Adaptive Impedance Control for Robot Based on Estimation of Environmental Parameters

Abstract

In order to realize precise contact tasks with an unknown environment, robotic force controllers have to adapt themselves to the unknown environment. Some impedance controllers are designed for several representative environmental parameters, A BP neural network is proposed to determine the one-to-one mapping relations between the environmental parameters and the impedance parameters. However, it is difficult to accurately know the environmental parameters in the case of a changing environment, RLS is proposed to estimate environmental parameters, then determine the impedance coefficients to control the robot. Simulations prove that the controller designed is feasible and effective.