Optimal tracking control of the ship course system with partially-unknown dynamics

摘要

In this paper, an online adaptive optimal controller is obtained by using the policy iteration (PI) based integral reinforcement learning (IRL) technique for the ship course-tracking control the ship course-tracking system with partially-unknown dynamics. The performance index is defined to be quadratic with respect to the state and control input to obtain the linear quadratic tracking algebraic Riccati equation (LQT ARE). The IRL technique is employed to solve the LQT ARE problem online without the need for complete knowledge of the system dynamics. The simulation results via Matlab verified the effectiveness of the algorithm.