Optimal control of robot behaviour using language measure

Abstract

This paper presents optimal control of robot behaviour in the discrete event setting. Real signed measure of the language of supervised robot behaviour serves as the performance index for synthesis of the optimal policy. The computational complexity of control synthesis is polynomial in the number of states of the deterministic finite state automaton model that is generated from the regular language of the unsupervised robot behaviour. The results of simulation experiments on a robotic test bed are presented to demonstrate the efficacy of the proposed optimal control policy.