Volume 6     Volume 5     Volume 4     Volume 3     Volume 2     Volume 1

Issue 4     Issue 3     Issue 2     Issue 1

Implementation of CMOS fuzzy controllers as mixed-signal integrated circuits
I. Baturone, S. Sanchez-Solano, A. Barriga, J. L. Huertas
Inst. de Microelectron. de Sevilla, Spain
vol. 5, no. 1, pp. 1-19, Feb. 1997

This paper discusses architectural and circuit-level aspects related to hardware realizations of fuzzy controllers. A brief overview on fuzzy inference methods is given focusing on chip implementation. The singleton or zero-order Sugeno's method is chosen since it offers a good tradeoff between hardware simplicity and control efficiency. The CMOS microcontroller described herein processes information in the current-domain, but input-output signals are represented as voltage to ease communications with conventional control circuitry. Programming functionalities are added by combining analog and digital techniques, giving rise to a versatile microcontroller, capable of solving different control problems. After identifying the basic component blocks, the circuits used for their implementation are discussed and compared with other alternatives. This study is illustrated with the experimental results of prototypes integrated in different CMOS technologies.

A VLSI fuzzy expert system for real-time traffic control in ATM networks
G. Ascia, V. Catania, G. Ficili, S. Palazzo, D. Panno
Istituto di Inf. e Telecommun., Catania Univ., Italy
vol. 5, no. 1, pp. 20-31, Feb. 1997

Concerns a fuzzy logic-based system which has been purposely designed to achieve real-time traffic control in high-speed networks using the asynchronous transfer mode (ATM) technique. One of the most critical functions is "policing", which has the task of ensuring that each user source complies with the traffic parameters negotiated in the call setup to avoid network congestion. This function is difficult to implement on account of certain conflicting requirements such as selectivity and responsiveness. This is confirmed by the severe limits affecting the most popular mechanisms proposed so far, based on conventional logic. The capacity to formalize approximate reasoning processes offered by fuzzy logic is exploited to derive rules of behavior for a policer starting from the know-how of an expert. We address two key issues related to the implementation of the fuzzy policer. The first focuses on the possibility of hardware implementation of the mechanism using VLSI technology; we present the design of a VLSI fuzzy processor which exhibits a level of performance of over 3 MFLIPS. The second issue concerns the suitability of applying the fuzzy policer to the policing of several classes of sources to reach high levels of cost effectiveness and scalability.

Fuzzy logic architecture using subthreshold analogue floating-gate devices
G. F. Marshall, S. Collins
DRA-CIS Div., Malvern, UK
vol. 5, no. 1, pp. 32-43, Feb. 1997

A subthreshold fuzzy logic architecture is proposed which includes floating-gate devices acting as nonvolatile analogue memories. The use of floating-gate devices which are embedded within the architecture reduces the size and power consumption of the system. Using the proposed subcircuits, a system containing 75 rules can be implemented in less than 5 mm/sup 2/ while consuming 500 mu W. The system is, therefore, suitable for integration within a smart sensor, an application area in which low-power consumption and compactness are potentially critical.

The synthesis of compact fuzzy neural circuits
J. F. Hurdle
Div. of Geriatrics, Utah Univ., Salt Lake City, UT, USA
vol. 5, no. 1, pp. 44-55, Feb. 1997

Concerns Smart Parts, a class of fuzzy neural hardware. Smart Parts are "smart" in that they can learn an input-output mapping implicit in a data set. They are "parts" in that they are small high-speed fuzzy neural processors meant to provide the fuzzy hardware designer customized functionality in a small package. They are application-specific. The paper focuses on the tool assisting this synthesis, TROUT. Using design heuristics that favor small size and high speed, TROUT can produce Smart Parts circuit specifications virtually automatically. It starts by minimizing the complexity of the target application data set in a way that simplifies the eventual implementation. It chooses a neural net or fuzzy network model from a small library that best suits the data set. In this paper, we detail the fuzzy min-max classifier model (FMM). TROUT optimizes FMM learning parameters to produce the smallest circuit offering the highest input vector throughout. Two architectural insights make the synthesis tractable. First, the FMM network architecture is structurally adaptive. Second, asynchronous circuit-design techniques are used because they simplify the synthesis process by eliminating clock scheduling. The output from TROUT is very high-level hardware description language (VHDL) code that can be used to synthesize the circuit in any of a number of circuit technologies. We outline the synthesis process and provide a circuit example based on the public domain wine classification data set.

Nonsingleton fuzzy logic systems: theory and application
G. C. Mouzouris, J. M. Mendel
Dept. of Electr. Eng. Syst., Univ. of Southern California, Los Angeles, CA, USA
vol. 5, no. 1, pp. 56-71, Feb. 1997

In this paper, we present a formal derivation of general nonsingleton fuzzy logic systems (NSFLSs) and show how they can be efficiently computed. We give examples for special cases of membership functions and inference and we show how an NSFLS can be expressed as a "nonsingleton fuzzy basis function" expansion and present an analytical comparison of the nonsingleton and singleton fuzzy logic systems formulations. We prove that an NSFLS can uniformly approximate any given continuous function on a compact set and show that our NSFLS does a much better job of predicting a noisy chaotic time series than does a singleton fuzzy logic system (FLS).

Selection of appropriate defuzzification methods using application specific properties
T. A. Runkler
Corp. Technol., Siemens AG, Munich, Germany
vol. 5, no. 1, pp. 72-9, Feb. 1997

Defuzzification is used to transform fuzzy inference results into crisp output. The standard defuzzification methods fail in some applications. It is, therefore, important to select appropriate defuzzification methods depending on the application. This paper presents some of the most important defuzzification methods and investigates their properties. With three application examples, it illustrates how to select appropriate defuzzification methods using application specific properties.

Radar tracking for air surveillance in a stressful environment using a fuzzy-gain filter
K. C. C. Chan, V. Lee, H. Leung
Dept. of Comput., Hong Kong Polytech., Kowloon, Hong Kong
vol. 5, no. 1, pp. 80-9, Feb. 1997

We present a fuzzy-gain filter for target tracking in a stressful environment where a target may accelerate at nonuniform rates and may also complete sharp turns within a short time period. Furthermore, the target may be missing from successive scans even during the turns, and its positions may be detected erroneously. The proposed tracker incorporates fuzzy logic in a conventional alpha - beta filter by the use of a set of fuzzy if-then rules. Given the error and change of error in the last prediction, these rules are used to determine the magnitude of alpha and beta . The proposed tracker has the advantage that it does not require any assumption of statistical models of process and measurement noise and of target dynamics. Furthermore, it does not need a maneuver detector even when tracking maneuvering targets. The performance of the fuzzy tracker is evaluated using real radar tracking data generated from F-18 and other fighters, collected jointly by the defense departments of Canada and the United States. When compared against that of a conventional tracking algorithm based on a two-stage Kalman filter, its performance is found to be better both in terms of prediction accuracy and the ability to minimize the number of track losses.

Subsystem inference representation for fuzzy systems based upon product-sum-gravity rule
Y. Yam
Dept. of Mech. & Autom. Eng., Chinese Univ. of Hong Kong, Shatin, Hong Kong
vol. 5, no. 1, pp. 90-107, Feb. 1997

A new representation which expresses a product-sum-gravity (PSG) inference in terms of additive and multiplicative subsystem inferences of single variable is proposed. The representation yields additional insight into the structure of a fuzzy system and produces an approximate functional characterization of its inferred output. The form of the approximating function is dictated by the choice or polynomial, sinusoidal, or other designs of subsystem inferences. With polynomial inferences, the inferred output approximates a polynomial function the order of which is dependent on the numbers of input membership functions. Explicit expressions for the function and corresponding error of approximation are readily obtained for analysis. Subsystem inferences emulating sinusoidal functions are also discussed. With proper scaling, they produce a set of orthonormal subsystem inferences. The orthonormal set points to a possible "modal" analysis of fuzzy inference and yields solution to an additive decomposable approximation problem. This work also shows that, as the numbers of input membership functions become large, a fuzzy system with PSG inference would converge toward polynomial or Fourier series expansions. The result suggests a new framework to consider fuzzy systems as universal approximators.

On hardware support for interval computations and for soft computing: theorems
H. T. Nguyen, V. Kreinovich, V. Nesterov, M. Nakamura
Dept. of Math. Sci., New Mexico State Univ., Las Cruces, NM, USA
vol. 5, no. 1, pp. 108-27, Feb. 1997

This paper presents a rationale for providing hardware supported functions of more than two variables for processing incomplete knowledge and fuzzy knowledge. The result is in contrast to Kolmogorov's (1957) theorem in the numerical (nonfuzzy) case.

A method for design of a hybrid neuro-fuzzy control system based on behavior modeling
Li Wei
Dept. of Comput. Sci. & Technol., Tsinghua Univ., Beijing, China
vol. 5, no. 1, pp. 128-37, Feb. 1997

It is known that control signals from a fuzzy logic controller are determined by a response behavior of a controlled object rather than its analytical models. That implies that the fuzzy controller could yield a similar control result for a set of plants with a similar dynamic behavior. This idea lends to modeling of a plant with unknown structure by defining several types of dynamic behaviors. On the basis of dynamic behavior classification, a new method is presented for the design of a neuro-fuzzy control system in two steps: 1) we model a plant with unknown structure by choosing a set of simplified systems with equivalent behavior as "templates" to optimize their fuzzy controllers off-line; and 2) we use an algorithm for system identification to perceive dynamic behavior and a neural network to adapt fuzzy logic controllers by matching the "templates" online. The main advantage of this method is that convergence problem can be avoided during adaptation process. Finally, the proposed method is used to design neuro-fuzzy controllers for a two-link manipulator.

Designing fuzzy controllers from a variable structures standpoint
J. S. Glower, J. Munighan
Dept. of Electr. Eng., North Dakota State Univ., Fargo, ND, USA
vol. 5, no. 1, pp. 138-44, Feb. 1997

A procedure is presented for designing fuzzy controllers based upon variable structures techniques. Three such controllers are presented: the fuzzy equivalents of sliding-mode controllers, saturating controllers, and tanh controllers. By using an approach based upon variables structures (VSS) techniques, the stability of each of these controllers is assured. By using a sliding surface, the order of the rule base is reduced to size r*m, where r is the number of inputs and m is the number of fuzzification levels. This combination makes the proposed design procedure able to generate simple controllers with guaranteed stability properties. To illustrate the proposed design procedure, fuzzy controllers are designed for a ball-and-beam system. It is demonstrated that in spite of this system being a fourth-order unstable system, the proposed design procedure results in simple stable fuzzy controllers.

On the stability of fuzzy systems
M. A. L. Thathachar, P. Viswanath
Indian Inst. of Sci., Bangalore, India
vol. 5, no. 1, pp. 145-51, Feb. 1997

Studies the global asymptotic stability of a class of fuzzy systems. It demonstrates the equivalence of stability properties of fuzzy systems and linear time invariant (LTI) switching systems. A necessary and sufficient condition for the stability of such systems are given, and it is shown that under the sufficient condition, a common Lyapunov function exists for the LTI subsystems. A particular case when the system matrices can be simultaneously transformed to normal matrices is shown to correspond to the existence of a common quadratic Lyapunov function. A constructive procedure to check the possibility of simultaneous transformation to normal matrices is provided.