This thesis is devoted to detection of 3-dimensional point obstacles on the plane by using accumulated scan line images. Real time image processing is enabled by accumulating only one scan line from each image frame. And the change of motion of the feature in image is small because of the short time between image frames, so it doesn't take much time to track features. To obtain recursive optimal obstacles' position and robot motion along to the motion of camera, Kalman filter algorithm is applied. From Kalman filter in case of the fixed environment, 3-dimensional obstacles point map can be obtained. Position and motion of moving obstacles can also be obtained by pre-segmentation. The last, to solve the stereo ambiguity problem from multiple matches, the camera motion is actively used to discard mis-matched features.
To get relative obstacles' distance from camera, parallel stereo camera setup is used. For experiments test vehicle is made. It is made up of 486-PC and DSP image processing board and it is driven by two step motors. Scan line images are stored on hard disk and transferred to the workstation, so stereo matching and Kalman filter algorithm are performed in the workstation. The experiment to find environment map is performed on-line.