Force field manipulation for an ethernet-case assembly task

Robot manipulation for delicate industrial tasks has been a challenge and playing an important role in developing intelligent industrial robots. In this paper, we present force-field manipulation for an ethernet-case assembly task. Since the assembly trajectory is delicate, programming the trajectory is tedious and difficult by a robot teach pedant. To plan the assembly trajectory easily, the idea is to break the problem into two steps: first, the force field and reference trajectory of the task are built using human demonstrated trajectories collected by a haptic device. Then using the force field, the reference trajectory is re-planned until the assembly task is finished. Thus, the proposed method presents the task model represented by the force field and the reference assembly trajectory obtained by Gaussian Mixture Regression (GMR) to achieve the assembly task easily. In our experiment, we provide an extensive validation of the proposed approach which shows good results with a less time to finish the assembly task by a robot.