一类非线性MIMO系统鲁棒自适应神经网络DSC设计  

DSC design of a robust adaptive NN control for a class of nonlinear MIMO systems

作　　者：李铁山[1,2] 王晓飞[1] 杨新宇[2] 

机构地区：[1]上海交通大学船舶海洋与建筑工程学院,上海200030 [2]大连海事大学航海学院,辽宁大连116026

出　　处：《哈尔滨工程大学学报》2009年第2期121-125,共5页Journal of Harbin Engineering University

基　　金：国家自然科学基金资助项目(60874056,50779033);中国博士后科学基金资助项目(20070420101);上海市博士后科研资助计划资助项目(07R214128);国家高技术研究发展(863)计划资助项目(2007AA11Z250)

摘　　要：为了研究一类多输入多输出强非线性系统的自适应跟踪问题,采用RBF神经网络逼近模型不确定性,外界干扰和建模误差采用非线性阻尼项进行补偿,并将动态面控制与Nussbaum增益技术结合,提出了一种鲁棒自适应神经网络跟踪控制算法.该算法不仅能够解决系统中控制方向完全未知问题和可能存在的控制器奇异值问题,而且能够避免传统后推方法的计算膨胀问题,从而大大降低了控制器的复杂性,使之易于工程实现.同时,该算法保证了闭环系统的稳定性,并具有良好的鲁棒性.仿真结果验证了控制器的有效性.A problem in adaptive tracking control was considered for a class of highly nonlinear multi-input multi- output （MIMO） systems with both unknown system nonlinearities and unknown virtual control gain nonlinearities. By employing a radial basis function （RBF） neural network （NN） to approximate uncertain functions, and a non- linear damping item to compensate for both external disturbances and modeling errors, a robust adaptive neural net- work control algorithm was developed based on dynamic surface control （DSC） and the Nussbaum gain approach. The proposed algorithm not only both solves problems of unknown control direction and possible controller singulari- ty, but also solves the problem of ＂explosion of complexity＂ in the conventional backstepping method, reducing the computational load of the algorithm and making it convenient to implement in applications. In addition, the algo- rithm has good robustness, guaranteeing stability in a closed-loop system. Simulation results validated its effective- ness and performance.

关 键 词：多输入多输出系统 神经网络 动态面控制 NUSSBAUM增益 

分 类 号：TP273.2[自动化与计算机技术—检测技术与自动化装置]