Neural network based robust hybrid control for robotic system: an H∞ approach
Neural network based robust hybrid control for robotic system: an H∞ approach
复制标题
DOI：
10.1007/s11071-010-9902-4
复制
发表时间：
2011-09
影响因子：
5.6
通讯作者：
Jinzhu Peng;Jie Wang;Yaonan Wang
中科院分区：
文献类型：
--
作者：
Jinzhu Peng;Jie Wang;Yaonan Wang
提出了一种新颖的机器人系统鲁棒混合跟踪控制。这种混合控制方案将计算扭矩控制（CTC）与神经网络、变结构控制（VSC）和非线性H∞控制方法相结合。假设机器人系统的标称系统完全已知，采用CTC方法对其进行控制。设计神经网络来逼近参数不确定性，使用VSC来消除逼近误差的影响，并采用H∞控制来实现期望的鲁棒跟踪性能。基于Lyapunov稳定性定理，可以保证闭环中的所有信号都是有界的，并且通过采用所提出的鲁棒混合控制来实现指定的H∞跟踪性能。通过对二连杆机器人的计算机仿真验证了该控制方案的有效性。
A novel robust hybrid tracking control for robotic system is proposed. This hybrid control scheme combines computed torque control (CTC) with neural network, variable structure control (VSC) and nonlinearH∞control methods. It is assumed that the nominal system of robotic system is completely known, which is controlled by using CTC method. Neural network is designed to approximate parameter uncertainties, VSC is used to eliminate the effect of approximation error, andH∞control is employed to achieve a desired robust tracking performance. Based on Lyapunov stability theorem, it can be guaranteed that all signals in closed loop are bounded and a specifiedH∞tracking performance is achieved by employing the proposed robust hybrid control. The validity of the control scheme is shown by computer simulation of a two-link robotic manipulator.