| Formation flying of multiple spacecraft is an enabling technology for many future space missions. By separating a monolithic structure into several spacecraft that collectively act as a single structure, virtual spacecraft of unlimited size can be flown. This will enable new scientific missions, based on distributed yet coordinated measurements. Examples include stellar interferometry missions and gravity field mapping.; One of the fundamental requirements of formation flying is precise knowledge of the relative position and attitude between the vehicles. The Global Positioning System (GPS) has been shown to be an accurate position and attitude sensor both on earth and in space. The use of GPS in space is limited to orbits with adequate visibility to the NAVSTAR GPS constellation, typically LEO. For higher elevation orbits, such as MEO, GEO, and highly elliptical orbits, visibility of the GPS constellation can easily fall to two or even zero satellites. In such conditions, insufficient measurements are available for a formation of vehicles to solve for all relative position and attitude states. However, by augmenting the available NAVSTAR satellite signals with GPS pseudolites placed onboard the vehicles in the formation, sufficient signals are available to solve for relative position and attitude between all vehicles. The onboard pseudolite system can even operate during times of total occlusion of the satellites. Thus, an onboard pseudolite system can be used in environments where the NAVSTAR satellites are known to be totally unavailable, such as deep space.; This dissertation presents the development of a relative positioning sensor for a fleet of vehicles, combining information from onboard pseudolites with information from the GPS satellites. Relative positioning is conducted on a fleet of three prototype spacecraft, in a laboratory environment as proof of concept. System robustness is improved through a novel carrier phase reinitialization approach. |
