Robot Control Using Disparate Multiple Sensors

Abstract

A simple and efficient control algorithm that combines several sensors in order to realize the positioning task of a robot end effector is presented in this paper. The multiple-sensor control was designed as a part of the task function approach. A particular choice of the task function allows the researcher to simplify the design of the control law and the stability analysis. This global controller is based on the weighted sum of individual task functions. The approach was applied to the control with vision and force sensors. In spite of its simplicity, this approach provides satisfactory experimental results. Improvements in the positioning of cumbersome objects were obtained using cameras observing different parts of a scene. Moreover, peg-in-hole insertion experiments involving large initial errors were performed using a seven-axis robot manipulator without any computation of the peg trajectory by combining vision and force sensors.