Dictionary-Based Compressive SLAM

Abstract

Obtaining a compact representation of a given landmark map built by mapper robots is a critical issue for recent simultaneous localization and mapping (SLAM) applications. This “map compression” problem is explored from a novel perspective of the dictionary-based data compression approach in the paper. The primary contribution of the paper is proposal of an incremental compression approach for simultaneous mapping and map-compression applications. An incremental map compressor is presented by employing a modified random sample consensus (RANSAC) map-matching technique and the compact projection technique. Experiments evaluate the presented techniques in terms of compression speed, compactness of data and structure, and an application to the compression distance.