Title :
Inductive inference of logical DES controllers using the L* algorithm
Author :
Yang, Xiaojun ; Lemmon, Michael ; Antsaklis, Panos
Author_Institution :
Dept. of Electr. Eng., Notre Dame Univ., IN, USA
Abstract :
It is well known that learning procedures such as the L* algorithm will infer minimal deterministic finite automatons (DFA) in polynomial time through the use of membership and equivalence queries. This paper introduces a modification of the L*-algorithm that can be used for the inductively inferring optimal logical DES controllers in which prior knowledge of the plant is confined to a finite lookahead window of predicted behaviours
Keywords :
computational complexity; control system synthesis; discrete event systems; finite automata; inference mechanisms; learning (artificial intelligence); real-time systems; L* algorithm; deterministic finite automatons; discrete event systems; equivalence queries; inductive inference; learning procedures; logical DES controllers; optimal control; polynomial time; Automatic control; Control system synthesis; Discrete event systems; Doped fiber amplifiers; Inference algorithms; Learning automata; Marine vehicles; Optimal control; Polynomials; Uncertainty;
Conference_Titel :
American Control Conference, Proceedings of the 1995
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-2445-5
DOI :
10.1109/ACC.1995.532099
