Adaptive unified Impedance and Admittance control using online environment estimation

Abstract

Impedance and Admittance Control are two effective control methods when dealing with dynamic interactions between a robotic system and its environment. Impedance control performances excel when in contact with stiff environments. Conversely, admittance control is better suited when interacting with a soft environment. The Unified Impedance and Admittance control method aims to provide excellent control over a large array of possible environmental stiffness by constantly interpolating between the two. Because the switching rate governing the interpolation and the environmental parameters greatly impacts the performances of the controller, an adaptive strategy is developed in this paper. With the use of online environmental estimation, the practicality of the control is enhanced and less time is spent in offline training. The advantage of this approach is demonstrated through simulations of a 2 degrees of freedom manipulator.