Towards an approach of synthesis, validation and implementation of distributed control for AMS by using events ordering relations

Abstract

We study distributed control synthesis and validation for automated manufacturing systems (AMS) in the framework of supervisory control theory. To reduce the size of the control problem, we view an AMS as comprised of asynchronous subsystems which are coupled through imposed logical Boolean specifications. The principle of the distributed control approach is the decomposition of the global monolithic control action into local coordinated control strategies for the individual subsystems. Owing to its importance in a distributed scheme, the order in which events occur arouses interest. By extending our previous results, we develop a set of rules of events precedence ordering, under which the control strategy via decomposition promises the subsystems synchronisation and coordination. We show how these rules contribute to reduce the size of the controller models used in the verification/validation and implementation steps. The effectiveness of the proposed approach is demonstrated by means of an industrial AMS example.