Model free adaptive optimal tracking controller design for AFS/DYC based integrated chassis control system
Zhi‐Jun Fu, Bin Li, Xiaobin Ning · 2017
In view of the problem of poor robustness and adaptability in current AFS/DYC-based vehicle integrated chassis control system, a model free optimal tracking control approach based on approximate dynamic programming theory is proposed. Adaptive optimal control law is obtained from online solution of the Hamilton-Jacobi-Bellman (HJB) equation using a recurrent NN identifier and a critic NN to identify the unknown dynamics and the optimal value function, respectively. A novel improved adaptive law for the critic NN is proposed to achieve fast convergence. Lyapunov theory proves that the proposed method can make the system state track the desired trajectory and stabilize the error dynamic in an optimal way. Model free and self-adaptive properties of the proposed approach provides a new solution for the integrated AFS and DYC controller design instead of the commonly used model based method. Simulation results show that the proposed method demonstrates strong robustness and good adaptability in terms of enhancing vehicle stability performance when encountering uncertain cornering stiffness.