基于在线神经网络的无人机着陆飞行自适应逆控制器设计

基于在线神经网络设计了无人机着陆飞行自适应逆控制器。根据时标分离的原则，将无人机系统分解为快慢不同的四个回路，采用动态逆的方法设计快回路、慢回路和非常慢回路控制器，并且在慢回路和非常慢回路用基于在线神经网络的干扰观测器逼近无人机所受的扰动和动态逆误差，降低了控制器对干扰和模型精确度的要求，增强了控制器的鲁棒性。仿真结果说明所设计的无人机着陆控制器是非常有效的。

Design of Self Adaptive Inversion Flight Controller for UAV Landing Based on On-line Neural Network

A self adaptive inversion controller for UAV landing based on on-line neural network is designed in this paper. According to the time scale separation principle, the UAV system is divided into four subsystems. The dynamic inversion controllers are designed respectively for the quick loop, the slow loop and the more slow loop. The disturbance observer based on on-line neural network is used to approximate the disturbance and to on-line compensate the model inversion error. It reduces the restrain conditions on disturbance and the demand of model accuracy. Therefore, it improves the robust performance of the controller. Simulation demonstrated that the proposed landing controller is feasible and very effective.