摘要
利用将Backstepping方法和神经网络相结合,设计了一类非线性不确定系统的鲁棒自适应控制器。该文利用在线神经网络控制器,从而,不要求系统不确定部分有界。通过李亚谱诺夫方法,验证了系统的稳定性;阐述了该系统具有较好的鲁棒性。最后,将所设计的控制器用于仿真实例,得到很好的结果,说明了该方法的有效性。
An intelligent backstepping control scheme is proposed for a class of nonlinear uncertain systems with complete state available. The idea on second order sliding mode control is used for reference, therefore a feedback linearization controller is designed in the sense of the backstepping technique. A RBF network is employed to online approximated nonlinear uncertain system. This control can ensure the stability of closed loop system and the asymptotical convergence of tracking error, moreover posses robustness to uncertainties. The effectiveness of the proposed control scheme is verified by a simulated result.
出处
《计算机仿真》
CSCD
2004年第5期71-73,77,共4页
Computer Simulation
关键词
不确定性
神经网络
鲁棒性
Uncertainty
Neural network
Robustness
