Radar-Based Localization Using Visual Feature Matching

Mohamed Elkholy, Mohamed Elsheikh, and Naser El-Sheimy

Abstract: In unmanned ground vehicles (UGV), the global navigation satellite system (GNSS) is the primary sensor used to estimate the vehicle’s position. However, the GNSS signal suffers from blockage or multipath error, especially when driving through canyons or tunnels or beside high buildings, e.g., downtown areas. Thus, frequency modulated continuous wave (FMCW) 360o Radar is used to compensate for the GNSS outage. This paper proposes a novel approach based on the Oriented FAST and Rotated BRIEF (ORB) method to detect features from Radar scans and matching them to estimate the vehicle’s pose. The experimental work proved the efficiency of the proposed method over the traditional techniques, e.g., Iterative Closest Point (ICP), Normal Distribution Transform (NDT), and other traditional visual features detecting methods, e.g., SURF and FAST.
Published in: Proceedings of the 34th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2021)
September 20 - 24, 2021
Union Station Hotel
St. Louis, Missouri
Pages: 2353 - 2362
Cite this article: Elkholy, Mohamed, Elsheikh, Mohamed, El-Sheimy, Naser, "Radar-Based Localization Using Visual Feature Matching," Proceedings of the 34th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2021), St. Louis, Missouri, September 2021, pp. 2353-2362.
https://doi.org/10.33012/2021.17918
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In