Vision-based lane detection methods often rely on lane markings in image streams as the only source of information to determine lane boundaries. However, different road conditions such as strong shadows and lighting changes present significant challenges in extracting lane marking features, which may result in inaccurate or inconsistent lane boundary detections. This paper proposes GPS-Guided Lane Detection (GPS-LD) as a robust lane detection method that uses conventional image processing techniques for feature extraction and global route data as prior information to determine lane boundaries. The proposed method uses the route from the mission planner to determine the shape of the road in front of the vehicle. The road shape extracted from the route data is then used as a reference to cluster lane marking features and to verify the lane boundary detection results. The proposed method is evaluated on the KITTI dataset and the detection performance is comparable to state-of-the-art lane detection methods.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    GPS-guided Vision-based Lane Detection for Autonomous Vehicles


    Contributors:


    Publication date :

    2023-09-24


    Size :

    3648222 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Vision Guided Lane Transition

    Jochem, T. / Pomerleau, D. / Thorpe, C. et al. | British Library Conference Proceedings | 1995



    Lane estimation for autonomous vehicles using vision and LIDAR

    Huang, Albert Shuyu | DSpace@MIT | 2010

    Free access

    Recent Advances in Computer Vision Technologies for Lane Detection in Autonomous Vehicles

    Anto, Harshitha Devina / Malathi, G. / Kumar, G. Bharadwaja et al. | Springer Verlag | 2023


    Hybrid machine vision method for autonomous guided vehicles

    Lu, Jian / Hamajima, Kyoko / Ishihara, Koji | SPIE | 2003