论文部分内容阅读
操纵面故障会直接影响无人机执行任务的效率和生存力。针对某多操纵面无人机给出了一种神经网络自适应重构控制方法,并在理论上证明了该方法的有效性。该重构控制方法可对建模不准确和操纵面故障(卡死、受损或松浮)等因素引起的逆误差进行有效补偿。仿真结果表明,所设计的神经网络重构控制方法,不仅可以补偿系统建模误差和参数摄动对控制系统的影响,还可以对故障操纵面在线重构控制律,使操纵面故障飞机能够继续执行任务,提高了飞机的稳定性、操纵性和生存力。
Control surface failure will directly affect the efficiency and survivability of the UAV mission. Aiming at a multi-control surface UAV, a neural network adaptive reconfiguration control method is given, and the validity of the method is proved theoretically. The reconfigurable control method can effectively compensate for the inverse error caused by inaccurate modeling and control plane failures (stuck, damaged or loose). The simulation results show that the proposed neural network reconfiguration control method can not only compensate the influence of system modeling error and parameter perturbation on the control system, but also reconstruct the control law of the fault control surface online and make the control plane fault continue. Performing the mission improved the stability, maneuverability and viability of the aircraft.