Error Dynamics Based Dual Heuristic Dynamic Programming for Self-Learning Flight Control

A data-driven nonlinear control approach, called error dynamics-based dual heuristic dynamic programming (ED-DHP), is proposed for air vehicle attitude control. To solve the optimal tracking problem, augmented system defined by derived dynamics and reference trajectory so that actor neural network can learn feedforward feedback terms at same time. During online self-learning process, learns policy minimizing system’s value function. The input identified recursive least square (RLS) output of critic are used to update network. In addition, total uncertainty term also RLS, which compensate caused inaccurate modeling, parameter perturbation, on. outputs ED-DHP include rough trim surface, from network, compensation. Based on this scheme, complete knowledge not needed, offline learning unnecessary. verify ability ED-DHP, two numerical experiments carried out based established morphing model. One sinusoidal signal a fixed operating point, other guidance command with process variable points. simulation results demonstrate good performance validate robustness scheme