Powered Parafoil Lateral-directional Attitude Angle Control with Adaptive Neural Network

A lateral-directional relative yaw angle control scheme based on adaptive neural network used to identify online the dynamic erroris proposed to design the command-tracking controller for a powered parafoil system (PPS). The PPS has been studied as a two-body system, similar to a double pendulum, with flexible structure and 8 degrees of freedom (DOF); and, it is difficult to present a precision model. Due to the uncertainties of modeling and the external disturbances, an adaptive parameter adjuster is introduced into the RBF neural network to approximate the dynamic error, which effectively improves the original system's characteristic of poor damping of attitude oscillation. And the adaptive neural network system could not only improve the robustness of the system further, but also improve the tracking precision in the process of adaptive parameters converging. Simulation results show that the proposed scheme can track the maneuver command with higher precision and faster response speed than the natural one, and it is applicable to design the PPS attitude angle controller.