引用本文:李鹏飞,张银河,张蕾,王晓华,王文杰.考虑误差补偿的柔性关节机械臂命令滤波反步控制[J].控制理论与应用,2020,37(8):1693~1700.[点击复制]
LI Peng-fei,ZHANG Yin-he,ZHANG Lei,WANG Xiao-hua,WANG Wen-jie.Command-filtered backstepping control with error compensation for flexible joint manipulator[J].Control Theory and Technology,2020,37(8):1693~1700.[点击复制]
考虑误差补偿的柔性关节机械臂命令滤波反步控制
Command-filtered backstepping control with error compensation for flexible joint manipulator
摘要点击 2537  全文点击 906  投稿时间:2019-10-21  修订日期:2020-02-11
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2020.90872
  2020,37(8):1693-1700
中文关键词  柔性关节机械臂  命令滤波  反步控制  滤波误差补偿
英文关键词  flexible-joint manipulator  command filtering  backstepping control  filtering error compensation
基金项目  国家自然科学基金(51607133), 陕西省科技厅工业攻关项目(2016GY-136)资助, 重点产业创新链(群)-工 业领域(2019ZDLGY01-08).
作者单位邮编
李鹏飞 西安工程大学 电子信息学院 710048
张银河 西安工程大学 陕西省产业用纺织品协同创新中心 
张蕾* 西安工程大学 电子信息学院 710048
王晓华 西安工程大学 电子信息学院 
王文杰 西安工程大学 电子信息学院 
中文摘要
      针对柔性关节机械臂在运动过程中易产生振动, 使力和位置等控制精度难以保证等问题, 提出了一种改进 的命令滤波反步控制方法. 首先, 采用二阶命令滤波器得到反步控制方法中所需的虚拟控制函数及其导数, 避免了 对虚拟控制函数的多阶求导所导致的计算爆炸问题. 随后, 为了消除引入命令滤波器所产生的滤波误差, 设计了滤 波误差补偿机制, 并基于李雅普诺夫稳定性理论证明了该策略可保证闭环跟踪误差系统的稳定性. 最后, 针对单关 节柔性机械臂进行MATLAB仿真, 证明了该方法的有效性. 实验结果表明, 加入滤波误差补偿后, 跟踪精度提高了 12.7%, 滤波误差明显减少.
英文摘要
      Aiming at the problem that the flexible joint manipulator is subjected to vibration during the movement, which makes the control accuracy of force and position difficult to be guaranteed, a modified command filtering backstepping control strategy with error compensation is proposed for flexible joint manipulator. Firstly, the second-order command filter is adopted to derive the virtual control and its derivative. Thus the problem of “explosion of complexity” caused by recursive calculation of the partial derivatives of virtual control inputs is avoided. Secondly, in order to reduce the filtering error caused by adopting command filter, an error compensation mechanism is designed and the stability of the tracking error closed-loop system is verified based on Lyapunov stability theory. Finally, simulation results on a single-link flexible manipulator show that the tracking accuracy is increased by 12.7% and the filtering error is obviously reduced, which proves the effectiveness of the proposed method.