一种基于神经网络的快速回馈递推自适应控制

针对不确定严格反馈块控非线性系统,提出了一种快速回馈递推自适应控制方法。系统的不确定性及外界干扰由RBF神经网络在线逼近,利用动态面控制技术简化回馈递推方法的控制律,同时改进参数自适应律,使在线调整自适应参数显著减少,提高了控制算法的计算效率。基于Lyapunov方法证明了闭环系统所有信号有界,跟踪误差指数收敛到有界紧集内。最后进行了空天飞行器飞行控制系统设计,并在高超声速的条件下对其进行了仿真验证,结果表明了该方法的有效性。

A Fast Adaptive Backstepping Method Based on Neural Networks

A fast adaptive backstepping method based on neural networks is presented for a general class of strict-block-feedback uncertain nonlinear systems.RBF neural networks are employed to compensate for the uncertainties and disturbances on line.The problem of explosion of complexity inherent in the existing backstepping method is eliminated and the complexity of control law is reduced by incorporating dynamic surface control technique into the design framework of backstepping methodology.Adaptive tuning laws are improved so that the number of parameters needed to be adapted online decreases obviously and calculation efficiency is increased greatly.All signals in the closed-loop system are guaranteed to be ultimately bounded by Lyapunov approach and outputs of the system exponentially converge to a small neighborhood of the desired trajectory.Finally,the flight control system of an Aerospace Vehicle(ASV) is designed based on the proposed method.Six-degree-of-freedom(6 DOF) numerical simulation results demonstrate the effectiveness and robustness of the control scheme.