Human-Like Robot Impedance Regulation Skill Learning from Human-Human Demonstrations

--Humans are experts in collaborating with others physically by regulating compliance behaviors based on the perception of their partners' states and the task requirements. Enabling robots to develop proficiency in human collaboration skills can facilitate more efficient human-robot collaboration (HRC). This paper introduces an innovative impedance regulation skill learning framework for achieving HRC in multiple physical collaborative tasks. The framework is designed to adjust the robot compliance to the human partner's states while adhering to reference trajectories provided by human-human demonstrations. Specifically, electromyography (EMG) signals from human muscles are collected and analyzed to extract limb impedance, representing compliance behaviors during demonstrations. Human endpoint motions are captured and represented using a probabilistic learning method to create reference trajectories and corresponding impedance profiles. Meanwhile, an LSTM-based module is implemented to develop task-oriented impedance regulation policies by mapping the muscle synergistic contributions between two demonstrators. Finally, we propose a whole-body impedance controller for a human-like robot, coordinating joint outputs to achieve the desired impedance and reference trajectory during task execution. Experimental validation was conducted through a collaborative transportation task and two interactive T ai Chi pushing hands tasks, demonstrating superior performance from the perspective of interactive forces compared to a constant impedance control method. OLLABORA TIVE robots (cobots) have emerged as a solution for more efficient human-robot collaboration (HRC) in both industrial and domestic scenarios. Co-manipulation outperforms fully robotic manipulation by offering enhanced flexibility and effectiveness while surpasses fully human manipulation by reducing labor costs, maintaining concentration, and minimizing errors due to fatigue [1]. This work was supported in part by the Research Grants Council of the Hong Kong SAR under Grant 24209021, 14222722, 14211723 and C7100-22GF and in part by InnoHK of the Government of Hong Kong via the Hong Kong Centre for Logistics Robotics. Darwin Caldwell is with the Department of Advanced Robotics, Istituto Italiano di Tecnologia, 16163 Genoa, Italy (e-mail: darwin.caldwell@iit.it).