During Apollo lunar surface traverses (e.g., Apollo 14), astronauts could not locate the geological target due to spatial disorientation, which substantially compromised science productivity, and potentially mission safety and success. Navigation satellites analogous to GPS are not available to lunar surface explorers; the lack of atmosphere and familiar objects of known scale, the reduced gravity and the different reflection property of lunar surface make it difficult for astronauts to judge distances and slopes and to recognize landmarks. In 2004, a space exploration policy to return humans back to the moon was established by President Bush. Recently, manned missions to a NEA (Near Earth Asteroid) attract a broader interest within both the NASA and planetary research community. Spatial disorientation is a problem that threatens the productivity and safety of EVA traverses on the Moon or the NEA in the future missions.
This dissertation aims at developing techniques enabling a precise, robust, and extensible astronaut navigation system on lunar surface for future landed mission to help the astronauts to overcome the spatial disorientation. The navigation system is based on an integrated orbital-ground sensor and data network that includes data from lunar orbiters and sensors mounted on the astronauts' suits. Lunar orbital sensors include the Lunar Reconnaissance Orbiter Camera (LROC) and the Lunar Orbiter Laser Altimeter (LOLA) systems carried onboard the Lunar Reconnaissance Orbiter (LRO). The LRO data is used to provide a global DEM (Digital Elevation Model) of very high resolution (up to 1 m). Suit-mounted sensors include a boot-mounted IMU (Inertial Measurement Unit), a sky camera on helmet and a pair of stereo cameras mounted on the chest of the astronaut as well as a display system and other components. During the lunar traverse, the IMU provides accurate distance estimation through ZUPT (Zero Velocity Update), and the stereo-vision sensors correct any drift in the azimuthal angle in the trajectory reconstructed by the IMU. An EKF (Extended Kalman Filter) is designed to incorporate the data from the IMU, the chest-mounted stereo-vision sensors, and the sky camera to obtain the high accuracy navigation solution. Promised results from field experiments in lunar like environments show that the navigation system can support a walkback scenario (providing astronauts with guidance when returning to the lander/module) of at least 6 km. This dissertation presents a method combining the star tracking technology and the IMU for positioning on the lunar surface. The experiment shows that at current configuration, the system is able to achieve an accuracy of 3 km on the lunar surface. The combination of the boot-mounted IMU and the sky camera for navigation is tested in experiments. The system reaches a disclosure error of 0.57% over a traverse of 360 m. The navigation system presented in this dissertation is capable of providing precise navigation information to enable lunar astronauts to safely navigate on the lunar surface, locate science targets, and return safely to the lander/module.