机械臂任务空间鲁棒Terminal滑模控制  

Task Space Robust Terminal Sliding Mode Control for Robotic Manipulators

作　　者：赵东亚 赵永瑞[2] 张兰 

机构地区：[1]中国石油大学(华东)化学工程学院,青岛266555 [2]中国石油大学(华东)机电工程学院,青岛266555

出　　处：《机械工程学报》2012年第5期1-9,共9页Journal of Mechanical Engineering

基　　金：国家自然科学基金(61004080);山东省自然科学基金(ZR2011FM003);中央高校基本科研业务费专项资金;科技人员服务企业行动计划(2009GJC60044);青岛市科技计划(08-1-3-36-jch);青岛市经济技术开发区重点科技发展计划(2011-2-52)资助项目

摘　　要：不确定性和外部干扰是制约机器人控制精度的主要因素,传统Terminal滑模控制方法未能考虑机械臂运动学不确定性。提出一种新的机械臂任务空间鲁棒连续非奇异Terminal滑模跟踪控制方法。该方法具有较强的鲁棒性,能够有效克服运动学不确定性、动力学不确定性和外部干扰,保证机械臂任务空间跟踪误差有限时间稳定,将现有机械臂Terminal滑模控制方法从关节空间扩展到任务空间。通过模型转换,同时考虑机械臂运动学不确定性和动力学模型不确定性。使用矢量和矩阵范数定义,给出系统不确定性的界定方法。运用Lyapunov再设计方法和有界稳定理论,分析系统稳定性,明确给出系统调节时间和跟踪误差剩余集的估计方法。仿真中,通过与现有机械臂关节空间Terminal滑模跟踪控制方法比较,验证了所提方法的有效性。Uncertainty and external disturbance are the main factors which degrade the robot control precision.Kinematic uncertainty is not considered by conventional terminal sliding mode control.A new robotic manipulator task space robust tracking control approach is studied with continuous non-singular terminal sliding mode.The proposed control approach has strong robustness and can deal with kinematic and dynamic uncertainty.It achieves finite-time stability for robot task space tracking control.The existing robot terminal sliding mode control is extended by the proposed approach form joint space to task space.Kinematic uncertainty and dynamic uncertainty are considered together in the controller design.By using norm definitions of vector and matrix,a measure approach is presented for system uncertainty.In light of Lyapunov redesign method and bounded stability theory,the corresponding stability analysis is presented and the estimated methods for settling time and tracking error residual set are also developed explicitly.In simulation,the proposed approach is compared with the existing robot joint space terminal sliding mode tracking control.These comparisons validate the effectiveness of the proposed approach.

关 键 词：机器人 鲁棒控制 控制系统 不确定系统 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]