Guaranteed SLAM - A pure interval approach

Abstract

This paper presents a new Simultaneous Localisation And Mapping (SLAM) method which takes into account all measurements provided by a 2D LiDAR sensor and generates a Constraint Satisfaction Problem (CSP). Interval methods are used to solve the CSP in a guaranteed way. The approach proposed in this paper does not make any assumptions about the probability distribution of the measurements and does not require any probabilistic methods for feature detection. The SLAM problem is solved in a guaranteed way by purely using only interval methods.