A method for mobile robot obstacle avoidance based on stereo vision

An obstacle avoidance method is researched based on stereo vision by using Pioneer3-AT wheeled mobile robots as research platform. Locating obstacles is realized quickly by means of image segmentation and stereo vision algorithm which can separate obstacles from the background and stereo match their contour features with the stereo vision calibration results to get the spatial point for 3D reconstruction. The obstacle avoidance method uses stereo vision and sonar sensors working cooperatively to get the information of the obstacles in the vicinity of the robot. Fuzzy logic control algorithm is adopted to avoid collision and bypass the enroute obstacles. In the development environment of Visual C++ and OpenCV, the validity and effectiveness of the method has been demonstrated in achieving the task of evading obstacles. The method is simple and quick which is significant for the further navigation.