Development of a physiological signals enhanced teleoperation strategy

In this paper, we have developed a teleoperation method with novel features, such that the human operator could manipulate the telerobot using both physical and physiological means. Physiological studies have shown that our central neural system (CNS) is able to adapt muscle force, stiffness and damping to perform different tasks under various environments. However, there has been very little research in the robotics communities to incorporate human motor skills of muscle force, stiffness and damping adaptation into teleoperation. As a matter of fact, muscle activities regulated by CNS can be represented by surface electromyography (sEMG) measured by electrodes attached on the skin of human limbs. Therefore, sEMG based teleimpedance control has been recently developed to transfer the variable stiffness from human operator to a robot for flexible manipulation [1], [2], [3]. In this paper, we have developed a personalised variable gain control, whereas the control gain is set according to muscle activation extracted from sEMG signals, and force feedback is employed such that human operator is able to sense the circumstance in a haptic manner and to adapt muscle contraction subconsciously as if they are directly interacting with the environment. The proposed method enables an operator to adjust tracking speed and stiffness of the telerobot easily on the fly, and thus greatly improves a user´s manipulation ability for delicate tasks, especially in the presence of motion amplification and varying payload on the telerobot. The effectiveness of the proposed approach has been tested and confirmed by comparative experiments, in which a 7-DOF (degrees of freedom) robotic arm of Baxter® robot is employed.