Title :
Research of Mobile Robot SLAM Based on EKF and its Improved Algorithms
Author :
Chen, Chen ; Cheng, Yinhang
Author_Institution :
Sch. of Electron. & Inf. Eng., Beijing Jiaotong Univ., Beijing, China
Abstract :
Extended Kalman filter (EKF) based solution is one of the most popular techniques for solving mobile robot simultaneous localization and mapping (SLAM) problem. In this paper, the basic algorithm of EKF based SLAM and its improved algorithms are introduced. The improved algorithms are mainly on two aspects: data association and computational complexity. First, the classical data association algorithm, individual compatibility nearest neighbor (ICNN), is presented. And two improved methods including batch validation gating and multi-hypothesis are also introduced. Then, partitioned updates and submapping methods are introduced as the main ones of reducing computational complexity. Some representative improved algorithms are presented. These algorithms enable EKF to solve the mobile robot SLAM problem in cluttered and large scale environments.
Keywords :
Kalman filters; computational complexity; mobile robots; nonlinear filters; sensor fusion; batch validation gating; classical data association algorithm; computational complexity; data association; extended Kalman filter; individual compatibility nearest neighbor; mobile robot simultaneous localization and mapping; multihypothesis; partitioned updates; submapping methods; Computational complexity; Distributed computing; Electronic mail; Information technology; Intelligent robots; Mobile robots; Partitioning algorithms; Predictive models; Robot sensing systems; Simultaneous localization and mapping; EKF; SLAM; computational complexity; data association; mobile robot;
Conference_Titel :
Intelligent Information Technology Application, 2009. IITA 2009. Third International Symposium on
Conference_Location :
Nanchang
Print_ISBN :
978-0-7695-3859-4
DOI :
10.1109/IITA.2009.381
