| In many autonomous mobile robotic applications, the capability of finding a good path is of primary importance, especially for the places which are cluttered by random obstacles that may appear on the robot's path.;Our research focuses on using a color monocular vision system for the self-localization of a mobile robot at the startup phase, while in the navigation phase, the proposed algorithm combines the information collected from the monocular vision system and a belt of ultrasonic sensors mounted on the robot to detect the obstacles and avoid them as it approaches the predefined goal. The experimental results confirm the robustness of our method.;Numerous approaches for autonomous mobile robot navigation and obstacle avoidance have already been proposed that utilize different types of sensors as the range finders. The most effective methods are usually the ones that take advantage of sensor fusion to address the shortcomings due to each individual sensor. |
