Aiming at the problems of inaccurate workpiece identification and low grasping efficiency caused by mutual occlusion of workpiece during grasping by using vision system, a trajectory optimization method of manipulator based on vision was proposed. The image information of the workpiece was collected by the camera, and the “hand-eye” coordinate relationship was established by calibrating the manipulator coordinate system, the workpiece coordinate system and the camera coordinate system, and the workpiece was identified and positioned based on the point cloud data. An improved hybrid genetic-whale trajectory planning algorithm was proposed to control the workpiece grasping process. Three important joints of the manipulator were simulated and tested. The simulation results show that the improved hybrid genetic-whale algorithm has faster convergence speed and stronger searching ability, and the optimized grasping time is reduced by 2.3 s compared with the basic genetic algorithm. The experimental results show that the grasping success rate of the manipulator based on point cloud recognition reaches 93.75%, which greatly improves the grasping efficiency and verifies the effectiveness of the algorithm.
LI Dongmin, DU Hao, ZHAO Liuyang, CAO Luhu, WANG Tong, WANG Yu, MA Wenping, LUAN Hengxuan. Research on Trajectory Optimization of Manipulator Based on Vision[J]. Machine Tool & Hydraulics,2023,51(8):35-41