Adaptive Dynamic Programming-Based Active Disturbance Rejection Control for Autonomous Vehicle Tracking

Nowadays, autonomous driving technology is more closely connected with human society, but the complex urban road environment puts higher requirements on the autonomous vehicle's robustness and fast response ability. To solve the problems mentioned above, this paper proposes a trajectory tracking control method with anti-interference and high real-time performance. The active disturbance rejection control approach is utilized to reject external disturbance and internal model uncertainties with the model predictive control controller. To speed up the computation, parameterized adaptive dynamic programming is used to estimate the optimal control sequence. A double-lane change demonstration is used to test and verify the proposed method. The experimental results show that our method improves the anti-interference ability of the system while speeding up the control speed. The proposed method can maintain the lateral error within 0.09m even with imposing 0.05rad disturbance to the input, and the single-step control time is half the traditional MPC.