Title:  Appearance-based Vehicle Localization across Seasons in a Metric Map

Committee:

Dr. Frank Dellaert, CS, Chair , Advisor

Dr. Anthony Yezzi, ECE, Co-Advisor

Dr. Jim Regh, IC

Dr. Patricio Vela, ECE

Dr. Henrik Chistensen, ECE

Dr. Gabe Sibey, CU Boulder Computer Science

Abstract: 

This dissertation presents a vision-only approach to the outdoor localization problem. The system provides for real-time, metric localization of a moving camera (on a vehicle) in a pre-built 3D map, which is inherently robust with respect to appearance changes. This is achieved by utilizing a novel spatio-temporal map (STM) representation which is built up from multiple drives worth of stereo camera data, as well as a localization module which efficiently retrieves landmarks from the STM to perform appearance-based localization in real-time. The map encodes the landmark visibility structure of the datasets which were captured to build the map, and this structure is then exploited for efficient localization. Experiments on real data validate that the proposed method works better than conventional approaches.