柔性关节协作机器人自适应跃度的鲁棒跟踪控制  

Robust Tracking Control of Adaptive Jerk for Flexible-Joint Co-Robot

作　　者：舒鑫东 孔祥杰 陈思鲁 李荣 张驰 杨桂林 SHU Xindong;KONG Xiangjie;CHEN Silu;LI Rong;ZHANG Chi;YANG Guilin(Zhejiang Key Laboratory of Robotics and Intelligent Manufacturing Equipment Technology,Ningbo Institute of Materials Technology and Engineering,Chinese Academy of Sciences,Ningbo 315201,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院宁波材料技术与工程研究所,浙江省机器人与智能制造装备技术重点实验室,宁波315201 [2]中国科学院大学,北京100049

出　　处：《组合机床与自动化加工技术》2025年第3期114-118,123,共6页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金项目(U20A20282,U1813223);中国科学院机器人与智能制造创新研究院自主项目(C2021001);浙江省重点研发计划项目(2021C0169);宁波市重大科技任务攻关项目(2022Z038)。

摘　　要：介绍了由柔性关节构成的协作机器人平台,并基于降阶刚体动力学模型分析其系统误差动力学模型。针对协作机器人在运动控制过程中容易受到其关节柔性模态的干扰问题,开展基于组合误差和径向基神经网络模型的自适应前馈控制器设计,并利用自适应跃度设计了鲁棒积分控制,形成了“自适应前馈+反馈”的二元控制算法,进而基于李雅普诺夫稳定性理论证明了闭环系统的跟踪误差能够渐近收敛。通过实际的机器人跟踪轨迹实验验证了该控制器的跟踪性能。In this paper,a collaborative robot platform composed of flexible joints is introduced,and its systematic error dynamics model is analyzed based on the reduced-order rigid body dynamics model.To solve the issue that co-robots are easily disturbed by their flexible modes in motion control,an adaptive feedforward control is designed based on the combined error and radial basis neural network model,and a robust integral control is designed by using the adaptive jerk to form a 2-DoFs control algorithm,in terms of"adaptive feedforward+feedback",and the asymptotic convergence of the tracking error of the closed-loop system is proved based on the Lyapunov stability theory.Experiment is conducted to validate performance of the proposed controller.

关 键 词：柔性关节 协作机器人 轨迹跟踪 自适应鲁棒控制 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]