查看文章

This paper presents techniques developed for lane following and obstacle detection capabilities for autonomous vehicles. The proposed system is based on a single 120 degrees fish eye camera and a 180 degrees scanning LIDAR sensor. For the lane detection a modified lane marking technique was used representing the position of the lines making the lane with respect to the position of the camera mounted on the vehicle. The paper describes implementation of the lane detection system using Hough transform to detect the lanes. Based on the detected lane markings, the position of the relative vehicle is determined as to whether it is in its lane or if it has drifted outside the lane. The system shows the offset amount, as a percentage of distance, it has drifted from the centre of the lane. The SICK LIDAR LMS-291 S-O5 laser range sensor is adopted for the obstacle detection.