DocumentCode :
414318
Title :
A comparison of maximum likelihood methods for appearance-based minimalistic SLAM
Author :
Rybski, Paul E. ; Roumeliotis, Stergios I. ; Gini, Maria ; Papanikolopoulos, Nikolaos
Author_Institution :
Dept. of Comput. Sci. & Eng., Minnesota Univ., Minneapolis, MN, USA
Volume :
2
fYear :
2004
fDate :
April 26-May 1, 2004
Firstpage :
1777
Abstract :
This paper compares the performances of several algorithms that address the problem of Simultaneous Localization and Mapping (SLAM) for the case of very small, resource-limited robots. These robots have poor odometry and can typically only carry a single monocular camera. These algorithms do not make the typical SLAM assumption that metric distance/bearing information to landmarks is available. Instead, the robot registers a distinctive sensor "signature", based on its current location, which is used to match robot positions. The performances of a physics-inspired maximum likelihood (ML) estimator, the iterated form of the Extended Kalman Filter (IEKF), and a batch-processed linearized ML estimator are compared under various odometric noise models.
Keywords :
Kalman filters; image sensors; maximum likelihood estimation; mobile robots; path planning; appearance based minimalistic SLAM; batch processed linearized ML estimator; bearing information; distance information; distinctive sensor; iterated extended Kalman filter; maximum likelihood methods; monocular camera; odometric noise models; physics inspired maximum likelihood estimator; resource limited robots; robot positions; robot registers; simultaneous localization and mapping; Cameras; Computer science; Lenses; Maximum likelihood estimation; Mobile robots; Robot sensing systems; Robot vision systems; Robustness; Simultaneous localization and mapping; Workstations;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Robotics and Automation, 2004. Proceedings. ICRA '04. 2004 IEEE International Conference on
ISSN :
1050-4729
Print_ISBN :
0-7803-8232-3
Type :
conf
DOI :
10.1109/ROBOT.2004.1308081
Filename :
1308081
Link To Document :
بازگشت