Event-Triggering-Learning-Based ADP Control for Post-Stall Pitching Maneuver of Aircraft
提出一种改进的事件触发学习自适应动态规划方法，用于飞机失速后俯仰机动，实现鲁棒最优控制并降低计算成本，通过非线性扰动观测器抑制气动扰动，仿真验证了有效性。
In this article, an improved event-triggering-learning (ETL)-based adaptive dynamic programming (ADP) method for the post-stall pitching maneuver of aircraft is proposed to achieve the robust optimal control and reduce the computational cost. First, a feedforward control with the nonlinear disturbance observer (NDO) technique is designed to attenuate the adverse effects caused by the unsteady aerodynamic disturbances. Subsequently, the ADP method with a critic neural network which is constructed to approximate the value function in the Hamilton-Jacobi-Bellman equation is employed to conduct the optimal control of aircraft. In addition, to reduce the computational cost of learning, the event-triggering (ET) mechanism with an improved ET condition is applied. The Lyapunov stability theory is utilized to prove that all signals in the closed-loop control system are uniformly ultimately bounded. Finally, simulation results are presented to illustrate the effectiveness of the proposed ETL-based ADP method.