Collaboration robotic compliance grasping based on implicit human - computer interaction

Abstract

An implicit human-computer interaction technique, which combines surface Electromyography and inertial information, is proposed to realize the compliant grasping of cooperative robots. First, the mapping between human hand space and the end of the cooperative robot is established by IMU. Gesture recognition is carried out through surface EMG to determine the robot posture, as to realize operator's remote operation on robot. In addition, in order to solve the problems of collision, gesture recognition and block holding force, a compliant control algorithm based on force feedback is proposed. Besides the muscle force model is applied to predict the opponent force, so that the autonomous compliance is achieved.