Adaptive Identifier-Critic-Based Optimal Tracking Control for Nonlinear Systems With Experimental Validation

Adaptive Identifier-Critic-Based Optimal Tracking Control for Nonlinear Systems With Experimental Validation

Adaptive Identifier-Critic-Based Optimal Tracking Control for Nonlinear Systems With Experimental Validation 150 150 UKAEA Opendata
UKAEA-RACE-PR(21)02

Adaptive Identifier-Critic-Based Optimal Tracking Control for Nonlinear Systems With Experimental Validation

This article presents and practically validates an identifier-critic-based approximate dynamic programming (ADP) method to online address the optimal tracking control problem for nonlinear continuous-time unknown systems. The imposed assumption on precisely known system dynamics is obviated via a neural network (NN) identifier. A static control is first adopted to retain the steady-state tracking response, while an optimal control derived via the ADP  method is proposed to regulate the tracking error by minimizing a cost function. A critic NN is then trained online to obtain the solution of the associated Hamilton–Jacobi–Bellman (HJB) equation. The learning of the identifier NN and critic NN is performed online simultaneously by tailoring a novel adaptation method, which can guarantee the convergence of the estimated NN weights. Consequently, the critic NN can be used to construct the optimal control policy directly, such that the actor NN used in the previous ADP schemes is avoided. Simulations are performed to verify the suggested control, and experiments on a helicopter plant are carried out to show its feasibility and improved control response.

Collection:
Journals
Journal:
IEEE Transactions on Systems, Man, and Cybernetics: Systems
Publisher:
IEEE
Published date:
07/02/2020