Computationally Efficient Low-Infrastructure Navigation Solutions Using Target Localization Algorithms
Bong-Jun Yang, E. Daniel Wang, and Joseph Loof, Georgia Tech Research Institute
Location: Ballroom B
Date/Time: Wednesday, Jun. 5, 9:15 a.m.
As future lunar missions to the moon consider exploring and utilizing moon resources, navigation on the moon surface is becoming an important problem for autonomous rover missions. In the literature, a low infrastructure, radiometric navigation system, which uses a minimum number of lunar-orbit satellites with an aid from a reference station, has been proposed for obtaining the position and the velocity of planetary surface users. In solving the position and the velocity of the user, whether the problem is solved in batch processing or in sequential processing, such as the extended Kalman filter, the previous approaches have treated pseudoranges and Doppler frequency measurements as the constraints that must be satisfied, and as a result, numerical procedures for obtaining solutions have been sensitive to initial guesses.
In this study, we show that the low-infrastructure navigation problem can be recast into that of the unknown target localization problem and therefore can be solved by computationally efficient algorithms that do not require initial guesses. The resulting position and the velocity can be fused with onboard sensors, such as Inertial Measurement Units (IMUs), leading to a loosely-coupled Inertial Navigation System/Lunar Navigation Satellite System (INS/LNSS) integration architecture on the moon.