基于BP神经网络的飞行器参数辨识与自适应控制

Parameter Identification and Adaptive Control of Aircraft Based on BP Neural Network

针对跨域空间飞行器气动参数非线性严重和具有大不确定性等特点,提出一种基于BP神经网络的飞行器离线参数辨识与在线自适应控制的方法。首先,临近空间飞行器进行风洞试验吹风得到的气动参数是典型的输入输出非线性系统,运用BP神经网络算法进行离线训练建立气动数据的辨识模型;其次,根据气动数据的辨识模型计算实时舵效变化参数,飞行器控制的增益根据舵效变化完成在线自适应调节,实现飞行器的自适应姿态控制;最后,进行数学仿真验证,结果表明,将BP神经网络应用于飞行器姿态控制中,能够实现控制参数的自适应调整,说明BP神经网络具有优良的逼近性能,最终提升了飞行器姿态控制系统的性能,提高了智能化设计水平。

A method of off-line parameter identification and on-line adaptive control of cross domain space vehicle based on BP neural network is proposed.Firstly,the aerodynamic parameters obtained from wind tunnel tests of near space vehicles are typical from input-output nonlinear systems,and the identification model of aerodynamic data is established by using off-line training with BP neural network algorithm;Secondly,the real-time rudder efficiency variation parameters are calculated according to the identification model of aerodynamic data,and the control gain of the aircraft is self-adaptive adjusted according to the change of rudder effect.Finally,the adaptive attitude control of the aircraft is implemented.The results show that the adaptive adjustment of control parameters can be realized by applying BP neural network to the attitude control of aircraft,which shows that BP neural network has excellent approximation performance and finally improves the performance of aircraft attitude control system and improves the ability of intelligent design.