A Controller Design of Variable Wingspan Aircraft Based on Adaptive Dynamic Programming

Variable wingspan aircraft have a wide range of applications by dynamically changing the wing shape and structure to adapt to different flight conditions and mission requirements. Thus, it can optimize the flight performance, improve the fuel efficiency and extend the flight envelope. In this paper, a control scheme based on adaptive dynamic programming (ADP) is proposed for the control problem of variable wingspan aircraft while considering the input bounded case. The article first establishes a longitudinal nonlinear dynamics model of the variable wingspan aircraft and carries out the linearization process, and then designs a control law based on the actor-critic mechanism of reinforcement learning. We utilize the approximation ability of neural network (NN) to accomplish the online learning process of the control strategy and verify the validity of the proposed algorithm through numerical simulation. By choosing an appropriate Lyapunov function, we can prove the convergence of this control algorithm. The simulation results show that the algorithm is able to impose significant constraints on the inputs and is more in line with the real situation than the ordinary ADP control method.