Motion Planning Combines Human Motion Prediction for Human-Robot Cooperation

Human safety is the most important re-quirement for human robot cooperation (HRC). Al-though all kinds of robots have come to our side, human safety is still an ongoing research problem for human-robot coexisting scenarios. Different from other motion planning algorithms that only consider dynamic obstacle avoidance in ordinary environment, this paper combines the user's head pose in a human-robot coexisting environment. First of all, based on the correlation between head rotation and wrist (dynamic obstacle) motion, this paper takes the user's head pose, wrist position and wrist speed as the input of prediction model based on LSTM(Long Short-Term Memory), and then predicts the wrist position. Secondly, we propose a motion planning method combined with predicted wrist position, which can ensure that the distance between the robot and the user's wrist is in a safe range and the path length of manipulator end-effector is shorter. Finally, we demonstrate our idea on a 7-DOF TIAGo robot for human-robot cooperation task. Results show that the robot becomes sense motive, and responds to human motion quickly and efficiently.