Lessons Learned from Operating C/A-Code COTS GPS Receivers on Low-Earth Orbiting Satellites for Navigation

Abstract

Since June of 1993, an experimental GPS receiver system has been orbiting the earth aboard a small, low-altitude, polar-orbiting satellite called RADCAL. The purpose of the experiment was to prove the concept of using GPS for satellite navigation. If successful, the system would also provide a backup to the satellite's primary navigation beacon. The goal: provide position and velocity data to an accuracy of three to five meters, and provide attitude data to within a degree. The configuration of the RADCAL GPS experiment precluded realtime feedback loops for navigation; the data was stored and downloaded after a designated collection period. On the ground, a lengthy process was used to yield the position and attitude data days after the collection event. The GPS receivers and ground equipment were configured in several modes; they ultimately yielded a position accuracy of five meters, and attitude of two degrees. This was the original goal, and the experiment was considered successful. However, one of the receivers failed in November 1993, and the other failed in January 1995. The GPS receivers were commercially available and not spaceflight proven; they were suspected of being vulnerable to single-event upsets and latchups. This turned out to be the cause of the failure of both receivers. The interface between the GPS receivers and RADCAL's other subsystems proved to be the area which could not tolerate corrupt data. The single-event latchups problems would ultimately lead to the failure of the receivers. These difficulties, as well as other lesser obstacles, provide a host of lessons learned for future satellite navigation systems.