Intelligent driving has been prevailing worldwide and is also challenging, which can be complicated by the factors of human drivers. In this paper, a novel collision avoidance strategy is proposed to enhance driving safety in highway overtaking by comprehensively considering the driver’s steering intent. First, in order to capture the driver’s operational characteristics from the driving data, we formulate the prediction of the driver’s steering intent and the ego vehicle’s states as a multivariate time series (MTS) forecasting problem, which is then handled by deep learning with a time pattern attention mechanism (DL-Attn). Second, a predictive risk field (PRF) model is proposed to quantify the real-time overtaking risk based on the above prediction results. Then, the overtaking is evaluated via a personalized risk threshold which can be set for a specific driver via experiments. Next, a linear time-varying model predictive control (LTV-MPC) -based assistance controller is designed so as to interfere in the risky overtaking and take over the ego vehicle from the driver to avoid possible collisions. And the feasibility and stability of the closed system are ensured theoretically. Finally, experiments are carried out in three typical cases. The results demonstrate that the proposed strategy can not only effectively improve driving safety for highway overtaking, but also identify safe overtaking to avoid unnecessary interference.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    A novel collision avoidance strategy for highway overtaking considering the driver’s steering intent


    Beteiligte:
    Zhang, Zijun (Autor:in) / Liang, Weihe (Autor:in) / Zhang, Han (Autor:in) / Zhao, Wanzhong (Autor:in) / Wang, Chunyan (Autor:in) / Huang, Heng (Autor:in)


    Erscheinungsdatum :

    01.03.2025




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    OP-CAS: Collision Avoidance with Overtaking Maneuvers

    Cha, Eun Sang / Kim, Kee-Eung / Longo, Stefano et al. | IEEE | 2018


    / COLLISION AVOIDANCE/OVERTAKING APPARATUS OF UNMANNED SURFACE VEHICLE

    SEHUN KIM / CHOI BYUNHWOONG / YEONGHUN KONG et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    Self-learning-based interpretation of driver's intent for evasive steering

    ZHANG GUOGUANG / KOSIAK WALTER K | Europäisches Patentamt | 2024

    Freier Zugriff

    SELF-LEARNING-BASED INTERPRETATION OF DRIVER'S INTENT FOR EVASIVE STEERING

    ZHANG GUOGUANG / KOSIAK WALTER K | Europäisches Patentamt | 2023

    Freier Zugriff

    Self-Learning-Based Interpretation of Driver’s Intent for Evasive Steering

    ZHANG GUOGUANG / KOSIAK WALTER K | Europäisches Patentamt | 2024

    Freier Zugriff