Stereo vision guided control of a Stewart platform

This paper presents a new implementation of closed-loop pose control for a Stewart platform. We use stereo vision to measure the pose of the Stewart platform in real time. Since only three feature points are necessary for pose measurement, the computational requirements for image processing and calculation are small. The measured pose provides feedback to the inverse kinematics of the Stewart platform so that the pose error can be compensated. Experimental results are provided to show the ability of the method to control the pose.