Velocity Decomposition Based Planning Algorithm For Grasping Moving Object

An online planning method is proposed for an industrial manipulator to grasp a moving object whose motion is not long-term predictable. Due to the limited time when the moving object stays within the limited workspace of the fixed manipulator, the manipulator has to grasp the object before it leaves the workspace. The planning algorithm brings the end effector of the manipulator to the vicinity of the object quickly and makes it match the pose of the object at first, then grasps the object. In terms of the states of the object and the end effector, the velocity of the end effector is decomposed to three directions. The accelerations of each direction are planned to make sure that the end effector can achieve stably tracking of the moving object in a short time. According to these accelerations, the velocity of the end effector and the joints velocities are obtained through pseudo inverse of the Jacobian matrix of the manipulator. Several simulation examples show that the proposed method can finish the grasping tasks faster than conventional methods.