The design of logic controllers for event-driven systems continue to rely largely on intuitive methods rather than on formal techniques. This approach results in a control code that requires extensive verification, is hard to maintain and modify, and may even fail at times. Supervisory control theory (SCT) provides a formal approach to logic control synthesis. In order to demonstrate the usefulness of the supervisory control theory in manufacturing systems, an educational test-bed that simulates an automated car assembly line has been built using LEGO/spl reg/ blocks. Finite state machines (FSMs) are used for modeling operations of the assembly line, and for the specifications that accomplish the task of successfully completing the assembly repeatedly. Using the technique of SCT, we derive a supervisor that enforces the specifications while offering the maximum flexibility of assembly. Subsequently a controller is extracted from the maximally permissive supervisor for the purpose of implementing the control by selecting, when possible, at most one controllable event from among the ones allowed by the supervisor. Testing to check the correctness of the control code is reduced, since the controller is guaranteed to enforce the specifications.
[1]
P. Ramadge,et al.
Modular Supervisory Control of Discrete Event Systems
,
1988
.
[2]
Vijay K. Garg,et al.
Modeling and Control of Logical Discrete Event Systems
,
1994
.
[3]
Ratnesh Kumar,et al.
A computer implementable algorithm for the synthesis of an optimal controller for acyclic discrete event processes
,
1999,
Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C).
[4]
Lynn Underwook,et al.
Intelligent Manufacturing
,
1998
.
[5]
Karl-Erik Årzén.
Grafcet for intelligent supervisory control applications
,
1994,
Autom..
[6]
P. Ramadge,et al.
Supervisory control of a class of discrete event processes
,
1987
.
[7]
F. Duncan,et al.
Programmable logic controllers: architecture and applications
,
1990
.
[8]
Shengbing Jiang,et al.
Modeling discrete event systems with faults using a rules based modeling formalism
,
2002,
Proceedings of the 41st IEEE Conference on Decision and Control, 2002..