Adaptive neural tracking control of constrained waverider vehicles via single-network adaptive dynamic programming

The approximate optimization is considered though adaptive dynamic programming (ADP) method for waverider vehicles (WVs). This method can achieve the stability, robustness, and near optimization of the tracking problem for WVs. First, baseline controllers are designed for both velocity subsystem and altitude subsystem via neural network approximation. Meantime, novel auxiliary systems are implemented to compensate the desired control law with actuator saturation. Then, transient optimal controllers are developed using single-network adaptive critic method. Moreover, the stability of closed-loop systems, and the convergence of both the tracking error and the auxiliary system are theoretically proved. Finally, the superiority of the proposed method is proved via compared simulation results.

Automated summary

This study presents an intelligent optimization control method for waverider vehicles based on adaptive dynamic programming (ADP). The method achieves stability, robustness, and near optimization of the tracking problem for waverider vehicles by designing baseline controllers, implementing novel auxiliary systems, and developing transient optimal controllers using a single-network adaptive critic method.