基于RBF神经网络滑模自抗扰的四旋翼飞行器控制

Control of Quadrotor aircraft based on RBF neural network sliding mode and active disturbance rejection control

针对四旋翼飞行器参数不确定性和易受外部干扰的问题,提出一种以径向基函数(Radial Basis Function,RBF)神经网络滑模控制(RBF-SMC)为外环、自抗扰控制为内环的内外环嵌套的控制策略.首先,根据牛顿-欧拉方程建立四旋翼飞行器数学模型,对四旋翼飞行器的位置和姿态进行控制,外环采用RBF神经网络滑模控制,基于RBF神经网络用来对模型干扰不确定项进行补偿,对其增益进行实时调节,抑制滑模控制带来的“抖振”问题,内环采用自抗扰控制,通过扩张状态观测器可以实现观测和补偿内部耦合等参数不确定性和外部干扰;然后,使用Lyapunov方法证明闭环系统稳定性;最后,应用Matlab/Simulink平台进行仿真实验.实验结果表明,设计的控制器不仅能抑制外界干扰,而且能实现期望飞行轨迹的准确跟踪.

To address the problems of parameter uncertainty and susceptibility to external disturbance of Quadrotor aircraft,a nested inner-outer ring control strategy is proposed,with the RBF(Radial Basis Function)neural network sliding-mode control(RBF-SMC)as the outer ring,and with active disturbance rejection control(ADRC)as the inner ring.Firstly,a mathematical model of the four-rotor aircraft is built according to the NewtonEuler equation,which can be used to control the position and attitude of the four-rotor aircraft.The RBF neural network sliding-mode control is adopted for the outer ring.Uncertain terms of model disturbance are compensated based on the RBF neural network,its gain is adjusted in real time,and the"chattering"problem caused by slidingmode control is suppressed.The active disturbance rejection control is adopted for the inner ring.The extended state observer(ESO)can be used for observation,and to compensate for parameter uncertainty(such as internal coupling)and external disturbance.Then,the Lyapunov method is used to prove the stability of the closed-loop system.Finally,simulation experiments are carried out in the Matlab/Simulink platform.The simulation results show that the designed controller can not only suppress external disturbance,but also achieve accurate tracking of expected flight trajectory.