Adaptive Concurrent Learning Algorithm Based on Pontryagin's Maximum Principle for Nonlinear System Optimal Tracking Control with State Inequality Constraints

In this paper, an adaptive iterative learning algorithm is proposed to solve the optimal tracking control problem (OTCP).Unlike the existing method, we select the finite-time horizon cost function to measure tracking performance.Our method doesn't require an accurate system dynamic and the concurrent learning (CL) technique is utilized to learn the system identification model.Based on identification model, we present our concurrent iterative learning algorithm under the Pontryagin's framework to learn the solution for OTCP.The proposed algorithm overcomes the limitation of Adaptive Dynamic Programming (ADP) methods when dealing with time-varying systems or situations involving state inequality constraints.The algorithm's effectiveness is demonstrated through a numerical simulation.