Jing-Yue Lin

Verifying a class of nondeterministic discrete event systems in a generalized temporal logic

A formal verification method is proposed for a class of nondeterministic discrete event systems in which point probability distributions are known. Such systems are modeled by bounded stochastic models that specify their structures. A linear-time temporal logic is generalized to an uncertain environment in order to establish a framework for system verification. Soundness and completeness of the proof system are also discussed. Properties of this class of system are verified by deducing temporal logic specifications of desired behavior from descriptions of the system dynamics. The formulas do not mention probabilities explicitly, so that the specifications and verification are completed qualitatively without using the probability theory. The verification is illustrated by an example of a flexible manufacturing system. >

Asymptotic behaviour of output feedback for a class of non-deterministic discrete event systems

A class of non-deterministic discrete event systems (DES) where the point probability distributions are known is considered. The non-deterministic DES models are introduced by using minimax algebra to formalize the treatment of time sequences, and by the usual classical algebra to introduce a markovian structure that assigns the transition probabilities to events. Based on these representations, the closed non-deterministic DES models are derived by output feedback control. The asymptotic behaviour of the system is investigated to predict the long range performance of discrete event processes. The results are illustrated by some examples of flexible manufacturing systems.

A Temporal Logic Approach to the Control of Discrete Event Systems

Recently there has been considerable interest in the development of a control theory for discrete event dynamical systems (DEDS). Temporal logic has been applied into the control problems for DEDS in [1-2] and [5]. In this paper, we will define an event structure for describing discrete event behavior. A temporal logic model (TLM) is then built for modelling DEDS. A composition operation on temporal logic models will be described to provide the basis for our required control action. This will kad naturally to, in Section 5, the outline of a procedure for developing a controller which, when composed with the plant, will ensure that the closed-loop system has the required behavior.

Optimal robust stabilization compensator design : a deterministic approach

The problem of designing multivariable stabilization controllers that satisfy an optimal condition is considered in this paper. Specifically, it is required to find a robust stabilization controller so that closed-loop stability and asymptotic regulation occur, and also so that a quadratic performance index will be minimized. Necessary and sufficient conditions for a robust optimal output-feedback stabilization are presented. The gradient is considered over a functional defined on matrix spaces. Conditions for an optimal point are derived for this kind of functional. Finally, an algorithm for the robust optimal output-feedback design and an example are given.

Specification of Intelligent Controllers for Discrete Event Systems in a Temporal Logic Framework

Abstract In this paper, a temporal logic model is used for designing controllers. The plant is formalized as a system driven by inputs and events that force a sequence of states in the system state space. Temporal logic models are built for the plant, and controller; the closed loop system specification are also written in the same language. A synthesis method for building an intelligent controller is developed based on the reachability property. An example demonstrates the viability of this approach.