Development of An Impedance-based Human-robot Hybrid Interaction System with RNN Force Estimator on Visual Tasks

This article develops a human-robot hybrid interaction system in order to involve both visual servoing task and human-robot interaction control tasks in one schema. A variable impedance strategy is proposed which is designed that the variable impedance parameter is adjusted by interaction behavior, so as to realize the target observation and sight wandering in visual tasks. The adaptive recurrent-neural-network-based(RNN-based) force estimator is adopted to estimate the human-robot interaction force in the sliding mode controller, in order to ensure the performance of the sliding mode controller of the robot. Experimental results verify the effectiveness of the proposed force-estimator-based control and adaptive variable impedance regulation in physical human-robot interaction tasks.