Impedance adjustment for learning human variable impedance skill

In this paper, a Gaussian Mixture Model (GMM) based skill transfer framework is discuss and a task phase segmentation method based on human operation intention is proposed. Based on this skill transfer framework, an impedance adjustment algorithm for phase switching is also put forward to improve the performance of autonomous execution of robot skills. The proposed algorithm can solve the problem of task failure caused by position error during automatic robot performance of low-impedance skills without affecting compliance of the low-impedance part of the variable impedance skill transfer.