Y. Lu, V. Polotski, and J. Sasiadek (Canada)
Mobile robot, localization, segmentation, data association.
This paper addresses the problem of outdoor localization using a 2D laser range finder for a real size mobile robot. A priori map of the environment should be known. Using robot pose estimate provided by a basic navigation system, the local map is scanned; both the real scans and the map scans are approximated by line segments and compared against each other. Data association and matching algorithms have been developed, which allows for a significant improvement of the robot pose estimate. The results have been implemented and tested on the mobile robot executing a patrol mission.
Important Links:
Go Back
