Title :
An SLAM algorithm based on improved UKF
Author :
Qu, Liping ; He, Shuiqing ; Qu, Yongyin
Author_Institution :
Dept. of Electr. Inf. Eng., Coll. Univ. of Beihua, Jilin, China
Abstract :
Because of using system nonlinear model directly UKF overcomes the shortcomings of the methods such as EKF that they easily introduces truncation errors in the process of lining model .So it is widely used in SLAM problem. Because the square root of filter has the advantages that it can ensure the covariance matrix nonnegative, a square root version of the UKF was included in the SLAM problem that improve the performance of UKF-SLAM algorithm. Simulation result shows that this algorithm is effective.
Keywords :
Kalman filters; SLAM (robots); covariance matrices; mobile robots; nonlinear filters; covariance matrix; filter square root; improved UKF-based SLAM algorithm; lining model process; mobile robot; nonlinear model; square root version; truncation errors; Covariance matrix; Filtering algorithms; Kalman filters; Mathematical model; Simultaneous localization and mapping; Mobile Robot; SLAM; Unscented Kalman filter;
Conference_Titel :
Control and Decision Conference (CCDC), 2012 24th Chinese
Conference_Location :
Taiyuan
Print_ISBN :
978-1-4577-2073-4
DOI :
10.1109/CCDC.2012.6243112
