Digital test fields for automated and connected vehicles have become increasingly important in recent years. Smart roadside infrastructure sensors, which digitally record the surrounding traffic, are used in this context to transmit traffic data for real-time and offline applications. The challenge in using such infrastructure sensors in real traffic lies in determining optimal locations and orientations for the best possible coverage of the observed road elements. Previous optimization approaches usually consider only one sensor per location and have only been evaluated on simplified 2D road geometries. We present an approach that uses a genetic algorithm (GA) to automatically optimize multiple sensor locations on a 3D digital map. Our approach is generic and applicable to any real-world road geometry. By defining separate priority areas within the digital map, we provide the GA with additional requirements and constraints to enable real-world deployment of automated vehicles on the investigated road elements. We evaluate modularly adaptable sensor setups consisting of two LiDARs and two cameras per site. The results of two different positioning modes are plausibilized with the help of a final sensor accuracy analysis and comparison with an already real existing setup. The evaluation shows that our algorithm is fully suitable for future use in the planning of digital test fields for automated and connected mobility or smart city applications.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Generic Approach to Optimized Placement of Smart Roadside Infrastructure Sensors Using 3D Digital Maps


    Contributors:


    Publication date :

    2022-10-08


    Size :

    691944 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English