在复杂大气条件下的飞机自动着陆控制器设计与仿真

Controller Design and Simulation of Aircraft's Automatic Landing Under Complex Atmospheric Conditions

对飞机整个自动着陆过程的控制进行了设计,主要采用逆系统方法设计控制律,并用神经网络对控制律进行了鲁棒补偿,神经网络的学习规则为带有死区的关于神经网络灵敏度的线性函数。对于在自动着陆过程中可能遇到的紊流和风切变等几种典型大气情况进行了分析,并将所设计的着陆控制律在上述复杂大气条件下进行了仿真验证。仿真结果表明,所设计的自动着陆系统对复杂大气条件具有鲁棒性,着陆过程中实际飞行高度与期望飞行高度的误差在合理的范围之内。

The control of aircraft＇s automatic approach and landing was designed, mainly using inverse system method, compensated by neural network for robustness. The learning rule of neural network was a linear function of the neural network＇s sensitivity with dead zone. Some typical atmospheric conditions, such as turbulence and windshear, were analyzed, which might occur in landing. Some simulations were made to verify the control law under above-mentioned complex atmospheric conditions. The simulation results show that the automatic landing system has good robustness to complex atmospheric conditions, and the error between the real flight altitude in landing and reference flight altitude is in a reasonable range.