New methods for Computer Vision based Low-Cost Mobile 3D Mapping and Map Update for Autonomous Driving

    Project: Research

    Project Details

    Description

    This project will investigate cost-effective methods for constructing 3D HD (High Definition) environment map that is potentially useful for autonomous vehicle navigation in a semi-structured and pre-visited urban environment, using low-cost vehicle-board multi-modality sensors (e.g. stereo vision camera, ultrasound, radar or low-cost Lidar). The methods are targeted at cloud-sourcing scenario where each individual vehicle is able to build a map, compare it with the pre-stored map or map in the cloud, and make suggestions with respect to on-line map update or on-demand map update. This research is focused on the study and experimentation of new methods for creating such map, storing and transmitting such map, as well as map comparison and updating.
    StatusActive
    Effective start/end date13/03/1913/03/25

    Fingerprint

    Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.