[1]彭金柱,卞英楠,周树亮.基于DRNN网络的轮式机器人鲁棒 Hi infinity 控制[J].郑州大学学报(工学版),2018,39(04):64-69.[doi:1013705/j.issn.1671-68332018.01.016]
　Peng Jinzhu,Bian Yingnan,Zhou Shuliang.A Stable NRobust Control for Wheeled Robotic System Based on DRNN Network and H∞ Methods[J].Journal of Zhengzhou University (Engineering Science),2018,39(04):64-69.[doi:1013705/j.issn.1671-68332018.01.016]
点击复制

基于DRNN网络的轮式机器人鲁棒 Hi infinity 控制()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
39卷
期数:
2018年04期
页码:
64-69
栏目:
出版日期:
2018-07-22

文章信息/Info

Title:
A Stable NRobust Control for Wheeled Robotic System Based on DRNN Network and H∞ Methods
作者:
彭金柱; 卞英楠; 周树亮
郑州大学 电气工程学院,河南 郑州,450001
Author(s):
Peng Jinzhu;  Bian Yingnan;  Zhou Shuliang
School of Electrical Engineering, Zhengzhou University, Zhengzhou, Henan 450001
关键词:
Keywords:
Diagonal Recurrent Neural Network;  Control Moment Method;  H∞ Control;  Uncertainty;  Lyapunov Stability
DOI:
1013705/j.issn.1671-68332018.01.016
文献标志码:
A
摘要:
针对非完整约束的两轮移动机器人系统中存在建模误差及外扰的情形，提出了一种结合对角递归神经网络和非线性H∞方法的控制策略。利用对角递归神经网络逼近建模不确定的非线性项， H∞控制则用来实现期望的鲁棒跟踪性能。基于Lyapunov稳定性理论，整个系统跟踪误差闭环有界。此外，在外界干扰仅仅积分有界的情形下，系统仍能满足具体的鲁棒跟踪性能。最后，对于相同外扰及不确定性下的移动机器人，将其与控制力矩法进行对比，仿真结果表明所提方法是有效的。
Abstract:
In order to solve the modeling errors and external disturbances in the system,a stable robust hybrid tracking control scheme for wheeled robotic system with nonholonomic constraint is proposed. The hybrid control scheme combined DRNN neural network with nonlinear H∞ method. DRNN neural network was employed to approximate the system uncertainty terms,and H∞ control was utilized to achieve a desired robust tracking performance. Based on Lyapunov stability theory, the tracking errors of the closed-loop system were bounded. In addition, a specified H∞ tracking performance was obtained by the proposed robust hybrid control even though the disturbances were merely integral bounded. Finally, the proposed control scheme was compaed with the computed torque control for nonholonomic mobile robotic system under the uncertainties and external disturbances. Simulation experiments showed that the proposed control strategy was effective.
更新日期/Last Update: 2018-07-26