一类不确定非线性系统的回馈递推滑模鲁棒控制器设计  

Robust Backstepping Sliding Mode Controller Design Approach for a Class of Uncertain Nonlinear Systems

作　　者：曹邦武[1] 姜长生[1] 

机构地区：[1]南京航空航天大学自动化学院

出　　处：《宇航学报》2005年第6期818-822,共5页Journal of Astronautics

摘　　要：针对一类多输入多输出不确定非线性系统,用一个RBF神经网络获得未建模动态和外来扰动的估计值,将回馈递推思想与滑模控制相结合,逐步回馈递推,设计了鲁棒控制器。将这种方案用于某型侧滑转弯导弹的飞行控制系统设计。其中,为克服导弹作动器的位置饱和、速率饱和问题,改进了滑动面的符号函数,并采用模糊逻辑对滑动面的斜率进行实时调整。仿真表明了该控制方法的有效性。For a class of multi-input-multi-output uncertain nonlinear system, a robust controller based on backstepping and sliding mode control was designed. The unknown dynamics and external disturbances were estimated by a RBF neural network. Then the scheme was applied to design the autopilot system of a skid-to-slide missile. Considering the missile＇s actuator position saturation and velocity saturation, the sign function of the slide mode was improved and fuzzy logic was introduced to adjust the slopes of the sliding mode surfaces. The simulation results show its effectiveness.

关 键 词：回馈递推 滑模 模糊控制 神经网络 导弹 飞行控制 作动器饱和 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]