Building a global map of the environment of a mobile robot: the importance of correlations

Author

Castellanos, J.A. ; Tardós, J.D. ; Schmidt, G.

Author_Institution

Dept. de Inf. e Ingenieria de Sistemas, Zaragoza Univ., Spain

Volume

2

fYear

1997

fDate

20-25 Apr 1997

Firstpage

1053

Abstract

The work presented in this paper is aimed at evaluating the influence of correlations between map entities on the process of robot relocation and global map building of the environment of a mobile robot navigating in an indoor environment. An EKF filter approach, supported by a probabilistic model to represent uncertain geometric information, is used to process the information obtained by the sensors mounted on the robot. We have developed two approaches, first, considering the existence of correlations, and second assuming independence between entities of the map. We have experimented with the mobile robot MACROBE, using its laser rangefinder

Keywords

Kalman filters; laser ranging; mobile robots; nonlinear filters; path planning; probability; MACROBE; correlations; extended Kalman filter; global map building; indoor environment; laser rangefinder; mobile robot; probabilistic model; uncertain geometric information; Buildings; Indoor environments; Information filtering; Information filters; Laser modes; Mobile robots; Navigation; Path planning; Robot sensing systems; Solid modeling;

fLanguage

English

Publisher

ieee

Conference_Titel

Robotics and Automation, 1997. Proceedings., 1997 IEEE International Conference on

Conference_Location

Albuquerque, NM

Print_ISBN

0-7803-3612-7

Type

conf

DOI

10.1109/ROBOT.1997.614274

Filename

614274