AMC 2020 Schedule

ND-000353 Extending Dynamic Movement Primitives towards High-Performance Robot Motion

ND-000035 Development of Pushing Control Mechanisms for Generator Inspection Robot

ND-000345 Distributed Interpolation: Synchronization of motion-controlled axes with coordination vector and decentralized segment controllers

ND-000957 Collaborative Transport by Mecanum Mobile Robots using Reaction Torque Observer

ND-000515 Event-Triggered Sliding Mode Control Strategies for Positioning Systems: An Experimental Assessment

ND-000906 Fractional-Order System Identification of Viscoelastic Behavior: A Frequency Domain Based Experimental Study

ND-000728 On Orbital Stabilization as an alternative to Reference Tracking Control

ND-001309 A Fractional-Order Control Approach to Ramp Tracking with Memory-Efficient Implementation

ND-000892 Regulation of Penetration Rate and Drilling Power in Rotary Drilling Systems

When I started my research works with a name of Intelligent Space, the main idea was "We should use our technologies to support our human activities with keeping good health conditions". It means that the intelligent space should understand human behaviors and to provide proper physical support by using robotics. The Intelligent Space is focusing to fuse IT and Robotics in our daily life. I believe that such research direction is still important even if actual state of ongoing intelligent space is remaining in lower stages. In this talk I will show some current results of monitoring human health conditions, assisting human mobilities and elementary technologies for actuators from a view point of Intelligent Space to discuss our important research issues.

Do you also have a motion system that has the same error in each task? Iterative Learning Control (ILC) can achieve perfect performance for your system. A general learning framework is presented that exploits measured error signals from previous tasks. By employing very simple models, both fast and safe learning is achieved, guaranteeing a reduction of the error in each experiment. Typically, perfect performance is achieved in only five to ten iterations. A complete design framework for motion systems is provided, while at the same time touching upon the essential theoretical foundations, including non-causality of the optimal design and the connection to traditional feedback and feedforward designs. Finally, recent approaches are explored that facilitate the implementation on industrial systems, including flexibility for a large class of tasks and multivariable systems.

Learning for Advanced Motion Control

ND-001023 Development of Compact Linear Actuator Combining DC motor and Cylindrical Cam for Tactile Display

ND-000698 A High-Torque Density Compliant Actuator Design for Physical Robot Environment Interaction

ND-000477 Human-Adaptive Impedance Control Using Recurrent Neural Network for Stability Recovery in Human-Robot Cooperation

ND-001163 Novel Algorithm for Position/Force Control of Multi-DOF Robotic Systems

ND-001279 Transparent Torque Sensor-less Impedance Rendering for Low-cost Direct Drive Motor

ND-001171 Development and Basic Analysis of Novel Flexible Linear Motor

ND-000205 Soft Boom Cylinder Control Using Disturbance-Observer-Based Equivalent Hydraulic System for Electric Excavator

ND-000604 An Approach to Force Control by Model Predictive Velocity Control with Constraints

ND-000752 A Method to Make a Robot Understand What was a Target Object in Motion Copying System

ND-001295 Ripple Minimization for Harmonic-geared Series Elastic Actuator under Force Control

ND-001147 Feedback Controller Design Based on H-infinity Control Theory in Dynamically Substructured System

ND-000175 Experimental comparison of velocity estimators for a control moment gyroscope inverted pendulum

ND-000531 Controller design of mass flow rate loop for high-precision pneumatic actuator

ND-000442 A simple quasi-LPV approach to control design of an industrial wind turbine

ND-001015 Multi-System Iterative Learning Control: an Extension of ILC for Interconnected Systems.

ND-001155 Proposal of Automatic Power Plug Insertion Control for Electric Vehicle with In-Wheel-Motors

ND-001112 3K Compound Planetary Reduction Gearbox With Non-backlash Mechanism

ND-001198 Basic Study on Regenerative Air Brake Using Observer-based Thrust Control for Electric Airplane

ND-000213 Path Planning for Perception-Driven Obstacle-Aided Snake Robot Locomotion

ND-000736 Basic Idea of Quadrant Dynamic Programming for Adaptive Cruise Control to Create Energy Efficient Velocity Trajectory of Electric Vehicle

ND-000418 Motion Control of Large Inertia Loads Using Electrohydrostatic Actuation

ND-001244 Obstacle Avoidance in Path Following using Local Spline Relaxation

ND-000116 Robust Controller Design for Ball Screw Drives with Varying Resonant Mode via mu-synthesis

In the first part of the talk we present a new approach to state observation, called Parameter Estimation-based Observers (PEBO) whose main idea is to translate the state estimation problem into one of estimation of constant, unknown parameters. The class of systems for which is applicable is identified via two assumptions related to the transformability of the system into a suitable cascaded form and our ability to estimate the unknown parameters. The first condition involves the solvability of a partial differential equation while the second one requires some persistency of excitation-like conditions. We present also PEBO in a unified framework together with the - by-now classical - Kasantzis-Kravaris-Luenberger and Immersion and Invariance observers. In the second part we show that, for systems for which a linear regression-like relation is available, it is possible to combine PEBO with a new estimation technique called Dynamic Regressor Extension and Mixing (DREM). This new technique, called DREMBAO, is used to generate adaptive observers. PEBO and DREMBAO are shown to be applicable to position estimation of a class of electromechanical systems - including motors and MagLev systems - and for speed observation of a class of mechanical systems. The performance of these observers is compared with high-gain and sliding mode observers. As expected, it is shown that - in the presence of noise - the performance of the two latter designs is significantly below par with respect to the other techniques.

ND-000744 On Frequency Response Function Identification for Advanced Motion Control

ND-000124 Resonant Frequency Damping Disturbance Observer based Robot Servo System

ND-000701 Suppressing Position-Dependent Disturbances in Repetitive Control: With Application to a Substrate Carrier System

ND-000558 Vibration Amplitude Suppression Control of Industrial Machine Driven at Resonance Frequency

ND-001139 Contact Force Control Based on Force Estimation in Bimorph-type Piezoelectric Actuators

ND-001236 Verification of Double Hand Teleoperation System Using Haptic Forceps Robots and LCLM Platform

ND-001252 Experimental Setup for a Novel Mechanical Force Generator

ND-000655 Coactivation Method of Antagonistic Muscle Pairs Using Common and Differential Modes for Functional Electrical Stimulation Control

ND-000574 Minimum-Energy State Determination of an Underactuated Suction Cup Gripper Grid

ND-000566 Feature Extraction and Generation of Robot Writing Motion Using Encoder-Decoder Based Deep Neural Network

ND-000612 Admittance Control Based on a Stiffness Ellipse for Rapid Trajectory Deformation

ND-001201 Standing Assistance Control based on Voluntary Body Movement within Safety Tolerance

ND-001317 Offline and Online Tyre Model Reconstruction by Locally Weighted Projection Regression

ND-000469 Cooperative Adaptive Cruise Control Algorithms for Vehicular Platoons Based on Distributed Model Predictive Control

ND-000779 Modeling and field-experiments identification of vertical dynamics of vehicle with active anti-roll bar

ND-000647 Real-Time Model Predictive Control for a Parallel Hybrid Electric Vehicle using Outer Approximation and Semi-Convex Cut Generation

Reproducing the capabilities of visual sensing that one can find in nature would provide a very powerful and highly desired tool for robots. Ideally this enables a robot to perceive and interpret its surrounding so that it can use this information to execute different tasks within a real world environment. As robots operate in various environments (indoors, in space, in air, underwater) equipped with different sets of visual sensing devices (standard cameras, time-of-flight cameras, structured light cameras, hyperspectral imager) this makes the generic "interpretation of the world around a robot" very challenging. In this presentation I wish to introduce you to certain aspects within the world of "robotic vision" - where we try to teach machines to understand, plan and act in an intelligent way. In particular we will be concerned with how robots might build concepts about objects, understand relations between objects and understand the 3D structure of the surrounding world. The analysis of motion in the world of a robot is also an integral part of this understanding and important for many robotic control tasks.

ND-000264 Energy Analysis Method and Walking Simulation with Exoskeleton Assistive Devices

ND-000027 Remote Control Method with Tactile Sensation for Underwater Robot with Magnetic Coupling

ND-000582 Space-variant Color Point Cloud Measurement System - Enomous Data Reduction using Saliency Map -

ND-001228 Road and Intersection Detection Using Convolutional Neural Network

ND-000981 Haptic Interface for Virtual Reality based on Hybrid Cable-Driven Parallel Robot