The Khepera CMU Camera Turret is based on the famous CMUcam2 module from Carnegie Mellon University. It provides embedded image processing functions for the Khepera robot and allows partial download of the image to the robot for user specific processing. The turret is however not designed to allow the complete image download to a remote PC for image processing on a computer. Remote image processing should be performed using a K2D vision turret. With the CMU Camera Turret, vision based experiments with multiple robots are possible, as each robot can be fully independant without cables and without a remote operation program. All image preprocessing functions are performed on the turret processor to keep the robot main processor free for other tasks.