Nonlinear Disturbance Observer for External Force Estimation in a Cooperative Robot*

In human-robot cooperation, where the robot shares workspace with humans, safety of the operator becomes a major concern. To this aim, the robot is required to detect forces from the human operator and the environment, and react to them accordingly. Since force sensors can be very expensive, force estimation methods are proposed. In this paper, the goal is to estimate the external forces acting on the end-effector of the robot. These forces make torques at the joint space. To estimate the applied joint space external torques, a nonlinear disturbance observer is proposed. The estimated torque can be converted into task space force, using the Jacobian matrix. The suggested method is demonstrated on a WallMoBot, which is designed to help the operator to install heavy glass panels. Simulation results and preliminary experimental results are presented to show the validity of the proposed observer in estimating the applied joint space external torques of the WallMoBot.