A Method Based on Wearable Devices for Controlling Teaching of Robots for Human-robot Collaboration

A method based on wearable devices is proposed for controlling teaching of robots for human-robot collaboration. Positions at the end of arms and the angles at which each joint rotates are determined by calculating the Euler angles on different parts of human arms where the sensors are in the process of teaching. After these discrete data are coded and generally output by the Gaussian Mixture Model, continuous motion trajectories at the end of arms and corresponding joint angles are obtained. By solving the inverse kinematics equation of the robot at the end, a set of inverse solutions are obtained as a reference. The rest of the inverse solutions are not shown in this paper. At last, experiments demonstrated that the system make the robots' attitudes in the course of their motion closest to those of human arms during movements, in order to represent the motion process of human arms. The smooth and stable motion verifies that instructional personnel neither required teaching box nor pulled robotic arms during teaching. Indeed, they can teach robots about HRC simply through normal manual operations.