Direct dynamic control of a robot using an end-point mounted camera and Kalman filter position estimation

Abstract

The development of a control strategy for direct dynamic end-point control of a manipulator based on position estimates obtained from endpoint mounted sensors is presented. The feasibility of using an extended Kalman filter algorithm for relative position estimates based on task object feature locations extracted from images obtained from an end-point mounted camera has been studied. An experimental system has been constructed and the position estimation and control strategy implemented to test the performance of the design. Tests have demonstrated the successful operation of the control strategy developed and have also demonstrated the feasibility of using an extended Kalman filter for relative position estimates at a sample rate of 61 Hz.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>