Data-Driven Supervisory Control of Discrete-Event Systems With Forcible Events
Chao Gu, Chao Gao, Kai Cai · IFAC-PapersOnLine · 2024
This paper explores the problem of analysis and control towards model-unknown discrete-event systems. Using data, i.e. partly observed output event sequences and some pre-established knowledge, we apply an event-forcing mechanism to verify an extended concept of controllability, namely forcible-controllability, and develop control strategies with respect to the observed specification language. In particular, for the given data, a property namely forcible-informativity is proposed, representing a criterion for determining the forcible-controllability of the observed specification language. The data-driven supervisory control is also devised accordingly, if feasible.