Autonomous Navigation of Multiple Robots using Supervisory Control Theory

Abstract

This paper proposes a hybrid structure for autonomous navigation of multiple robots with tasks assigned by a centralized scheduler. Considering the robots as Discrete Event Systems (DES), the Supervisory Control Theory (SCT) is used to model and control individually the behavior of each robot of the system. Particularly, gathering deliberative and reactive motion planning algorithms through a structured procedure, it allows to obtain a hybrid planner able to perform safe navigation in a mapped workspace with unknown obstacles. Therefore, initially, the open-loop behavior and the reduced supervisors for the robots are provided and, later, a pseudocode describing the scheduler and its interaction with the robots is presented. Furthermore, as a case study, simulations considering mobile robots are carried out to corroborate the proposed framework.