Fast global scan matching for high-speed vehicle navigation

This paper presents a fast global scan matching technique for high-speed vehicle navigation. The proposed grid-based scan-to-map matching technique collectively handles unprocessed scan points at each grid cell as a grid feature. The grid features are transformed and located in the global frame and updated every time a new scan is acquired. Since registered and updated are only grid features, which are each the mean of scan points in a grid cell, the proposed grid feature matching technique is very fast. Representation for each grid cell by multiple grid features further maintains accuracy regardless of the grid size while fast processing is achieved. The technique is therefore suited for localization of high-speed vehicle navigation. Experimental results show the effectiveness of the proposed technique numerically and experimentally.