Dynamic Soaring Trajectory Optimization and Tracking with Adaptive Non-singular Fast Terminal Sliding Mode Control

Dynamic soaring is an effective method to improve the endurance time and flight distance of small unmanned air vehicles. However, effective use the gradient change of the wind field to plan and track UAV trajectory has always been the focus and difficulty of current research. To solve this problem, trajectory optimization and tracking control for dynamic soaring of small unmanned air vehicles are presented in this paper. Firstly, the six degrees of freedom dynamics model with a consideration of the wind information is established. Secondly, soaring trajectory planning is formulated as an optimization problem and solved by using the Gaussian Pseudospectral method. Thirdly, a cascade dual loop scheme is used to design a tracking controller, where the outer loop employs a PID controller with virtual control signals for position control. A sliding mode control with adaptive non-singular fast terminal sliding mode control is proposed for inner loop attitude control. This method is robust to system parameter perturbation and uncertainties. Simulation shows that the proposed method can generate the trajectory in the gradient wind for dynamic soaring and ensure the aircraft tracks this trajectory even if in the presence of disturbance.