High-performance Load Torque Compensation of Industrial Robot using Kalman-filter-based Instantaneous State Observer

Abstract

Robust motion control against dynamic torque is required for rapid and precise motion control of industrial robots. In this regard, a disturbance observer (DOB) is widely used to achieve robust motion control. In general, it is difficult to achieve robust motion control against a step load torque because the DOB exhibits an estimation delay. To overcome this problem, this paper proposes a new method involving the use of a Kalman-filter-based instantaneous state observer for load torque compensation. The proposed method achieves the instantaneous load torque estimation of a two-inertia system using a load-side acceleration sensor. Torque compensation based on instantaneous torque estimation is highly robust against the insertion of a step load torque. The effectiveness of the proposed method is confirmed by performing both a numerical simulation and experiments using an industrial robot arm.