Video Guidance Sensor for automated capture

Abstract

The paper describes a video-based sensor, the Video Guidance Sensor, which has been developed for use in automated docking systems or for automated capture with robotic systems. The sensor itself consists of several basic components: IR diode illuminators, a solid-state camera, a frame grabber, and a microprocessor. The target consists of three circular retroreflectors evenly spaced in a line with the middle reflector mounted on a short pole. Each retroreflector has a narrow bandpass optical filter in front of it. From this sensor and target configuration, the positions and attitudes of the target relative to the sensor can be determined. The position information can be differenced to with respect to time in order to obtain the relative positional and angular velocities. That information is sufficient to perform automated docking, berthing, or capture. The paper gives details of the sensor's make-up, including accuracies, noise characteristics, and operating ranges.