Skip to Main Content
Navigation is a broad topic that has been receiving considerable attention from the mobile robotics community. The ability to move safely in the environment is a fundamental capability for most applications. Most previous work on this subject is focused on obstacle avoidance and path planning in indoor environments using range sensors such as lasers and sonars. This paper addresses the problem of vision-based outdoor navigation. More specifically, we evaluate segmentation algorithms and combine different computer vision techniques to extract environmental information in order to autonomously conduct a mobile robot in a navigable path. As a validation to the proposed techniques, we present results obtained from field experiments.