检测机械手轨迹突变点识别方法  

Recognition Method for Testing Manipulator Trajectory Change-point

作　　者：赵新玉[1] 温欣 段晓敏[2] ZHAO Xinyu;WEN Xin;DUAN Xiaomin(School of Material Science and Engineering,Dalian Jiaotong University,Dalian 116028;School of Science,Dalian Jiaotong University,Dalian 116028)

机构地区：[1]大连交通大学材料科学与工程学院,大连116028 [2]大连交通大学理学院,大连116028

出　　处：《机械工程学报》2020年第18期15-21,共7页Journal of Mechanical Engineering

基　　金：国家重点研发计划(2016YFF0203000);辽宁省教育厅(JDL2019013)资助项目。

摘　　要：结合六轴机械手和超声无损检测技术实现自动检测,是检测自动化领域重要发展方向。超声检测要遵循运动平稳原则,检测轨迹中姿态变化较大会导致机械手速度突变,造成超声耦合不佳影响检测结果。针对上述问题,提出一种基于特殊欧几里得群信息几何结构的轨迹突变点识别方法,以轨迹当前点和前后若干点的黎曼均值差异为突变点检测依据,来预先识别出轨迹突变点,进而在这些突变点附近进行速度平滑处理。利用六轴机械手运动学反解方法验证了基于黎曼均值方法识别轨迹突变点的有效性。进一步,开展了机械手平台运动实验,识别突变点并降低邻近三点速度,使机械手各轴检测运动更加平稳。A six-axis manipulator is used in ultrasonic nondestructive testing technology to realize automatic testing,which is an important development direction in the field of testing automation.In ultrasonic testing,the principle of smooth motion is followed.Sudden change in the speed of the manipulator is occurred due to large change in posture in the detection trajectory,and testing results is affected by poor coupling.To solve the above problems,a change-point detection method based on the matrix information geometry of special Euclidean group is proposed.Once a change-point occurs,the distance between the current point and the Riemannian mean of its neighbor points should be a local maximum.According to this principle,the change-points of the trajectory are identified in advance,and then the velocity is smoothed near these change-points.The effectiveness of the Riemannian mean method is verified by the inverse kinematics solution method of six-axis manipulator.Furthermore,the manipulator platform movement experiment is carried out to identify the change-points and reduce its speed near the change-points,so that the detection movement of each axis of the manipulator is more stable.

关 键 词：六轴机械手 突变点识别 信息几何 黎曼均值 

分 类 号：TB553[理学—物理]