of objects defined after a semiotic sign model. Simulation experiments using unicycle robots are presented that illustrate the interactions within a team of robots equipped with skills similar to those used in human-robot interactions.

A HOLONIC FAULT TOLERANT MANUFACTURING PLATFORM WITH MULTIPLE ROBOTS

Theodor Borangiu, Florin Daniel Anton, Silvia Tunaru and Anamaria Dogar

To be competitive, manufacturing should adapt to changing conditions existing in the market. The greater variety of products, the possible large fluctuations in demand, the shorter lifecycle of products expressed by a higher dynamics of new products, and the increased customer expectations in terms of quality and delivery time are challenges that manufacturing companies have to deal with to remain competitive. Besides these market-based challenges, manufacturing firms also need constantly to adapt to newly developed processes and technologies and to rapidly changing environmental protection regulations. Modern automated manufacturing systems need robotized material-conditioning systems able to move materials efficiently throughout the entire production area. This involves not only moving and storing materials, but also identifying, locating, qualifying, controlling and measuring them during processing and transportation. The objective of the proposed project is the design, implementing, testing and validation of a holonic, fault-tolerant manufacturing control platform integrating multiple robots with visual guidance for on demand material conditioning and automated visual inspection.

LOGGING, ALERT & EMERGENCY SYSTEM FOR ROAD TRANSPORT VEHICLES - An Experimental eCall, Black-box and Driver Alerting System

Javier Fernández, Fernando Cantalapiedra, Mario Mata, Veronica Egido and Sergio Bemposta

This paper describes the experimental platform developed at UEM, mounted on a conventional vehicle. It monitors most of the driver’s actions on the controls of the vehicle, logs the vehicle speed and position using a GPS, detects and recognizes vertical traffic signs, and records the last seconds of the trip with a panoramic video camera. If an accident occurs, the system calls emergency services (112 in Spain) sending vehicle position information (via SMS) and opening a voice channel.

A NEW HYBRID SAMPLING STRATEGY FOR PRM PLANNERS - To Address Narrow Passages Problem

Sofiane Ahmed Ali, Eric Vasselin and Alain Faure

The probabilistic path planner (PPP) is a general planning scheme that yields fast robot path planners for a wide variety of problems, involving high degree of freedom articulated robots, non holonomic robots, and multiple robots. This paper presents a new probabilistic approach for finding paths through narrow passages. Our probabilistic planner follows the general framework of probabilistic roadmap (PRM), but to increase sample density in difficult areas like narrow passages, we define two sampling constraints in order to get much more points than a classic PRM gets in such areas. We simulate our planner in 2D environments and the simulations results shows good performance for our planner.

DEVELOPMENT OF HIGH PERFORMANCE SERVO DRIVE/ANTI DRIVE MECHANISM FOR BACKLASH REMOVAL

I. Askari, S. A. Hassan, M. Altaf, A. Azim, M. B. Malik and K. Munawar

In electromechanical drives, there is always a backlash between any pair of gears. Due to this, it is almost impossible to realize a high accuracy and high performance drive. However such drives are crucial in today’s modern electromechanical systems. A high performance drive/anti-drive servomechanism is developed to eliminate the effect of backlash. The concept utilizes redundant unidirectional drives to assure positive coupling of gear meshes at all times. Based on this concept, a methodology for enumeration of admissible redundant-drive backlash free mechanism has been established. The angular displacement is achieved as a difference of two torques. These torques can be controlled by a high performance control system. A controller model will be designed to move a single degree of freedom platform up to a desired span with a payload.

OPTIMAL PLANNING FOR AUTONOMOUS AGENTS UNDER TIME AND RESOURCE UNCERTAINTY

Aurélie Beynier, Laurent Jeanpierre and Abdel-Illah Mouaddib

In this paper we develop an approach for optimal planning under time and resource uncertainty with complex task dependencies. We follow the line of research described by (Bresina, 2002), overcoming limitations of existing approaches: they only handle simple time constraints and they assume a simple model of uncertainty concerning action durations and resource consumptions. In many domains such as space applications (rovers, satellites), these assumptions are not valid. We present an approach that considers temporal and resource constraints and deals with uncertainty about the durations and resource consumptions. From an acyclic mission graph, we automatically build a Markov Decision Process that can be optimally solved by dynamic programming. Experimental results prove that this approach allows for considering large mission graphs.

RANGE DETERMINATION FOR MOBILE ROBOTS USING ONE OMNIDIRECTIONAL CAMERA

Ola Millnert, Toon Goedemé, Tinne Tuytelaars, Luc Van Gool, Alexander Hüntemann and Marnix Nuttin

We propose a method for computing the absolute distances to obstacles using only one omnidirectional camera. The method is applied to mobile robots. We achieve this without restricting the application to predetermined translations or the use of artificial markers. In contrast to prior work, our method is able to build absolute scale 3D without the need of a known baseline length, traditionally acquired by an odometer. Instead we use the ground plane assumption together with the camera system's height to determine the scale factor. Using only one omnidirectional camera our method is proven to be cheaper, more informative and more compact than the traditional methods for distance determination, especially when a robot is already equipped with a camera for e.g. navigation. An additional advantage is that it provides more information since it determines distances in a 3D space instead of one plane. The experiments show promising results. The algorithm is indeed capable of determining the distances in meters to features and obstacles and is able to locate all major obstacles in the scene.

OPTICAL FLOW NAVIGATION OVER ACROMOVI ARCHITECTURE

Patricio Nebot and Enric Cervera

Optical flow computation involves the extraction of a dense velocity field from an image sequence. The purpose of this work is to use the technique of optical flow so that a robot equipped with a color camera can navigate in a secure way through an indoor environment without collide with any obstacle. In order to implement such application, the Acromovi architecture has been used. Acromovi architecture is a distributed architecture that works as middleware layer between the robot architecture and the applications, which allows sharing the resources of each robot among all the team. This middleware is based on an agent-oriented approach.

POWER ESTIMATION FOR REGISTER TRANSFER LEVEL BY GENETIC ALGORITHM

Yaseer A. Durrani, Teresa Riesgo and Felipe Machado

In this paper, we propose a new genetic algorithm (GA) based macromodeling technique for the power dissipation of intellectual property (IP) components to their statistical knowledge of the primary inputs. Our technique can handle combinational and sequential circuits at register transfer level. During power estimation, the sequence of input stream is generated by a GA using input metrics. Monte Carlo zero delay simulation is performed and power dissipation is predicted by a macromodel function. In experiments with IP macro-blocks, the results are effective and highly correlated, with an average error of 1%. Our model is parameterizable and provides accurate power estimation.

MANAGING CONTROL ARCHITECTURES DESIGN PROCESS - Patterns, Components and Object Petri Nets in Use

Robin Passama, David Andreu, Christophe Dony and Thérèse Libourel

The paper presents a methodology for the development of robot software controllers, based on actual software component approaches and robot control architectures. This methodology defines a process that guides developers from the analysis of a robot controller to its execution. A proposed control architecture pattern, useful for analysis and integration of expertise during design process is presented. A dedicated component-based language, focusing on reusability and upgradeability of controller architectures parts, proposes notations to design and implements software architectures.

REACTIVE SIMULATION FOR REAL-TIME OBSTACLE AVOIDANCE

Mariolino De Cecco, Enrico Marcuzzi, Luca Baglivo and Mirco Zaccariotto

This paper provides a new approach to the dynamic path planning and obstacle avoidance in unknown and dynamic environments. The system is based on the interaction between four different modules: the Path Planner, a Graph, the “Sentinel”, and the module which computes the Reactive Simulation. The reactive simulation takes in account the kinematics model of the vehicle and the actual state conditions to make a real-time simulation in order to predict the trajectory of the differential drive robot that would allow the safe reaching of the local target.

THE LAGR PROJECT - Integrating Learning into the 4D/RCS Control Hierarchy

James Albus, Roger Bostelman, Tsai Hong, Tommy Chang, Will Shackleford and Michael Shneier

The National Institute of Standards and Technology’s (NIST) Intelligent Systems Division (ISD) has been a part of the Defense Advanced Research Project Agency (DARPA) LAGR (Learning Applied to Ground Ro-bots) Project. The NIST team’s objective for the LAGR Project is to insert learning algorithms into the modules that make up the 4D/RCS (Four Dimensional/Real-Time Control System), the standard reference model architecture to which ISD has applied to many intelligent systems. This paper describes the 4D/RCS structure, its application to the LAGR project, and the learning and mobility control algorithms used by the NIST team’s vehicle.

A PERFORMANCE METRIC FOR MOBILE ROBOT LOCALIZATION

Antonio Ruiz-Mayor, Gracián Triviño and Gonzalo Bailador

This paper focus on the particular problem of how to measure in a reproducible way the localization precision of Mobile Robots. In particular the localization algorithms that match the general prediction-correction loop model are considered. We propose a performance metric based on formalizing the error sources. The metric exposition is illustrated with an example of localization algorithm for a real mobile robot. This metric differs from others in the fact that it fulfils at the same time the following properties: to effectively measure the estimation error of a pose estimation algorithm, to be reproducible, to clearly separate the contribution of the correction part from the prediction part of the algorithm, and to make easy the analysis of the algorithm performance respect to the great number of influencing factors. The proposed metric will help the developers of Mobile Robots to validate the localization algorithms in a systematic and standard way, reducing design time and workload.

A SPECIFIC LOCOMOTION INTERFACE FOR VIRTUAL REALITY - Design of a Wheelchair Type Haptic

Cédric Anthierens, Jean-Luc Impagliazzo, Yves Dupuis and Eric Richard

This paper presents a recent advance in Virtual Reality (VR) related to building design or development of public places. Indeed, it focuses on design and implementation of a wheelchair type haptic to simulate difficulties of access and displacement of a disabled person in a wheelchair within these places. A VR platform equipped by this haptic system provides a Virtual Environment (VE) which represents either a street scene or an interior building scene. This VE should be useful for architects who want to evaluate, from the design phase, facilities efficiency of their design dedicated to disabled persons. The first part of this paper deals with the lack of consideration about the accessibility of disabled persons in everyday places and thus needs to improve the built facilities quoted by a dedicated law. The second part deals with specifications and expected results of the VR platform. Next part will focus on the mechatronical design and explains how each part of the interface works to finally satisfy a good rendering and a high level of realism according to our main goal defined before. The implementation phase and integration of this specific behavioural interface into the VR platform will be presented in the fourth part. The last part, just before the conclusion and perspectives, discusses tests and final results obtained with a total immersion within a simulated interior environment.

PERFORMANCE ANALYSIS OF CSMA/CA PROTOCOL IN IEEE 802.11 NETWORKS USING BACKOFF MECHANISM

Amith M. N.

The distributed coordination function (DCF) in the IEEE 802.11 standard for wireless LAN is based on the Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) like medium access control protocol. This collision avoidance is implemented by means of backoff procedure which uses a rotating window mechanism. This paper presents a simulation analysis of the slot selection probabilities and the backoff mechanism. The obtained simulations allow us to determine the effective throughput versus offered load for different values of the contention window parameter and the number of the contending stations. The choice of CWmin and CWmax parameters are analyzed.

INTERACTION CONTROL EXPERIMENTS FOR A ROBOT WITH ONE FLEXIBLE LINK

L. F. Baptista, J. M. M. Martins and J. M. G. Sá da Costa

One of the major drawbacks of flexible-link robot applications is its low tip precision, which is an essential characteristic for applications with interaction control with a contact surface. In this experimental work, interaction control strategies considering rigid and flexible contact surfaces are applied on a two degrees of mobility flexible-link manipulator. The applied strategies are based on the closed-loop inverse kinematics algorithm (CLIK) to obtain the angular references to the joint position controller. The control schemes were previously tested by simulation and further implemented on the flexible-link robot. The obtained experimental results exhibit an excellent force tracking performance and reveal the successful implementation of this control architecture to real interaction control tasks.

NEUROLOGICAL AND ENGINEERING APPROACHES TO HUMAN POSTURAL CONTROL

Karim Tahboub, Thomas Mergner and Christoph Ament

This paper discusses the human postural control as a system engineering approach problem. Two main perspectives are considered: neurological and engineering. From the neurological perspective, the problem is described, main sensory systems are identified, sensor fusion is suggested, and control system architecture and details are presented. Experimental results on both human subjects and on a special-purpose humanoid agree with the presented architecture. On the other hand, the humanoid parameters are identified, the humanoid dynamic model is derived, external-disturbance estimation methods are presented, a control method for stabilizing the body motion and then for robust tracking of voluntary motion in the presence of external disturbances is shown. This constitutes an engineering approach to this problem. Simulation results are given and it is shown that the presented method is capable of estimating the disturbances and for controlling the motion

EXTRACTION OF SIGNIFICANT REGIONS IN COLOR IMAGES FOR LANDMARK IDENTIFICATION

Jose-Luis Albarral and Enric Celaya

To make vision-based robot navigation possible in outdoor environments, a robot must be able to detect and characterize relevant landmarks found in the environment so that they can be recognized later on. Our approach to landmark characterization involves a first processing of the image in which color-salient regions are selected as candidates that can possibly contain a landmark. A more detailed examination of these candidate regions is then performed in order to characterize a new landmark, or to recognize a previously seen one. In our approach, a region of interest is processed to extract the most relevant color-uniform regions, and obtain their spatial moments and topological structure that permit the landmark characterization. For this reason we have implemented a new segmentation algorithm fast and robust enough to be used by our landmark characterization system.

A NEW SENSORIAL AND DRIVING LOCOMOTION INTERFACE FOR VIRTUAL REALITY

Yves Dupuis, Jean-Luc Impagliazzo, Cédric Anthierens and Dominique Millet

This paper deals with the design of a 1D sensorial and driving locomotion interface for Virtual Reality applications able to simulate natural walking-in-place. The aim is to provide an unlimited roaming in a virtual world while physically walking in a constrained area. Most of existing locomotion interfaces do not allow to walk naturally in terms of steps length and frequency. Furthermore, we define the term “natural walking” in two complementary ways. The first one is devoted to biomechanical features of human walking, that is to say the position, speed and acceleration of human body parts. The second one is related to self-movement perception, namely the integration of multi-sensorial information such as kinaesthetic, visual and vestibular information. So, we designed our mechatronical interface using biomechanical and sensorial data of human walking. The interface is equipped with sensors in order to measure floor reaction forces onto the pedals and a video tracking device to measure the current positions of user’s feet. Since the program has been written in C++ language, it is easy to create new automata to control the interface for other applications such as running. Finally, the implementation of the interface with the virtual environment is described.

COOPERATIVE MAP BUILDING USING QUALITATIVE REASONING FOR SEVERAL AIBO ROBOTS

David A. Graullera, Salvador Moreno and Maria Teresa Escrig

The problem that a robot navigates autonomously through its environment, builds its own map and localizes itself in the map, is still an open problem. It is known as the SLAM (Simultaneous Localization and Map Building) problem. This problem is made even more difficult when we have several robots cooperating to build a common map of an unknown environment, due to the problem of map integration of several submaps built independently by each robot, and with a high degree of error, making the map matching specially difficult. Most of the approaches to solve map building problems are quantitative, resulting in a great computational cost and a low level of abstraction. In order to fulfil these drawbacks qualitative models have been recently used. However, qualitative models are non deterministic. Therefore, the solution recently adopted has been to mix both qualitative and quantitative models to represent the environment and build maps. However, no reasoning process has been used to deal with the information stored in maps up to now, therefore maps are only static storage of landmarks. In this paper we propose a novel method for cooperative map building based on hybrid (qualitative+quantitative) representation which includes also a reasoning process. Distinctive landmarks acquisition for map representation is provided by the cognitive vision and infrared modules which compute differences from the expected data according to the current map and the actual information perceived. We will store in the map the relative orientation information of the landmarks which appear in the environment, after a qualitative reasoning process, therefore the map will be independent of the point of view of the robot. Map integration will then be achieved by localizing each robot in the maps made by the other robots, through a process of pattern matching of the hybrid maps elaborated by each robot, resulting in an integrated map which all robots share, and which is the main objective of this work. This map building method is currently being tested on a team of Sony AIBO four legged robots.

PARTIAL STABILIZABILITY OF CASCADED SYSTEMS APPLICATIONS TO PARTIAL ATTITUDE CONTROL

Chaker Jammazi and Azgal Abichou

In this work, the problem of partial stabilization of nonlinear control cascade systems with integrators is considered. The latter systems present an anomaly, which is the non complete stabilization via continuous pure-state feedback, this is due to Brockett necessary condition. To cope with this difficulty we propose the partial stabilization. For a given motion of a dynamical system, say $x(t,\,x_0,\,t_0)=(y(t,\,y_0,\,t_0),\,z(t,\,z_0,\,t_0))$, the partial stabilization is the qualitative behavior of the $y$-component of the motion (i.e the asymptotic stabilization of the motion with respect to $y$) and the $z$-component converges, relative to to the initial vector $x(t_0)=x_0=(y_0,\,z_0)$. In the present work, we establish a new results for the adding integrators for partial stabilization, we show that if the control systems is partially stabilizable, then the augmented cascade system is partially stabilizable. Two applications are considered. The first one is devoted to partial attitude stabilization of rigid spacecraft. The second application is intended to the study of underactuated ship. Numerical simulations are given to illustrate our results.

ACO BASED METHOD COMPARATION APPLIED TO FLEET MANAGEMENT PROBLEM

Miriam Anton-Rodriguez, Daniel Boto-Giralda, Francisco J. Diaz-Pernas and J. Fernando Diez-Higuera

Road Transport enterprises do have the need of fleet management applications in order to upgrade their efficiency; the fulfilment of that need takes us in the search of optimization algorithms whose performance better suits not only the optimal route search problem, but the resource allocation too. ACO (Ant Colony Optimization) meta-heuristic has proven to be very useful when solving similar problems, but as ACO comes in several different flavours, to make the right algorithm choice is the first step in the search for a solution. This document presents a performance study made upon several ACO algorithms over the fleet management problem, with the objective of determining which one is the best finding the optimal solution in a reasonable amount of time.

MANAGEMENT OF A MULTICAMERA TRACKING SYSTEM

C. Motamed and R. Lherbier

This work is linked with the context of human activity monitoring and concerns the development of a multi-camera tracking system. Our strategy of sensors combination integrates the contextual suitability of each sensor with respect to the task. The suitability of a sensor, represented as a belief indicator, combines two main criteria. Firstly it is based on the notion of spatial “isolation” of the tracked object with respect to other object and secondly on the notion of “visibility” with respect to the sensor. A centralized filter combines the result of the local tracking estimations (sensor level) and then performs the track management. The main objective of the proposed architecture is to deal with the limitation of each local sensor with respect to the problem of visual occlusion.

GLOBAL OPTIMIZATION OF PERFORMANCE OF A 2PRR PARALLEL MANIPULATOR FOR COOPERATIVE TASKS

Héctor A. Moreno, J. Alfonso Pámanes, Philippe Wenger and Damien Chablat

In this paper the trajectory planning problem is solved for a 2PRR parallel manipulator which works in cooperation with a 1 degree-of-freedom (dof) platform. The whole kinematic chain is considered as a redundant 3-dof manipulator, and an algorithm is presented to solve the redundancy by using the joint velocities in the null space of the jacobian matrix. The internal motion of the assisted manipulator allows globally optimize the condition number of the jacobian matrix during the accomplishment of a desired task. Consequently, the accuracy of the manipulator is maximized and singular or degenerate poses are avoided. A case of study is presented to show the effectiveness of our approach.

A SEMANTICALLY RICH POLICY BASED APPROACH TO ROBOT CONTROL

Matthew Johnson, Jeffery Bradshaw, Paul Feltovich, Renia Jeffers, Hyuckchul Jung and Andrzej Uszok

In this paper we describe our approach to enhancing control of robotic systems by providing domain and policy services via KAoS. Recently developed languages such as OWL provide a powerful descriptive logic foundation that can be used to express semantically rich relationships between entities and actions, and thus create complex context sensitive policies. KAoS provides a tool to create policies using OWL and an infrastructure to enforce these policies on robots. We contend that a policy-based approach can provide significant advantages in controlling robotic systems and are a much more natural way for operators to manage multiple robots.

MULTIPROCESSOR ROBOT CONTROLLER - An Experimental Robot Controller for Force-Torque Control Tasks

István Oláh and Gábor Tevesz

There is an ongoing research and development on the field of hybrid position and force control of the assembly robots at the Department of Automation and Applied Informatics and the Department of Control Engineering and Information Technology, Budapest University of Technology and Economics. As a result an Experimental Robot Controller was built for the special needs of the project. Both the hardware and the software system of the controller are under continuous development. As the most recent achievement, a six-component force-torque sensor is built in and the development on the software system is currently dealing with the extension of the programming environment with force-torque control possibilities. There are numerous industrial applications for force and torque control (i.e. screw driver, welder), but a flexible equipment can provide much more than the possibility of solving a single task. This paper presents the overview of the robot controller hardware, separately detailing the force-torque sensor interface. The second half of this paper overviews the software system of the controller and the possibilities of its extensions for force control tasks.

ROBOT BEHAVIOR ADAPTATION FOR FORMATION MAINTENANCE

Maite López-Sánchez

Most often, autonomous robots maintain group formations by using global information such as the position of the group leader or even the position of all robots inside the formation. Alternative approaches to autonomous robot formations have considered local information, which is more realistic but presents some drawbacks such as troop deformation. In this paper we perform a step forward in local information usage for formation maintenance by analyzing a parameterization of different basic behaviors. Formation maintenance emerges from the combination of these simple behaviors, and its overall accuracy is empirically optimized by tuning behavior parameters. In particular, we study and characterize three different formations: queue or column (as for ants), inverted V or wedge (as for birds or planes) and rectangle (for manipulus antique roman troop formations). This paper describes simulated robots that incorporate a unique set of basic behaviors from which formation maintenance emerges. These simple behaviors provide robustness to the formation (that is, robots do not get lost) but do not prevent from formation deformations. Tackling this problem, we have identified a set of behavior parameters that we empirically tune in order to optimize the overall performance (that is, to minimize errors).

PERFORMANCE EVALUATION OF A CONTROLLED FLOW-SHOP SYSTEM WITH A TIMED PETRI NET MODEL

Loïc Plassart, Philippe Le Parc, Frank Singhoff and Lionel Marcé

This paper presents an original performance analysis applied to a flow-shop system driven by a set of local command units and a central controller. The performance evaluation is done with a timed coloured Petri net model. Simulation results show needs for bounding the controller response time in order to meet production targets.

PATTERN TRACKING AND VISUAL SERVOING FOR INDOOR MOBILE ROBOT ENVIRONMENT MAPPING AND AUTONOMOUS NAVIGATION

O. Ait Aider, G. Blanc, Y. Mezouar and P. Martinet

The paper describes a complete framework for autonomous environment mapping, localization and navigation using exclusively monocular vision. The environment map is a mosaic of 2D patterns detected on the ceiling plane and used as natural landmarks. The robot is able to localize itself and to reproduce learned trajectories defined by a set of key images representing the visual memory. A specific multiple 2D pattern tracker was developed for the application. It is based on particle filtering and uses both image contours and gray scale level variations to track efficiently 2D patterns even on cluttered ceiling appearance. When running autonomously, the robot is controlled by a visual servoing law adapted to its nonholonomic constraint. Based on the regulation of successive homographies, this control law guides the robot along the reference visual route without explicitly planning any trajectory. Real experiment results illustrate the validity of the presented framework.

A PATH PLANNING STRATEGY FOR OBSTACLE AVOIDANCE

Guillaume Blanc, Youcef Mezouar and Philippe Martinet

This paper presents an obstacle avoidance module dedicated to non-holonomic wheeled mobile robots. Chained system theory and deformable virtual zone principle are coupled to design an original framework based on path following formalism. The proposed strategy allows to correct the control output provided by a navigation module to preserve the robot security while assuring the navigation task. First, local paths and control inputs are derived from the interaction between virtual zones surrounding the robot and obstacles to efficiently prevent from collisions. The resulting control inputs and the ones provided by the navigation module are then adequately merged to ensure the success of the navigation task. Experimental results using a cart-like mobile robot equipped with a sonar sensors belt confirm the relevance of the approach.

ROBUST POSTURE CONTROL OF A MOBILEWHEELED PENDULUM MOVING ON AN INCLINED PLANE

Danielle Sami Nasrallah, Hannah Michalska and Jorge Angeles

The paper considers a specific class of wheeled mobile robots referred to as mobile wheeled pendulums (MWP). Robots pertaining to this class are composed of two wheels rotating about a central body. The main feature of the MWP pertains to the central body, which can rotate about the wheel axes. As such motion is undesirable, the problem of the stabilization of the central body in MWP is crucial. The novelty of the work presented here resides in the construction of a three-imbricated loop controller that delivers the full control strategy for the robot posture and copes with parameters uncertainties. Simulations on the performance of the controlled system are provided.

MTR: THE MULTI-TASKING ROVER - A New Concept in Rover Design

Antonios K. Bouloubasis, Gerard T. McKee, Paul M. Sharkey and Peter Tolson

In this paper we present the novel concepts incorporated in a planetary surface exploration rover design that is currently under development. The Multitasking Rover (MTR) aims to demonstrate functionality that will cover many of the current and future needs such as rough-terrain mobility, modularity and upgradeability. The rover system has enhanced mobility characteristics. It operates in conjunction with Science Packs (SPs) and Tool Packs (TPs) – modules attached to the main frame of the rover, which are either special tools or science instruments and alter the operation capabilities of the system.

SMOOTH TRAJECTORY PLANNING FOR FULLY AUTOMATED PASSENGERS VEHICLES - Spline and Clothoid based Methods and its Simulation

Larissa Labakhua, Urbano Nunes, Rui Rodrigues and Fátima S. Leite

A new approach for mobility, providing an alternative to the private passenger car, by offering the same flexibility but with much less nuisances, is emerging, based on fully automated electric vehicles. A fleet of such vehicles might be an important element in a novel individual, door-to-door, transportation system to the city of tomorrow. For fully automated operation, trajectory planning methods that produce smooth trajectories, with minimum associated jerk for providing passenger´s comfort, is required. This paper addresses this problem proposing a new approach that consists of introducing a velocity planning stage to generate adequate time sequences to be fed into the interpolating curve planners. Moreover, the generated speed profile can be merged into the trajectory for usage in trajectory-tracking tasks like it is described in this paper, or it can be used separately (from the generated 2D curve) for usage in path-following tasks. Three trajectory planning methods, aided by the speed profile planning, are analysed from the point of view of passengers' comfort, implementation facility, and trajectory tracking.

IMPROVED METHOD FOR HIGHLY ACCURATE INTEGRATION OF TRACK MOTIONS

Michael Kleinkes, Werner Neddermeyer and Michael Schnell

Modern Robotics today deals with increasing requirements on the flexible automation. One of this is the usage of linear tracks or even called 7$^{th}$ axis to extend the robots workspace. The inaccuracies of the linear track deteriorate the accuracy, which is in constrast to highly accurate robot systems needed for modern applications. To enhance the accuracy of the system consisting of robot and linear track, an identification of the non-linearities of the linear track is necessary. This article introduces an optimisation of a method for highly accurate integration of track motions where the profile of the linear track is identified by single coordinate systems along the track, combined by a cubic spline interpolation. Resulting there is a continous description of the track profile, depending on the current position of the robot on the linear track.

MULTIMODELLING STEPS FOR FREE-SURFACE HYDRAULIC SYSTEMS CONTROL

Eric Duviella, Philippe Charbonnaud and Pascale Chiron

The paper presents multimodelling steps for the design of free-surface hydraulic system control strategies. This method is proposed to represent simply and accurately the non-linear hydraulic system dynamics under large operating conditions. It is an interesting alternative to the use of Saint Venant partial differential equations because it allows the design, the tuning and the validation of control strategies. The multimodelling steps of the proposed method are performed in order to lead to the determination of a finite number of models. The models are selected on-line by the minimization of a quadratic criterion. The evaluation of the multimodelling method is carried out by simulation within the framework of a canal with trapezoidal profile.

BAYESIAN INFERENCE IN A DISTRIBUTED ASSOCIATIVE NEURAL NETWORK FOR ADAPTIVE SIGNAL PROCESSING

Qianglong Zeng and Ganwen Zeng

The primary advantages of high performance associative memory model are its ability to learn fast, to store correctly and to retrieve information similar to the human “content addressable” memory and it can approximate a wide variety of non-linear functions. Based on a distributed associative neural network, a Bayesian inference probabilistic neural network are designed, the learning algorithm and the underlying basic mathematical idea are presented for the adaptive noise cancellation. Simulation results using speech corrupted with low signal to noise ratio in telecommunication environment shows great signal enhancement. A system based on the described method can store words and phrases spoken by the user in a communication channel and subsequently recognize them when they are pronounced as connected words in a noisy environment. The method guarantees system robustness in respect to noise, regardless of its origin and level. New words, pronunciations, and languages can be introduced to the system in an incremental, adaptive mode.

DESIGN OF MAX-PLUS CONTROL LAWS TO MEET TEMPORAL CONSTRAINTS IN TIMED EVENT GRAPHS

Saïd Amari, Jean Jacques Loiseau, Claude Martinez and Isabel Demongodin

The aim of the work presented is the control of timed event graph in order to meet tight temporal constraints. The temporal constraint represents the maximal duration of a chemical or thermal treatment, for instance. We formulate the problem in terms of control of linear Max-Plus models. A method for the synthesis of a control law ensuring the meeting of constraints is first described for a single input single constraint. Then, the single input multi constraint problem is tackled and finally, the method is extended to the multi inputs, multi constraints problem. The proposed method is illustrated on an example.

GA-BASED APPROACH TO PITCH RECOGNITION OF MUSICAL CONSONANCE

Masanori Natsui, Shunichi Kubo and Yoshiaki Tadokoro

This paper presents a novel method for the pitch recognition of the musical consonance (i.e., unison or octave) using genetic algorithm (GA). GA is a kind of optimization techniques based on natural selection and genetics. In our method, the pitch recognition is performed by the following two-step procedure: (i) search space reduction using the comb filter estimation, and (ii) evolutionary parameter estimation of tone parameters such as notes and volumes by minimizing error between a target waveform and a synthesized waveform using sound templates with estimated parameters. The potential capability of the system is demonstrated through the pitch estimation of randomly-generated consonances. Experimental results show that the system can successfully estimate chords with more than 84% success rate for two-note consonances, and more than 71% success rate for three-note consonances.

PITCH ESTIMATION OF DIFFICULT POLYPHONY SOUNDS OVERLAPPING SOME FREQUENCY COMPONENTS

Yoshiaki Tadokoro, Masanori Natsui, Yasuhiro Seto and Michiru Yamaguchi

There are some difficult polyphony to estimate these pitches for transcription. This paper proposes two new methods for the pitch estimation of these difficult polyphony. One of them is based on the beat components of the polyphony analyzed by the short-time Fourier transform (STFT). The other is a method noticing the period of the residual signal after the elimination of polyphony components using a comb filter ( ). These methods are based on the fact that there is a small frequency difference between the real sound and the ideal one.

AUTOMATA BASED MODELLING AND SIMULATION - Application in an Industrial Software Environment

Vasileios Deligiannis and Stamatis Manesis

Research efforts were recently quite intense on handling systems whose behaviour of interest is determined by interacting continuous and discrete dynamics, that is the hybrid systems. Contemporary industrial systems are also hybrid and modelling such systems is always challenging. Hybrid automata and Petri Nets are the most used approaches to model hybrid systems. Despite academic efforts these two approaches did not meet wide acceptance when proposed for industrial use, mainly because they are application depended. In this paper, a recently proposed hyper-class of hybrid automata is presented, which seems to cover this weakness. Illustrating its use, an application of this new formulation method in an industrial software environment is given. The given example is taken from a chemical industry and uses PID controllers to control continuous variables, while the whole project was developed in a SCADA software platform.

TIMED DISCRETE-EVENT SYSTEM SUPERVISORY CONTROL FOR UNDER-LOAD TAP-CHANGING TRANSFORMERS

A. Afzalian, A. Saadatpoor and W. M. Wonham

Timed discrete-event systems (TDES) have so far not been used for modelling and control of electrical power system. Since these systems have both logical and temporal behavior, we propose to use TDES to address their control problems. Under-load tap-changing transformers (ULTC) which obviously have discrete-event behavior are widely used in transmission systems to take care of instantaneous variations in the load conditions in substations. ULTC may be controlled either automatically or manually. This paper discusses the modelling and synthesis of a timed discrete-event system supervisory controller for ULTC. Different modes of operation are considered and it is shown that the specifications are controllable and the closed loop control system is non-blocking. Protective system designers in electrical power systems can use the proposed approach to verify their required temporal and logical behavior.

DESIGN OF AN ITERATIVE LEARNING CONTROL FOR A SERVO SYTEM USING MULTI-DICTIONARY MATCHING PURSUIT

Iuliana Rotariu and Erik Vullings

Many motion systems repeatedly follow the same trajectory. However, in many cases, the motion system does not learn from tracking errors obtained in a previous cycle. Iterative Learning Control (ILC) resolves this issue by compensating for previous tracking errors, but it suffers from not being able to distinguish between tracking errors caused by machine dynamics versus errors caused by noise, and by trying to 'learn' the noise, additional errors are introduced. In this paper we address this issue by using the servo error signal by identifying the time-varying nonlinear effects, which can be learned and therefore improve the position accuracy, versus the stochastic effects, which cannot be learned. The identification of these effects is performed by means of time-frequency analysis of the servo error and therefore our goal is to obtain a high-resolution time-frequency energy distribution of the analyzed signal. Here we compare the servo error energy distribution by three means: (1) Wigner distribution; (2) adaptive signal decomposition over one dictionary of modulated versions of wavelets (simple atomic dictionary); (3) and by means of combining several simple atomic dictionaries into a complex atomic dictionary. We show that the latter approach leads to the highest-resolution energy distribution and tracking performance.

FEATURES EXTRACTION AND TRAINING STRATEGIES IN CONTINUOUS SPEECH RECOGNITION FOR ROMANIAN LANGUAGE

Corneliu Octavian Dumitru and Inge Gavat

This paper describes continuous speech recognition experiments for Romanian language, by using HMM modeling. The following questions are to be discussed: the realization of a new front-end reconsidering linear prediction, the enhancement of recognition rates by context dependent modeling, the evaluation of training strategies ensuring speaker independence of the recognition process without speaker adaptation procedures, by speaker selection for training. The experiments lead to a development of the initial system with a promising front-end based on PLP coefficients, second ranked for the recognition performance obtained, near the first ranked front-end based on mel-frequency cepstral coefficients (MFCC), but far better as the last ranked, based on simple linear prediction. Concerning the implemented algorithm for context dependent modeling, it permits in all situations enhanced recognition rates. The experiments made with gender speaker selection enhanced under certain conditions the recognition rate, proving good generalization properties especially by training with the male speakers database.

INTELLIGENT SIMULATOR DESIGN FOR DISTRIBUTED PIPELINE NETWORKS CONTROL

Yong Xu

Supervisory Control and Data Acquisition (SCADA) systems are widely used to meet the ever-increasing technological demands for monitoring and control of distributed system. An intelligent simulator is designed to enhance the conventional SCADA system. The new architecture can be exploited to develop integrated systems for complex distributed system management, performance prediction, fault detection and optimized operation.

ON JUST IN TIME CONTROL OF SWITCHING MAX-PLUS LINEAR SYSTEMS

Michel ALSABA, Sebastien Lahaye and Jean-Louis Boimond

Discrete event systems involving synchronization and delay phenomena can be described by a linear state representation over $(\max,+)$ algebra. Some discrete event systems involving choice phenomena could be transformed, under some conditions, into switching max-plus linear systems modeled as automata. The switching between states of these automata is governed by a switching variable. This paper deals with the just in time control of these switching max-plus linear systems. The control problem we propose is optimal under just in time criterion when the switching variable is given on the study horizon.

EVOLUTIONARY DATA MINING APPROACH TO CREATING DIGITAL LOGIC

James F. Smith III and ThanhVu H. Nguyen

A data mining based procedure for automated reverse engineering has been developed. The data mining algorithm for reverse engineering uses a genetic program (GP) as a data mining function. A genetic program is an algorithm based on the theory of evolution that automatically evolves populations of computer programs or mathematical expressions, eventually selecting one that is optimal in the sense it maximizes a measure of effectiveness, referred to as a fitness function. The system to be reverse engineered is typically a sensor. Design documents for the sensor are not available and conditions prevent the sensor from being taken apart. The sensor is used to create a database of input signals and output measurements. Rules about the likely design properties of the sensor are collected from experts. The rules are used to create a fitness function for the genetic program. Genetic program based data mining is then conducted. This procedure incorporates not only the experts’ rules into the fitness function, but also the information in the database. The information extracted through this process is the internal design specifications of the sensor. Significant experimental and theoretical results related to GP based data mining for reverse engineering and the related uncertainties will be provided.

THE USE OF MODULATING FUNCTIONS FOR IDENTIFICATION OF CONTINUOUS SYSTEMS WITH TIME-VARYING PARAMETERS

Witold Byrski and Jedrzej Byrski

In the paper the use of modulating functions for the optimal identification of the structure and parameters of continuous linear systems is presented. The modulating functions with compact support [0, h] are used in convolution filter for transformation of input/output signal derivatives. Based on pre-filtered functions continuous moving window [t-T, t] is used for on-line identification of piecewise constant parameters Q changes of linear system. Two versions of optimal quadratic method for identification are presented – with linear constraints on parameters Q and with quadratic constraints. The numerical results of some examples are shown.

FUZZY CLASSIFICATION BY MULTI-LAYER AVERAGING - An Application in Speech Recognition

Milad Alemzadeh, Saeed Bagheri Shouraki and Ramin Halavati

This paper intends to introduce a simple fast space-efficient linear method for a general pattern recognition problem. The presented algorithm can find the closest match for a given sample within a number of samples which has already been introduced to the system. The fact of using averaging and fuzzy numbers in this method encourages that it may be a noise resistant recognition process. As a test bed, a problem of recognition of spoken words has been set forth to this algorithm. Test data contain clean and noisy samples and results have been compared to a classic speech recognition method.

SONAR BUOYS: AN IMPROVED DESIGN APPROACH

K. Balasubramanian

Design approach for improved system performance of a microprocessor controlled sonar buoy performing surveillance of underwater objects is proposed. When launched into sea or ocean the microprocessor controlled buoy sets into action for automatic scanning of the underwater as to extract the object information and transmit the same by wireless to a remote ground station for further processing and taking final control action. System design outline for sonar buoy incorporating 11-cell replica correlation resulting in improved system performance is presented in this paper. Although the complexity of the hardware replica correlator is minimized using the recent digital delay lines the proposed microprocessor controlled buoy performs replica correlation through software and extracts object information conceding improved system performance.

CONTROLLING THE LORENZ SYSTEM WITH DELAY

Yechiel J. Crispin

A generalized method for adaptive control, synchronization of chaos and parameter identification in systems governed by ordinary differential equations and delay-differential equations is developed. The method is based on the Lagrangian approach to fluid dynamics. The synchronization error, defined as a norm of the difference between the state variables of two similar and coupled systems, is treated as a scalar fluid property advected by a fluid particle in the vector field of the controlled response system. As this error property is minimized, the two coupled systems synchronize and the time variable parameters of the driving system are identified. The method is applicable to the field of secure communications when the variable parameters of the driver system carry encrypted messages. The synchronization method is demonstrated on two Lorenz systems with variable parameters. We then apply the method to the synchronization of hyperchaos in two modified Lorenz systems with a time delay in one the state variables.

IDENTIFICATION OF SLOWLY TIME-VARYING SYSTEMS BASED ON THE QUALITATIVE FEATURES OF TRANSIENT RESPONSE A FROZEN-TIME APPROACH

Nelio Pastor, Juan J. Flores, Claudio R. Fuerte and Felix Calderón

A method for structural and parameter identification of a slowly time-varying systems is proposed. The frozen-time method is used in this analysis. By means of this method we obtain consecutive LTI models, which are identified in consecutive discrete instants using the Qualitative System Identification (QSI) Algorithm. The proposed algorithm models the behavior of the ODE’s coefficients means of polynomial functions. The algorithm models the variations of those coefficients though polynomials. An optimal model is obtained using Genetic Algorithms. The algorithm starts with a polynomial of second degree and tries to fit these polynomials, to the variations of the coefficients. If the degree of the polynomials is not enough it increases and repeats the process until achieving a good fit. The system was tested with simulated experiments in matlab, and then tested with the identification of a controlled experiment in a power systems laboratory. Keywords: Time-varying systems, LTI systems, Genetic Algorithms, Frozen-time approximation, Gradient optimization, System Identification.

A NEW METHOD FOR THE EVALUATION OF THE SIGNAL ACQUIRED FROM QUANTITATIVE SEISMOCARDIOGRAPH - Hardware and Software Solution for the new Field of Monitoring Heart Activity

Z.Trefny, J. Svacinka, S. Trojan, J. Slavicek, P. Smrcka and K. Hana

The device for quantitative seismocardiography (QSCG) detects cardiac vibrations as they affect the entire body; the measuring sensors (solid-state accelerometers) are usually placed in the plate of the chair – additional instruments applied on the proband’s body are not required. The results of the QSCG analysis are usable in various clinical fields. The first and most important step in the process of detection of significant characteristics of measured QSCG curves is to detect pseudo-periods in the signal regardless of the initial pseudo-period position. Other characteristics can be acquired by a relatively simple process over the appointed pseudo-period. We have developed the experimental equipment for the QSCG measuring and analysis. We have also developed special algorithms for preprocessing, segmentation and interactive analysis of the QSCG signal. In this contribution we will introduce technical principles of the quantitative seismocardiography and then we will focus on the original method for the basic segmentation of the QSCG signal in time-domain; the method is easy, robust and is appropriate for real-time QSCG processing.

SUFFICIENT CONDITION OF MAX-PLUS ELLIPSOIDAL INVARIANT SET AND COMPUTATION OF FEEDBACK CONTROL OF DISCRETE EVENT SYSTEMS

Mourad Ahmane and Laurent Truffet

Haar's Lemma (1918) provides the algebraic charcaterization of the inclusion of polyhedral sets. This Lemma has been involved many times in automatic control of linear dynamical systems when the constraint domains (state and/or control) are polyhedrons. More recently, this Lemma has been used to characterize stochastic comparison w.r.t linear orderings of Markov chains with different state space. Stochastic comparison is involved in the simplification of complex stochastic systems in order to control the approximation error made. In this paper we study the positive invariance of a max-plus ellispsoid by a max-plus linear dynamical system. We remark that positive invariance of max-plus ellipsoid is a particular case of polyhedron inclusion and we use Haar's Lemma to derive sufficient condition for the positive invariance. As an application we propose a method to compute a static state feedback control.

BAYESIAN ESTIMATION OF DISTRIBUTED PHENOMENA USING DISCRETIZED REPRESENTATIONS OF PARTIAL DIFFERENTIAL EQUATIONS

Felix Sawo, Kathrin Roberts and Uwe D. Hanebeck

In this paper we lay the foundation for a novel filtering technique for the fusion of two random vectors with imprecisely known stochastic dependency. This problem mainly occurs in decentralized estimation, e.g. of a distributed phenomenon, where the stochastic dependencies between the individual states are not stored. Thus, we derive parameterized joint densities with both Gaussian marginals and Gaussian mixture marginals. These parameterized joint densities contain all information about the stochastic dependencies between their marginal densities in terms of a parameter vector Xi, which can be regarded as a generalized correlation parameter. Unlike the classical correlation coefficient, this parameter is a sufficient measure for the stochastic dependency even characterized by more complex density functions such as Gaussian mixtures. Once this structure and the bounds of these parameters are known, bounding densities containing all possible density functions can be found.

RECONFIGURABLE HARDWARE IN-THE-LOOP SIMULATIONS FOR DIGITAL CONTROL DESIGN

Carlos Paiz, Christopher Pohl and Mario Porrmann

A framework to perform hardware-in-the-loop (HIL) simulations in the designflow of digital controllers, based on Field Programmable Gate Array (FPGA) technology, is presented. The framework allows the interaction of digital controllers, implemented on our rapid prototyping system, the RAPTOR2000, with a Matlab/Simulink simulation running on a host computer. The underlaying hardware and software designs supporting the interaction of the digital control and the simulation are presented. The designflow of FPGA-based digital controllers when using HIL is described and examples are given. Results from HIL simulations are presented, showing that the acceleration of the simulation increases with the complexity of the design when the number of I/Os stays constant.

ACCURATE OBSERVER FOR MULTI-FAULT DETECTION AND ISOLATION IN TIME VARYING SYSTEMS USING FAULT CHARACTERIZATION

Ryadh Hadj Mokhneche and Hichem Maaref

The usual observers up to now allowed the detection of faults in a parameter system via residue signals where each one is judicious to detect one or more faults. However in the event of occurrence of several faults on the same parameter, the residue signal of this observer will be able to detect them only if those are sufficiently spaced in time. But in the event of their occurrence at very close moments, they will be overlapped or compared to only one fault and having a more significant amplitude. Thus, if a possible fault compensation is carried out, it will be incorrect. In this paper, it is proposed then an accurate observer for fault detection and isolation for one or several faults on a same parameter and with a significant resolution. First, the characteristics of fault are shown to be used in a goal of determinating the types of possible detections. An application of simulation is detailed and achieved for fault detection in a sensor-based system, where the results are discussed. The succession effect of several faults is tested, at one time or different times, on the amplitude, sign and general form of these faults. In the end, the resolution of this observer is highlighted where a comparison between the usual observer and the accurate observer is discussed.

FAULT CHARACTERIZATION FOR MULTI-FAULT OBSERVER-BASED DETECTION IN TIME VARYING SYSTEMS

Ryadh Hadj Mokhneche and Hichem Maaref

Useful fault information, such as the amplitude and the sign, occurring during a time variable dynamic process are of capital importance to proceed correctly to fault compensation. The existing observers in literature, providing residues signals containing information on the presence or not of faults, do not take into account all of the faults when those occur at very close moments, which leads to an incorrect eventual compensation. This consideration is very significant for a correct dynamic control. In this paper, the characterization of the fault form in the time varying dynamic systems based on observers is proceeded to consider the detection of several faults some is their incidence moment and to take into account their amplitudes. The study of the several faults succession at the same moment or different moments, and of its consequences, is detailed. It is highlighted then the contribution of this characterization to fault detection and resolution where the interest to exploit these resolution in precise fault detection is shown.

NEURAL NETWORK BASED DATA FILTERING FOR POSITION TRACKING OF AN UNDERWATER VEHICLE

M. Ufuk Altunkaya, Serhat İkizoğlu and Fikret Gürgen

As a side effect of the developments in the mobile robotics, navigational technology has gained a leap recently. Although the most popular navigational aid for trajectory tracking is the Global Positioning System (GPS), it has also some disadvantages. Therefore attentions are drawn to other navigational devices such as Inertial Navigation Systems. Taking the underwater implementations of vehicle navigation into account, INS becomes a necessity due to communicational problems between the GPS and the satellites. At underwater vehicles Inertial Navigation Systems consisting of Inertial Measurement Units (IMU) such as accelerometers and gyros are used combined with other navigational devices like GPS or sonar. The error of the IMU output makes it necessary to be accompanied by an additional device. In this paper a neural network based filtering system introduced that is planned to be used for the trajectory tracking of an underwater vehicle.

ON THE STABILITY OF THE DISCRETE TIME JUMP LINEAR SYSTEM

Adam Czornik and Aleksander Nawrat

In this paper we investigate the relationships between individual mode stability and mean square stability of jump linear system. It is well known that generally stability of a dynamical system in all its modes does not guarantee stability of the jump linear system defined by all these modes. We present conditions under which stability of all modes implies the mean square stability of the overall system.

BOUNDS FOR THE SOLUTION OF DISCRETE COUPLED LYAPUNOV EQUATION

Adam Czornik and Aleksander Nawrat

Upper and lower matrix bounds for the solution of the discrete time coupled algebraic Lyapunov equation for linear discrete-time system with Markovian jumps in parameters are developed. The bounds of the maximal, minmal eigenvalues, the summation of eigenvalues, trace and determinant are also given.

A SAFETY SYSTEM FOR DYNAMIC VACUUM LIQUID NITROGEN PIPELINE - For World Class Manufacturing Operation in Semiconductor Industry

Vinayak Divate and Pichit Saengpongpaew

Liquid nitrogen is a colorless, odorless, extremely cold liquid and gas under pressure. It can cause rapid suffocation when concentrations are sufficient to reduce oxygen levels below 19.5%. Contact with liquid or cold vapors can cause severe frostbite. One volume of liquid nitrogen will expand to produce 696.5 equivalent volumes of gas. With this background Ln2 is being used in Semiconductor industry especially in testing operation. The cold tests are taken on products at a temp up to -40C to -60C. The Ln2 is used in testing machine to reduce test handler chamber temp. The chamber temp is required to maintain at -40 C with allowance of + or – 3C to get accurate results. So this paper describes how Dynaflex ln2 pipeline system can be maintained smartly to get maximum benefits with minimum unscheduled shutdown in semiconductor industry. This paper also gives the details about the safety system developed at SPANSION Thailand limited for handling of ln2 through Dynamic vacuum ln2 pipeline system.

LINEAR QUADRATIC GAUSSIAN REGULATORS FOR MULTI-RATE SAMPLED-DATA STOCHASTIC SYSTEMS

L. Armesto and J. Tornero

In this paper, linear quadratic Gaussian regulators are presented and formalized for multi-rate sampled-data stochastic systems using two well-known approaches: lifting technique and time-variant periodic modeling. It has been shown that both regulators are equivalent at the global frame-period with different computational costs and execution periods. An interesting analysis has been done to demonstrate the convergence of a periodic Kalman filter, used in the periodic regulator, into its equivalent continuous one (Bucy Kalman filter), when the periodicity ratio converges to infinity. In addition to this, in both regulators, multi-rate holds have been used, acting as interfaces between signals at different sampling rates, which may improve the system performance. A numerical example of LQG multi-rate control of a MIMO plant shows the application of both regulators, where in addition to showing the improvement with respect to the single-rate case.

A 16-BIT SWITCHED-CAPACITOR SIGMA-DELTA MODULATOR MATLAB MODEL EXPLOITING TWO-STEP QUANTIZATION PROCESS

Lukas Fujcik, Radimir Vrba and Miroslav Sveda

This paper presents a novel architecture of high-order single-stage sigma-delta (ΣΔ) converter for sensor measurement. The two-step quantization technique was utilized to design a novel architecture of ΣΔ modulator. The time steps are interleaved to achieve resolution improvement without decreasing of conversion speed. This technique can be useful for low oversampling ratio. The novel architecture was designed to obtain high dynamic range of input signal, high signal-to-noise ratio and high reliability. The proposed architecture of switched-capacitor (SC) ΣΔ modulator was simulated with blocks containing nonidealities, such as sampling jitter, noise, and operational amplifier parameters (white noise, finite dc gain, finite bandwidth, slew rate and saturation voltages). The novel architecture of SC ΣΔ modulator with two-step quantization process was designed and simulated in MATLAB SIMULINK.

EMBEDDED FPGA SOLUTION FOR WATER QUALITY MONITORING SYSTEM - Calibration and Measurement

Octavian Postolache, Jose Miguel Dias Pereira and Pedro Silva Girão

The paper presents a field operating water quality monitoring system based on real time controller and FPGA module. The system features functioning includes in-situ automatic cleaning and calibration of stand alone sensors such as turbidity, pH or conductivity, on-line measurement of water quality parameters using the calibrated sensors. In order to perform the above mentioned calibration and measurement tasks the system uses a set of centrifugal pumps and electrovalves and a corresponding embedded control materialized by the LabVIEW programmed FPGA module. The voltages associated with water quality measurement channels are acquired using a four channels analog input that work also under FPGA control. The data processing tasks are distributed between the FPGA module and the real-time controller included in the system. A practical approach concerning the sensor model implementation capabilities using the real-time controller (NI cRIO-9002) or FPGA (NI cRIO-9003) is also included. In order to provide the wireless remote control of the system an Ethernet – wireless bridge (IEEE802.11g) and client-server TCP software developed in LabVIEW was included in the system. A PDA based remote control solution was implemented to evaluate system performance.

SMART WIRELESS TIPPING-BUCKET RAIN GAUGE - Measurement and Automatic Dynamic Calibration

Octavian Postolache, Jose Miguel Dias Pereira and Pedro Silva Girão

The paper presents the design and implementation of a smart tipping-bucket rain gauge that uses a universal frequency do digital converter characterised by period and impulse counting measuring capabilities with online accuracy control and a serial interface connected to transmitter-receiver RF module that provides a wireless communication between the smart tipping-bucket rain gauge (TBR) and a host unit expressed by a FieldPoint real-time controller or a laptop PC associated with a weather monitoring network The TBR sensor tests in dynamic conditions is performed using FieldPoint based system. The system consist in a submersible pump that works under the FieldPoint control and assure the accurate control of water flow rates delivered to the rain gauge funnel. The rain gauge calibration ensures precise conversion of bucket tip times to actual rainfall rates. The data acquired during the calibration is stored in FieldPoint system memory and used for an accurate rain fall measurement after an intelligent data processing based on designed and implemented neural network. Data logging and data communication are parts of the LabVIEW real time software developed for the present system.

AN ALGORITHM EVALUATION TEST SUITE FOR BLIND SOURCE SEPARATION PROBLEM

Marina Charwath, Imke Hahn, Sascha Hauke, Martin Pyka, Slawi Stesny, Dietmar Lammers, Steffen Wachenfeld and Markus Borschbach

A first step to perform a competition of methods for source separation is the development of a testsuite that supports the development and evaluation of blind source separation (BSS) algorithms in a highly automated way. The concept of our testsuite is presented and it is shown how the testsuite can be used to apply BSS-algorithms to four standard sub-problems. To compare the performance of arbitrary algorithms on given problems the testsuite allows the integration of new algorithms and testing problems using well defined interfaces. A brief example is given by the integration of the FlexICA, EVD, EVD24 and the FastICA algorithm and our results achieved from automated tests and parameter optimizations.

A System for Analysis of the 3D Mandibular Movement using Magnetic Sensors and Neuronal Networks

Isa C. T. Santos, João Manuel R. S. Tavares, Joaquim G. Mendes and Manuel P. F. Paulo

In Dental Medicine, the study of the mandibular movement has an important role in the development of oral rehabilitation treatments, because it allows to determine if exists or not pathologies in the temporomandibular joints and helps the definition of adequate treatment plans. In this paper, is presented the development of a new system for the acquisition of the 3D mandibular movement. A common facial arc used in Dental Medicine was adapted as main support structure, and electromagnetic sensors were used to acquire the movement. To visualise and analyse in a personal computer the movement acquired, it was developed a computational application using the developing tool LabVIEW. In this work neural networks were employed to transform in cartesian coordinates the electrical signals obtained from the electromagnetic sensors.

Vehicle Variable Estimation in Diagnostic Context

Elie Accari, Denis Hamad and Chaiban Nasr

Safety in vehicles has many aspects and is implemented in different ways by manufacturers. With more safety systems to come, the vehicle will certainly start to have an operating system to manage the whole. Neural networks have an adaptative behavior that can be trained to meet new conditions and have a certain inherent degree of robustness when used as variable estimators. In this paper, we present a simplified model of the vehicle suitable to create neural network architectures that estimate the forces applied to the wheel as well as the vehicle body slip angle and yaw rate. For this purpose, we use the veDyna simulator which substitutes safely and economically real test vehicles. Typical extraneous and erroneous data are then presented to test the robustness of the network in order to judge the applicability of this approach from ideal exact calculation conditions to real life situations.

Smart Sensor Networking with ZigBee and Internet

Miroslav Sveda and Roman Trchalik

This paper deals with sensor networking based on ZigBee and Internet using IEEE 1451 smart transducer interface architecture. The contribution begins with introduction to the IEEE 1451 smart transducer - network interface for sensors and actuators as an emerging, standard-based networking framework. Next part of the paper reviews some concepts of ZigBee architecture aimed at connecting wireless sensors and actuators through ZigBee to Intranets or Internet. The kernel of the paper deals with design of a software architecture stemming from technical standards or standard proposals. This paper focuses namely on design of software architectures of communication interconnecting devices in between ZigBee and Internet.

Toward Automatic Defects Clustering in Industrial Production Process Combining Optical Detection and Unsupervised Artificial Neural Network Techniques

Matthieu Voiry, Kurosh Madani, Véronique Amarger and François Houbre

A major step for high-quality optical surfaces faults diagnosis concerns scratches and digs defects detection and characterisation in products. This challenging operation is very important since it is directly linked with the produced optical component’s quality. A new scratches and digs defects detection and characterisation method exploiting Nomarski’s microscopy issued imaging has been developed. The items detected using this high-performance approach can correspond to real defects on the structure but some dusts and cleaning marks are detected too. Thus, a classification phase is necessary to complete optical devices diagnosis. In this paper, we describe a data extraction method, which supplies pertinent features from raw Nomarski’s images issued from in-dustrial process. Then we apply this method to construct a database from real images. Finally we analyse the pertinence of features and the complexity of obtained database by clustering operation using an unsupervised Self Organizing Maps technique.

Modular Statistical Optimization and VQ Method for Image Recognition

Amar Djouak, Khalifa Djemal and Hichem Maaref

In this work, a modular statistical optimization method enriched by the introduction of VQ method dedicated to obtain the effectiveness and the optimal comuting time in images recognition system is poposed. To this aim, a comparative study of two RBF and an SVM classifiers are carried out. For that, features extraction is made based on used image database. These features are gathered into blocks. The statistical validation results allow thus via the suggested optimization loop to test the precision level of each block and to stop when this precision level is optimal. In the majority of the cases, this iterative step allows the computing time reduction of the recognition system. Finally, the introduction of vector quantization method allows more global accuracy to our architecture.

An Active Learning Approach for Training the Probabilistic RBF Classification Network

Constantinos Constantinopoulos and Aristidis Likas

Active learning for classification constitutes a type of learning problem where a classifier is gradually built by iteratively asking for the labels of data points. The method involves a data selection mechanism that queries for the labels of those data points that considers to be mostly beneficial for improving the performance of the current classifier. We present an active learning methodology for training the probabilistic RBF (PRBF) network which is a special case of the RBF network, and constitutes a generalization of the Gaussian mixture model. The method employs a suitable criterion to select an unlabeled observation and query its label. The proposed criterion selects points that lie near the decision boundary. The learning performance of the algorithm is tested with experiments on several data sets.

Distributed Product Development Of A Fuel-Injection System Using Multi-Agent

Zuhua Jiang, Shiwei Fu and Burkhard Lege

Multi-agent modeling has emerged as a promising discipline for dealing with decision making processes in distributed information system applications. One of these applications is the modeling of distributed design and analysis processes which can link up various designs and simulation processes to form a virtual consortium on a global basis. This paper proposes a multi-agent cooperative framework for the development of a fuel-injection-system including a fuel-injection-system and consisting of more than 90 parts. The meta-model of management agent and actor agent for the development of the fuel-injection-system is presented, and the architecture of the distributed multi-agent system for the development of a fuel-injection-system is discussed. The prototype system and some key agents in the distributed product development are introduced.

Blind Source Separation Based on a Single Observation

Damjan Zazula and Aleš Holobar

This paper deals with a novel approach to the compound signal decomposition. It takes advantage of blind source separation using the algorithm for convolution kernel compensation (CKC). We derive a version which cope with compound signals, mixtures of several source contributions, even if only a single observation is available. Our novel approach detects and separates the triggering instants of all source symbols which contribute to the processed observation. The obtained decomposition is very robust and accurate. We experimented with synthetic signals having characteristics similar to the electrocardiographic (ECG) signals. Also at signal-to-noise ratios (SNRs) as low as 0 dB, the obtained average true positive statistics for the detected source-symbol triggerings was 98+-1%, average false positive statistics 2+-1%, and false negative statistics 3+-2%.

Content-Adaptive Data Fusion

Paul Bao and Le Thanh Hai

We propose a novel image fusion scheme based on independent component analysis in which image is fused aimed at information maximization. In the scheme, a novel algorithm is presented which, based on specific fusing images, determines adaptively a specific weight for linear fusion of images using ICA. The scheme is established on the ICA maximum information principles and offers an efficient and adaptive image fusion process with the robustness under various fusion situations.

Hypothesis Testing as a Performance Evaluation Method for Multimodal Speaker Detection

Patricia Besson and Murat Kunt

This work addresses the problem of detecting the speaker on audio-visual sequences by evaluating the synchrony between the audio and video signals. Prior to the classification, an information theoretic framework is applied to extract optimized audio features using video information. The classification step is then defined through a hypothesis testing framework so as to get confidence levels associated to the classifier outputs. Such an approach allows to evaluate the whole classification process efficiency, and in particular, to evaluate the advantage of performing or not the feature extraction. As a result, it is shown that introducing a feature extraction step prior to the classification increases the ability of the classifier to produce good relative instance scores.

Facial SEMG for Speech Recognition Inter-Subject Variation

Sridhar P. Arjunan, Dinesh K. Kumar, Wai C. Yau and Hans Weghorn

The aim of this project is to identify speech using the facial muscle activity and without audio signals. The paper presents an effective technique that measures the relative muscle activity of the articulatory muscles. The paper has also tested the performance of this system for inter subject variation. Three English vowels were used as recognition variables. This paper reports using moving root mean square (RMS) of surface electromyogram (SEMG) of four facial muscles to segment the signal and identify the start and end of the utterance. The RMS of the signal between the start and end markers was integrated and normalised. This represented the relative muscle activity, and the relative muscle activities of the four muscles were classified using back propagation neural network to identify the speech. The results show that this technique gives high recognition rate when used for each of the subjects. The results also indicate that the system accuracy drops when the network trained with one subject is tested with another subject. This suggests that there is a large inter-subject variation in the speaking style for similar sounds. The experiments also show that the system is easy to train for a new user. It is suggested that such a system is suitable for simple commands for human computer interface when it is trained for the user.

Changes in EMG During Short Duration Supra Maximal and Long Duration Sub-maximal Exercise: A Comparative Study

Vijay Pal Singh, Dinesh Kant Kumar, Barbara Polus, Steve Fraser and Sonia Lo Guidice

Surface electromyogram (sEMG) is a non-invasive recording of the underlying muscle activities. It is used as a measure of the force of contraction, and changes in sEMG are used as an indicator for localised muscle fatigue. This paper reports a study undertaken to determine the difference in the change of sEMG due to fatigue resulting from short time sprint cycling, and long duration cycling. This paper reports the results of the experimental study of the two kinds of exercises i.e. short duration (supra-maximal) and the long duration (sub-maximal). The results indicate that measure of the spectrum shift was effective in the detection of fatigue in supra-maximal dynamic contraction but was not useful for fatigue caused due to long duration sub-maximal cycling.

ICA for Surface Electromyogram

Ganesh R. Naik, Dinesh K. Kumar, Sridhar P. Arjunan and M. Palaniswami

Surface electromyogram (SEMG) is an indicator of the underlying muscle activity and can be useful for human control interface. One difficulty in the use of SEMG for identifying complex movements is the mixing of muscle activity from other muscles, referred to cross-talk. Similarity in frequency and time domain makes the separation of muscle activity from different muscles extremely difficult. Independent Component Analysis (ICA) is a useful technique for blind source separation. This paper reports investigations to test the effectiveness of using ICA for such applications. It determines the impact of different conditions on the reliability of the separation. The paper reports the evaluation of issues related to the properties of the signals and number of sources. The paper also tests Zibulevsky’s method of temporal plotting to identify number of independent sources in SEMG recordings. The results demonstrate that ICA is suitable for SEMG signals when the numbers of sources are not greater than the number of recordings. The inability of the system to identify the correct order and magnitude of the signals is also discussed. It is observed that even when muscle contraction is minimal, and signal is filtered using wavelets and band pass filters, Zibulevsky’s sparse decomposition technique does not identify number of independent sources.

Comparison of Approximate Entropy Measure and Poincaré Plot Indexes for the Study of Gait Characteristics in the Elderly

Ahsan H. Khandoker, Marimuthu Palaniswami, and Rezaul K. Begg

Early identification of at-risk gait helps prevent falls and injuries. The aim of this study is to investigate the relationship between approximate entropy (ApEn) and Poincaré plot indexes of elderly gait patterns and to test whether ApEn could be used as a reliable gait identifier for falls-risk. Minimum foot clearance (MFC) data of 14 elderly and 10 elderly participants with a history of falls and balance problems were analyzed. The ApEn values of MFC were significantly correlated with Poincaré plot indexes of MFC in the healthy elderly group, whereas correlations were absent in the elderly fallers group. Mean ApEn in the fallers group (0.18±0.03) was significantly higher than that in the healthy group (0.13±0.13). The higher ApEn values in the fallers group might indicate increased irregularities in their gait patterns and a loss of gait control mechanism. Results are useful for the early diagnosis of common gait pathologies.

Automated Recognition of Human Movement States using Body Acceleration Signals

Md. Rafiul Hassan, Rezaul K. Begg, Ahsan H. Khandoker and Robert Stokes

Automated recognition of human activity states has many advantages, e.g., applications in the smart home environment for the monitoring of physical activity levels, detection of accidental falls in the older adults in the home environment or assessment of the recovery phase of patients living independently at home. In this paper, we describe an accelerometer-based system to recognize three activity states, e.g., steady state gait or walking, sitting and simulated sudden accidental falls. The recorded 3D movement accelerations of the trunk were processed using wavelets, and the features were extracted for recognition of movement states through the use of a fuzzy inference system. The system was trained and tested using 58 different data segments representing the three states. Cross-validation test results indicated an overall recognition accuracy by the machine classifier to be 89.7% with an ROC area of 0.83. The results suggest good potential for the system to be applied for various situations involving activity monitoring as well as gait and posture recognition. Further tests are required using various population groups.

The Effect of Shape Variables of Tibial Plateau on Tibio-femoral Movement Based on a Three-dimensional Anatomical Dynamic Model

N. Ekin Akalan, Mehmed Özkan and Yener Temelli

In this study the geometric and material properties of joint surfaces, bones, ligaments of the tibio-femoral joint is represented and passive knee flexion is simulated. The purpose of the study is to observe the effect of 11° tibial slope to the tibio-femoral movement. The contact forces between tibia and femur are defined as frictionless mathematical model. Tibial plateaus and condyles of femur are represented as ellipsoids as described in literature. Anterior, posterior cruciate ligaments, medial, lateral collateral ligaments are represented as non linear elastic springs. Knee flexion with and without internal-external torque are simulated, and the results are compared with the literature for slopped and flattened medial tibial plateau models. As a result, normal internal rotation of tibia and adduction ranges are achieved for unloaded condition in flattened model, but the knee flexion with forced internal/external rotation are out of normal range for both models.

SEMG for Identifying Hand Gestures using ICA

Ganesh R. Naik, Dinesh K. Kumar, Vijay Pal Singh and M. Palaniswami

There is an urgent need for establishing a simple yet robust system that can be used to identify hand actions and gestures for machine and computer control. Researchers have reported the use of multi-channel electromyogram (EMG) to determine the hand actions and gestures. The limitation of the earlier works is that the systems are suitable for gross actions, and when there is one prime-mover muscle involved. This paper reports overcoming the difficulty by using independent component analysis to separate muscle activity from different muscles and classified using backpropogation neural networks. The system is tested and found to be effective in classifying EMG.

ITIHand: A Real System for Palmprint Identification

José García-Hernández, Roberto Paredes, Ismael Salvador and Javier Cano

In the networked society there are a great number of systems that need biometric identification, so it has become an important issue in our days. Biometrics takes advantage of a number of unique, reliable and stable personal physiological features, to offer an effective approach to identify subjects. This identification can be based on palmprint features. At present work is described a real biometric identification system based on palmprints that uses local features.

A Novel Neural Eye Gaze Tracker

Diego Torricelli, Michela Goffredo, Silvia Conforto, Maurizio Schmid and Tommaso D’Alessio

A gaze tracking system, based on a novel neural approach, is proposed. The work is part of a wider research project concerning the development of human computer interfaces (HCI) addressed to disabled people, that could overcome the drawbacks of most of the existing methods for gaze tracking that require either intrusive devices or expensive equipment. This work, instead, aims at developing a low cost, completely non-intrusive and self-calibrating system. The method combines different techniques for three blocks in Eye Gaze Tracking, i.e. blink detection, feature extraction and neural computing, with a different approach with respect to the ones found in literature. First experimental results show good accuracy and robustness.

Continuous Blood Pressure Measurements in Stress Situations

Cord Volker Bauch and Dieter Barschdorff

When a person is exposed to physical and psychological challenge, the autonomic nervous system reacts in a way to cope the situation. Stress conditions are usually characterized by heart frequency or skin resistance changes. Though also the blood pressure is known to increase in stress situations, its measurement was not meaningful because of the insufficient time resolution of instruments using inflatable cuffs. We present the first model based continuous blood pressure determinations during stress tests. The measuring technique is based on the dependency of the systolic blood pressure on the pulse transit time and on individualized mathematical models. The Vienna Test System and car driving situations in the driving simulator "Nightdriver" are examined. With the new technique the blood pressure can be determined without interfering with the persons cognitive perceptions. It clearly correlates to different stress levels.

Human-Machine Interfaces Based on EMG and EEG Applied to Robotic Systems

Andre Ferreira, Wanderley Cardoso Celeste, Fernando Auat Cheein, Teodiano Freire Bastos-Filho and Mario Sarcinelli-Filho

Two different electro-biological signal based Human-Machine Inter- faces(HMIs)were developed: EMG and EEG based.Such interfaces present like main characteristics relatively simple acquisition and processing systems, which need of few hardware and software resources, so that they are computational and financial low cost solutions.Both interfaces have been applied to robotic systems and their performance have been shown up in such applications.The EMG based HMI was tested in a mobile robot,while the EEG based HMI was tested as much in a mobile robot as a robotic manipulator.

Comparison of Gene Selection and Machine Learning for Tumor Classification

Qingzhong Liu, Andrew H. Sung and Bernardete M. Ribeiro

Class prediction and feature selection are two learning tasks that are strictly paired in the search of molecular profiles from microarray data. In this paper, we compare three gene selection methods and several machine learning classifiers for tumor tissues based on gene expression data under different dimensions. Our experimental results confirm that feature selection and learning classifiers are both critical to the classification accuracy. We find that, overall, the gene selection method, SVM-RFE (Support Vector Machine Recursive Feature Elimination) is superior to the other two gene selection methods. Additionally, KFD (Kernel Fisher Discriminant) classifier, SVM with polynomial kernels, and KNN (K-Nearest Neighbor) have better classification accuracy than others. Under the high feature dimensions, the classification performances of RBF kernel classifiers decrease along with the increase of the feature dimension. These results provide useful guidelines for experiments on gene selection from microarry data.

Combining Neural Tracking and Control to Improve Rehabilitation of Upper Limb Movements in Hemiplegia

M. Goffredo, I. Bernabucci, M. Schmid, S. Conforto and T. D’Alessio

This paper aims at introducing a novel approach for assisting and re-storing upper arm movements in stroke patients. The presented system inte-grates advanced markerless motion analysis together with an artificial neural network controller for a biomechanical arm model. The keypoint of the project is to acquire kinematics information from the healthy arm of a stroke patient during planar arm movements and elaborate them in order to obtain a self-rehabilitative stimulation of the plegic arm of the same patient. The first ex-perimental tests show good results and allow to define working direction for the extension of the work and for its application in clinical contexts.

Shape Factor's Effect on a Dynamic Cleaners Swarm

Yaniv Altshuler, Israel A. Wagner and Alfred M. Bruckstein

This work discusses a possibility results for the Dynamic Cooperative Cleaners problem, and the relation of a specific geometric feature of the problem, know as the shape factor, to the efficiency of the swarm. Thedynamic cooperative cleaners problem assumes a grid, having "contamination" points or tiles that form a connected region of the grid. Several agents move in this contaminated region, each having the ability to "clean" the place it is located in. The contaminated tiles expand deterministically, simulating a spreading of contamination, or fire. The equivalence of this problem to another interesting multi agents problems was demonstrated another paper by the authors by utilizing results relevant to the problem in order to design a cooperative hunting protocol for a swarm of UAVs. This work enhances analytic results which were previously published by the authors, while discussing the effect of the region's shape factor (i.e. the ratio between the region's boundary and its area) and the swarm's cleaning efficiency. As a result, a tighter lower bound is produced, establishing a new and more generic impossibility result for the problem.

Cooperative Multi-Agent Approach to Dynamic Coverage in Multi-Robot Activities

Satoshi Kataoka and Shinichi Honiden

Dynamic coverage is a problem of multi-robot systems based on wireless ad-hoc networks. The issue of dynamic coverage occurs notably in post-disaster survivor rescue, search operation, and planet exploration. In this paper, we address the problem of frequency in dynamic coverage, which is to cover all the areas of a free space in the shortest possible time. We introduce an novel algorithm of dynamic coverage in a realistically restricted environment for robots. This approach improves efficiency of moving around and amount of communication in a simulation environment. The paper presents comprehensive experimental results and discusses future research directions related to dynamic coverage.

Effect of Communication Error on “Iterated Proposal–Voting Process”

Hiroshi Kawakami, Toshiaki Tanizawa, Osamu Katai and Takayuki Shiose

This paper proposes the framework of a simulation called ``iterated proposal--voting process'' and reports the effect of communication error on the process. This framework simulates the decision-making of a relatively small community. Constituents of the community try to decide on a common rule shared by the community via iteration of ``propose and vote.'' This kind of decision-making is frequently observed, especially in small communities such as families and groups of close friends. In this case, the results of decision-making do not always integrate the intentions of the members but in fact violate the intentions of certain members. Furthermore the decision process becomes complicated, or the so-called ``groupthink'' emerges. To investigate the above mentioned process, we developed a multi-agent model. In the model, each agent decides its action based on two criteria: satisfying ``physiologically fixed needs (PN)'' and satisfying ``social contextual needs (SNs),'' which sometimes conflict with each other. These criteria are based on a Nursing Theory that puts special emphasis on the relation between subjects and others. SNs are satisfied when such relations are balanced. Employing ``Hyder's theory of cognitive balance,'' SNs are evaluated for whether they are balanced. The simulation yields some interesting phenomena that are not observed by conventional static analyses, e.g. power indices.

Coordinated Transportation of a Large Object by a Team of Three Robots

Rui Soares and Estela Bicho

Dynamical systems theory is used in this work as a theoretical language and tool to design a distributed control architecture for a team of three robots that must transport a large object and simultaneously avoid collisions with either static or dynamic obstacles. This work extends the previous work with two robots. Here we increase the complexity by increasing the number of robots. The robots have no prior knowledge of the environment. The dynamics of behavior is defined over a state space of behavior variables, heading direction and path velocity. Task constrains are modeled as attractors (i.e. asymptotic stable states) of a behavioral dynamics. For each robot, these attractors are combined into a vector field that governs the behavior. By design the parameters are tuned so that the behavioral variables are always very close to the corresponding attractors. Thus the behavior of each robot is controlled by a time series of asymptotical stable states. Computer simulations support the validity of the dynamical model architecture.

Strategic Searching Approaches in a Multi-Robot System

Yan Meng and Ke Cao

This paper proposes two strategic searching approaches for a multi-robot system: utility greedy approach and game theoretic approach. It is assumed that a-priori probability of the target distribution is provided in a partially known dynamic environment. A utility greedy approach and a strategic game-theory based approach are proposed to optimize the searching task using a dynamic-programming based utility function. The pure Nash Equilibrium so-lution and the mixed-strategy equilibrium solutions for game-theory based ap-proach are provided. Extensive simulation results demonstrate that the pro-posed searching approaches have better searching performance and robustness compared to the other heuristic searching strategies.

Airship Formation Control

Estela Bicho, André Moreira, Sérgio Diegues, Manuel Carvalheira and Sérgio Monteiro

This paper addresses the problem underlying the control and coordination of multiple autonomous airships that must travel maintaining a desired geometric formation and simultaneously avoid collisions with moving or stationary obstacles. The control architecture is based on the attractor dynamics approach to behaviour generation. The airship physical model is presented and the mathematical background for the control architecture is explained. Simulations (with perturbations) with formations of two and three autonomous airships are presented in order to validate the architecture.

SWITCH! Dynamic Roles Exchange Among Cooperative Robots

Carlos E. Agüero, Vicente Matellán, José María Cañas and Víctor M. Gómez

Cooperation is improved every year in the RoboCup Competition. New techniques are proposed for sharing information among players, multi-robot localization, etc. This paper describes the dynamic roles exchange system protocol added to an existing RoboCup team. Dynamic role exchange contributes with fast ball founds, better field coverage and avoids player pushings. Our proposed is a description and implementation of this technique using Aibo robots.

Adaptive Control Network for Multi-Robot Exploration

Jose Vazquez and Malcolm Chris

This work addresses the problem of exploring an environment with a team of communicating robots. Exploration can be performed more efficiently when robots are able to communicate and coordinate their actions. We propose an adaptive control approach to keep the robots as a single connected network. In this approach a control network is created at the beginning of the exploration based on the communication network. As the robots traverse the environment the control network is updated to enhance connectivity. The approach has been implemented for LOS (Line of Sight) and RF (Radio Frequency) technologies. Our approach has been compared with coordination approaches that rely on fixed networks. The results demonstrate that our approach performs better than these fixed network approaches.

A Hybrid, Teleo-Reactive Architecture for Robot Control

Simon Coffey and Keith Clark

In this paper we describe the structure of a proposed hybrid architecture for robot control. A BDI-style planning layer manipulates a plan library in which plans are comprised of hierarchical, suspendable and recoverable teleo-reactive programs. We also present preliminary simulation and implementation work.

Dynamic and Distributed Allocation of Resource Constrained Project Tasks to Robots

Sanem Sariel and Tucker Balch

In this work, we propose a dynamic task selection scheme for allocating real-world tasks to the members of a multi-robot team. Tasks in our research are subject to precedence constraints and simultaneous execution requirements. This problem is known as the Resource Constrained Project Scheduling Problem (RCPSP) in operations research. Particularly, we also deal with the missions that may change their forms by introducing new online tasks during execution making the problem more challenging besides the real world dynamism. Unpredictability of the exact processing times of tasks, unstable cost values during runtime and inconsistencies due to uncertain information form the main difficulties of the task allocation problem for robot systems. Since processing times of tasks are not exactly known in advance, we propose a dynamic task selection scheme for the eligible tasks instead of scheduling all of them to prevent redundant calculations. In our approach, globally efficient solutions are attained by the mechanisms for forming priority based rough schedules and selecting most suitable tasks from these schedules. Rough schedules are formed by tentative coalition commitments which are agreed upon by robots for the tasks with simultaneous execution requirements. Since our method is for real world task execution, communication requirements are kept at minimum as much as possible. The approach is distributed and computationally efficient.

Levy Flights in the Stochastic Dynamics of Robot Swarm Gathering

Yechiel J. Crispin

We consider the problem of gathering a swarm of robots which is initially randomly dispersed over a domain in the plane. A stochastic method for the cooperative control of a swarm of mobile robots is presented. The network of mobile robots is modeled by a swarm performing a directed random walk. The swarm dynamics are governed by a system of stochastic difference equations. The motion is controlled by a robot leader, which transmits the coordinates of the gathering point to the swarm as the network cooperative control signal. We study the case where the control signal is corrupted by noise and find that the gathering process is robust to noise and efficient. The swarm dynamics display anomalous diffusion and Levy flights, where the robots move along straight lines over many time steps, followed by short random walks in the vicinity of the gathering point.