记笔记

Kinematic Control with Gravity Optimization for Redundant Manipulators under Model Uncertainties

Xiaojing LvZhihao XuYiqun GaoXuefeng ZhouXiaoyan LaoYing Chen

School of Aircraft Maintenance Engineering, Guangzhou Civil Aviation CollegeGuangdong Institute of Intelligent ManufacturingSouth China Robotics Innovation Research Institute

摘要：In this paper, trajectory tracking for redundant manipulators is considered. In real implementations, kinematic uncertainties of the robot are inevitable, making it difficult to achieve high-precision tracking control. At the same time,how to ensure the boundedness of the physical states remains challenging. Aiming at reducing energy consumption, with the aid of recurrent neural networks, we propose a RNN based controller, including a Jacobian adaption method. This control strategy is capable of handling system constrains as well as kinematic uncertainties. Stability of the closed-loop system is analyzed. Simulations shows the effectiveness of the proposed method.

关键词：

Gravity Optimization; Redundant Manipulator; Recurrent Neural Networks;

基金：

supported by Special Funds for Scientific and Technological Innovation and Cultivation of College Students in Guangdong Province（Grant No.pdjhb0693）； Guangzhou Science Research Plan-Major Project（Grant No.201804020095）； Postdoctoral working Station Scientific Research Foundation of the Guangdong Academy of Sciences（Grant No.2018GDASCX-1006）； Guangdong Science and Technology Plan Project（Grant No.20158010917001）； Guangdong Province Applied Science and Technology Research Project（Grant No.2015B090922010）； Guangzhou Science Research Project（Grant No.201707010457）； Science and Technology Planning Project of Guangdong Province,China（Grant No.2017A010103013）； Science and Technology Planning Project of Guangzhou（Grant N0.201803010106）；

会议名称：

第31届中国控制与决策会议

会议时间：

2019-06-03

会议地点：

中国江西南昌