In this paper a concept for a Hybrid Environment Model is proposed which aims at fulfilling the requirements of future ADAS functions. As model based tracking is not sufficient to provide innovative ADAS functions and automated driving with sufficient information about the static environment of a vehicle. Therefore, new solutions for environment modeling have to be established. The proposed model solves the requirements in a scalable and efficient way by adding an Occupancy Grid to the model-based tracking. The resulting hybrid environment model forms a sensor independent and extensible data basis which also is an abstraction layer between sensorics and functions and fosters modularization and re-usability. One reason often stated against the Occupancy Grid is its high demand for CPU, memory and bandwidth resources. It was shown that using data compression by a reduced quantization, run length encoding or a quad tree the resource demand can be reduced enough to make a deployment on standard automotive hardware feasible. Ongoing research is a further investigation of the compression methods applied. Additionally, for future assistance and automation in urban scenarios, the environmental model is complemented by a part which can represent typical urban information like curbstones or traffic lights in an efficient manner.


    Access

    Access via TIB

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    An efficient environmental model for automated driving


    Contributors:
    Grewe, Ralph (author) / Hohm, Andree (author) / Lueke, Stefan (author)


    Publication date :

    2013


    Size :

    14 Seiten, Bilder, 28 Quellen



    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English