Sensorless Force Control Algorithm based on Momentum Observer Technique

摘要

This paper presents a sensorless force control algorithm based on external force estimation technique for robotic manipulator systems. Typically, six-axis force torque or joint torque sensors are employed to measure external force induced by contact between the robot and the environment. However, attachment of such sensors to robotic manipulator systems is too costly and requires additional space in the body of a robotic manipulator. Our choice for solving this problem is to employ the momentum observer, which is one of the well-known external force estimation techniques for estimating external force, and the estimated external force is used as feedback for force control without using costly force torque or joint torque sensors. Stability of a closed-loop system where contact is in place and force control performance in such conditions have been analyzed based on the singular perturbation theory. In addition, a design guideline for applying the momentum observer approach we proposed has been put together according to the Lyapunov analysis method. We validate performance of the proposed algorithm using experiments with a six-axis industrial robotic manipulator (Robostar RA004). The experimental results show that the end-effector of the manipulator exerts desired force to contact object with error margin of 3.6%.