A Robot Humanoid Control Framework Through Human Arm Active Endpoint Stiffness and Direction Adaptive Compensation

Abstract

In the process of human-robot interaction (HRI), the controleffect cannot meet the needs of HRI tasks if a control strategy is developed solely from the robot's point of view. It's necessary to take into account the characteristics of the human operator. In this letter, a novel HRI framework is developed to realise robot humanoid control by transferring the human arm stiffness and human hand motion characteristics to the robot. Firstly, adaptive impedance control is used to adjust the robot's stiffness through human arm active endpoint stiffness, which is determined by the geometric information and muscle activation of the human arm. In addition, a direction adaptive compensation method is also proposed by introducing a direction compensating quantity to regulate the robot's trajectory for minimising the effect on the robot of changes in the motion direction of the human hand. The proposed method can not only stabilize the process of HRI but also decrease the interaction force to reduce the human operator's workload. Comparative experiments demonstrate the enhanced performance of the proposed method.