Robotic vision technology and algorithms for space applications

Abstract

The Space Station is planned as a multipurpose facility in which missions of long duration can be conducted. An important technology/technique area foreseen to increase productivity and enhance astronaut safety is the automation and robotics (A&R). In addition, increases in operational capability and flexibility are also envisioned through A&R. Initial thrusts of A&R in Space Station are in the areas of assembly, maintenance and repair, and routine inspections. One key element of the A&R systems is sensing for perception. Its primary goal is to provide information regarding the identification and position of the object in its environment relative to the system effector. This function involves isolation, description, identification, location, and data transmission. The type, volume, and precision/accuracy of the data needed depend on the nature of the task to be accomplished. Sensor complements may include both active and passive microwave and optical types with multifunction capability. In this paper, the vision data requirements are discussed from the standpoint of various applications. A review of the advanced systems technology is provided. The progress in the areas of sensor fusion and algorithm development is summarized. Future challenges in both sensor technology and algorithm development are elaborated.