LAUSR

Current FastSLAM algorithms face challenges such as heavy computing requirements and difficulty in enhancing estimation accuracy. This paper presents a fast algorithm of simultaneous localization and mapping (SLAM) based on combinatorial interval filters coupled with an improved box particle filter (IBPF) and extended interval Kalman filter (EIKF). First, strategies for improving box contracting and resampling are studied in depth via the linear programming contractor and dimension selection subdivision resampling methods. Then, we propose a weighted average based on a time-varying Markov model to increase the estimation accuracy of the EIKF. In this way, a kind of fast SLAM algorithm is designed through combinatorial synthetic integration, in which the IBPF algorithm is employed to realize simultaneous localization and the EIKF is utilized to build a map. A series of simulations and experiments demonstrate the superior performance of our interval filters based SLAM algorithm.