Showing papers by "Bernard P. Zeigler published in 1995"

Designing fuzzy net controllers using genetic algorithms

Abstract: As control system tasks become more demanding, more robust controller design methodologies are needed A genetic algorithm (GA) optimizer, which utilizes natural evolution strategies, offers a promising technology that supports optimization of the parameters of fuzzy logic and other parameterized nonlinear controllers This article shows how GAs can effectively and efficiently optimize the performance of fuzzy net controllers employing high performance simulation to reduce the design cycle time from hours to minutes Our results demonstrate the robustness of a GA-based computer-aided system design methodology for rapid prototyping of control systems >

DEVS Framework for Modelling, Simulation, Analysis, and Design of Hybrid Systems

Abstract: We make the case that Discrete Event System Specification (DEVS) is a universal formalism for discrete event dynamical systems (DEDS). DEVS offers an expressive framework for modelling, design, analysis and simulation of autonomous and hybrid systems. We review some known features of DEVS and its extensions. We then focus on the use of DEVS to formulate and synthesize supervisory level controllers.

Inductive Modeling: A Framework Marrying Systems Theory and Non-monotonic Reasoning

Abstract: This article develops a framework for inductive modelling that works at the input/output level of system description. Rather than attempt to construct a state-space model from given observed data, an inductive modeler can employ non-monotonic logic to manage a data base of observed and hypothesized input/output time segments. Also, some basic criteria are established to guide the evaluation of the inductive modeler's performance.

Alternative Analysis for Computational Holon Architectures.

Abstract: : This effort investigated programming languages computer processors and simulation techniques capable of supporting the development of an advanced framework for human performance process model research The framework, the holon cognitive architecture, conceptualizes the mind as consisting of a set of interactive agents These agents are abstract models of neuronal activity occuring in the brain This document contains an analysis of the computation requirements needed to simulate human performance process models based upon the holon cognitive architecture and suggest a growth path based upon a heterogeneous computing environment (KAR) P 3

DEVS-Based Endomorphic Agents: Control Through Deliberative and Reactive Planning

Abstract: A DEVS-based Endomorphic Agent employs models of the environment and itself to achieve autonomy in various way. In this paper, we present a framework for an agent able to flexibly reprogram itself under changing objectives or constraints such as might be provided in a flexible manufacturing environment. We show how the agent is able to employ its DEVS-models to develop control schemes for given objectives and constraints. The schemes can then be revised as required when a change in the desired objectives or constraints occurrs. The systems model knowledge representation thus gives the agent greater flexibility and autonomy than is possible with purely rective planners or supervisory controllers.