摘要
提出一种基于RBF神经网络的Terminal滑模控制方案,消除通常滑模控制的到达过程,保证跟踪误差在有限时间内趋于零。不需要对建模误差、模型摄动和外界干扰进行各种假设,通过在线调整RBF神经网络的权值来消除它们的影响。最后在高超声速条件下,对空天飞行器再入大气层姿态控制进行仿真,结果表明该方法的有效性。
A robust adaptive flight control based terminal sliding mode is proposed, which eliminates the reaching phase of common sliding mode control, and guarantees the tracking errors converge to zero in finite time. The effects caused by modeling errors, uncertainties and disturbances are cancelled by RBF neural network through adaptive tuning its weights and without any assumptions. Simulation shows the effectiveness of the presented method.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2008年第2期304-307,共4页
Systems Engineering and Electronics
基金
国家自然科学基金资助课题(90405011
90716028)
