Neuro-hierarchical sliding mode control for a class of under-actuated systems
Neuro-hierarchical sliding mode control for a class of under-actuated systems
复制标题
DOI：
10.1504/ijmic.2011.041779
复制
发表时间：
2011-08
期刊：
影响因子：
--
通讯作者：
D. Qian;Xiangjie Liu;J. Yi;Chengdong Li
中科院分区：
文献类型：
--
作者：
D. Qian;Xiangjie Liu;J. Yi;Chengdong Li
针对一类具有稳定平衡点的欠驱动系统，提出了一种神经递阶滑模控制器。这种控制器结合了神经网络的概念和分层滑模控制的方法。首先，针对该类系统设计了递阶滑模控制律。将系统划分为若干个子系统，定义了各子系统的滑动面。然后，选择其中一个子系统的滑动面作为第一层滑动面。第一层滑动面与另一个子系统的滑动面构成第二层滑动面。这个过程一直持续到所有子系统的滑动面都包括在内。控制律是由李雅普诺夫定理推导出来的。针对系统中的未知因素和不确定性，设计神经网络逼近递阶滑模控制律。从理论上证明了整个滑模面的渐近稳定性和网络权值的收敛性。仿真和物理实验结果验证了该控制器的有效性和鲁棒性。
A neuro-hierarchical sliding mode controller is presented for a class of under-actuated systems with a stable equilibrium point. Such controller is combined with the concept of neural networks and the methodology of hierarchical sliding mode control. At first, the hierarchical sliding mode control law is designed for the class as follows. The system is divided into several subsystems and the sliding surface of every subsystem is defined. Then, the sliding surface of one subsystem is selected as the first layer sliding surface. The first layer sliding surface is then to construct the second layer sliding surface with the sliding surface of another subsystem. This process continues until the sliding surfaces of all the subsystems are included. The control law is derived from Lyapunov theorem. By aiming at unknown factors and uncertainties, the neural networks are designed to approximate the terms of the hierarchical sliding mode control law. The asymptotic stability of the entire sliding surfaces and the convergence of the network weights are proven theoretically. Simulation and physical experiment results show the controller’s validity and robustness.