Concurrent learning-based online approximate feedback-Nash equilibrium solution of N-player nonzero-sum differential games

摘要

This paper presents a concurrent learning-based actor-critic-identifierarchitecture to obtain an approximate feedback-Nash equilibrium solution to aninfinite horizon N-player nonzero-sum differential game online, withoutrequiring persistence of excitation (PE), for a nonlinear control-affinesystem. Under a condition milder than PE, uniformly ultimately boundedconvergence of the developed control policies to the feedback-Nash equilibriumpolicies is established.