Vision-Aided Inertial Navigation Using a Geomatics Approach

Fadi A. Bayoud

Abstract:	This paper proposes and tests a method for bridging GPS outages during short and long periods with a vision-based inertial navigation. This method is similar to the Simultaneous Localisation And Mapping (SLAM), which is the problem of mapping the environment and at the same time using this map to determine the location of the mapping device. After 20 years of investigation, SLAM is still an open problem in the robotics community that searches a global, stable and efficient solution. This study falls in these trials, with a difference that it originates from the geomatics engineering perception of navigation and mapping. This paper solves the SLAM problem by integrating Photogrammetry and Inertial Measurement Unit (IMU) in a Kalman Filter. Briefly, from existing known features (ground control points, GCP) photogrammetric resection provides the position and orientation of the cameras that are integrated, after applying appropriate systems transformations, with the IMU data to produce a filtered position, which by its turn is used in the intersection to map more surrounding features, that are used as GCPs in the next epoch. Our methodology is presented, differences with other solution are pointed out, and numerical tests are discussed.
Published in:	Proceedings of the 18th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2005) September 13 - 16, 2005 Long Beach Convention Center Long Beach, CA
Pages:	2485 - 2493
Cite this article:	Bayoud, Fadi A., "Vision-Aided Inertial Navigation Using a Geomatics Approach," Proceedings of the 18th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2005), Long Beach, CA, September 2005, pp. 2485-2493.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In