Appearance-based mobile robot navigation using omnidirectional camera

Abstract

Appearance-based robot navigation has long been a fundamental goal in both robotics and computer vision research. While the problem is largely solved for robots equipped with active range-finding devices, for a variety of reasons, the task still remains challenging for robots equipped only with vision sensors. This paper proposes a appearance-based navigation system utilizing information from omnidirectional images captured by an omnidirectional camera as a primary sensor. The robot is able to localize with reasonable accuracy despite no metric measurements from the images. Real experiments to guide a Pioneer 3-AT mobile robot in a dynamic environment are presented, and the analysis of the results allows us to validate the proposed behaviour based navigation strategy.