Reinforecement learning-based optimal tracking control for wheeled mobile robot

摘要

This paper proposes a new method to design a reinforcement learning-based integrated kinematic and dynamic tracking control scheme for a nonholonomic wheeled mobile robot. The scheme uses just only one neural network to design an online adaptive synchronous policy iteration algorithm implemented as an actor critic structure. Our tuning law for the single neural network not only learns online a tracking-HJB equation to approximate both the optimal cost and the optimal control law but also guarantees closed-loop stability in real-time. The convergence and stability of the overall system are proven by Lyapunov theory. The simulation results for wheeled mobile robot verify the effectiveness of the proposed controller.