Model-free distributed coordination control for dual-arm reconfigurable manipulators with handling task constraints
2023 35TH CHINESE CONTROL AND DECISION CONFERENCE, CCDC(2023)
摘要
In this paper, a distributed coordination control of dual-arm reconfigurable manipulators based on neural networks learning is presented for handling tasks. Based on Newton-Euler algorithm, the dynamic models of the manipulators and grasped object are established through kinematic analyze, respectively. According to the load distribution method, the motion-induced force is effectively distributed to each arm of the manipulator, and then the dynamics of the single reconfigurable manipulator is obtained. An improved sliding mode function is designed to reflect the error of the joint trajectory and internal force. The radial basis function neural network (RBFNN) is utilized to learn the uncertain dynamics of reconfigurable manipulator. Then the model-free distributed coordination controller is obtained. The asymptotic stability of dual-arm reconfigurable manipulators is proved through Lyapunov stability theory. Finally, the validity of the model-free distributed coordination controller is verified by simulations.
更多查看译文
关键词
reconfigurable manipulators,neural network,dual-arm coordination control,sliding mode control
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要