基于接触状态识别的机器人操作充电枪寻孔策略  

Hole-finding strategy for charging gun operated by robot based on contact state recognition

作　　者：徐建明[1] 胡松达 董建伟 俞立[1] XU Jian-ming;HU Song-da;DONG Jian-wei;YU Li(College of Information Engineering,Zhejiang University of Technology,Hangzhou 310023,China)

机构地区：[1]浙江工业大学信息工程学院,杭州310023

出　　处：《控制与决策》2022年第7期1794-1802,共9页Control and Decision

基　　金：国家自然科学基金-浙江省自然科学基金联合基金两化融合项目(U1709213);国家自然科学基金面上项目(61374103).

摘　　要：针对机器人操作充电枪寻孔过程,研究一种基于接触状态识别的寻孔策略.基于以接触力为基本特征点的支持向量机(SVM)的线性分割算法,提出一种寻孔接触状态识别方法.针对端面接触状态、两点接触状态和三点接触状态,分别给出充电枪插头坐标系下的向心力方向、侧向力方向和扭矩方向的寻孔速度轨迹规划方法,并提出一种跟踪Z方向寻孔装配力与XY平面寻孔速度轨迹的力位混合控制方法.最后基于6自由度协作机器人、充电枪插头及插座的实验平台,通过寻孔操作实验验证所提出方法的正确性和有效性.Based on the recognition of the contact state,a hole-finding strategy is studied for the operation of the electric vehicle charging gun(EVCG).A method for identifying the contact state in the EVCG hole-finding process is proposed based on the linear segmentation algorithm of the support vector machine(SVM)with contact force as the basic feature.For the end-face contact state,two-point contact state,and three-point contact state,the hole-finding velocity trajectory planning methods of the centripetal force direction,the lateral force direction and the torque direction are respectively given under the coordinate system of the charging gun plug.Furthermore,a hybrid force/position control method is proposed for tracking the Z-direction hole-finding assembly force and the XY plane hole-finding velocity trajectory.Finally,based on the constructed 6-DOF collaborative robot and EVCG and its holder as the experimental platform,the accuracy and effectiveness of the proposed strategy are verified through experiments.

关 键 词：接触状态识别 寻孔策略 机器人 充电枪 支持向量机 轨迹规划 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]