Roll-yaw control of flying wing aircraft at a high angle of attack based on jet control

The complex flow field structure and the interaction between vortex structures make the flying wing configuration aircraft prone to transverse uncommanded motion at a high angle of attack. To suppress the uncommanded motion, two sets of jet actuators are arranged on the vehicle using two existing active jet control techniques, the control effect of the actuators is verified through wind tunnel force measurement experiments, and the mutual coupling relationship between the two sets of jet actuators is clarified. A virtual flight experiment is conducted in the wind tunnel to capture the uncommanded motion of the flying wing configuration aircraft in the transverse direction, and two methods, PID and deep reinforcement learning, are applied to suppress the uncommanded motion in this kind of highly coupled and nonlinear problem. The wind tunnel experiments show that the deep reinforcement learning method is more effective in controlling the highly coupled and nonlinear problem, and the trained intelligent model can effectively suppress the transverse uncommanded motion of the flying wing configuration aircraft model.