Distributed optimal cooperative control for nonlinear multi-agent systems with input saturation and full-state constraint

Abstract

The issue of distributed optimal cooperative control for nonlinear multi-agent systems (MASs) with input saturation and full-state constraint is discussed in this paper. Different with the traditional states constraint, the dynamic and asymmetric constraint conditions are considered and a barrier function is employed. Besides, a smooth function is constructed to map the input saturation. The adaptive dynamic programming (ADP) is utilized to seek distributed optimal control policy and the predefine cooperative performance index function is estimated by a critic neural network (CNN), in which a weight updating law is proposed to learn the ideal weights. The closed-loop system and weight estimation errors are proved to be uniformly ultimately bounded(UUB). Finally, simulation results testifies to the viability of the design algorithm.