During driving in real traffic, often conflict situations arise which do not lead to an accident; however, they offer a good indication of traffic (un)safety. There are many methods and indicators for classifying whether a current driving situation can be considered as a conflict or not (e.g., time to collision, post-encroachment time). However, not many approaches exist to predict how such conflicts will evolve once they have been identified and what the collision probability would be given the conditions of the conflict. Currently available methods make strong assumptions about driver behavior during the conflict, usually do not consider all participants involved in the conflict (only two vehicles), and have limited applicability to various types of real conflicts. In this paper, a Monte Carlo-based microsimulation approach is proposed to estimate the probability of collision for a conflict of any type. Unlike previous approaches, the proposed method can be used to simulate any type of conflict with an arbitrary initial conflict condition and an arbitrary number of vehicles in conflict. The method was developed based on real-world video data of a complex signalized intersection, from which conflicts were identified with the DOCTOR method. As proof of concept, the proposed method was applied to a real conflict involving four vehicles which was extracted from these video data and validated with a real traffic accident. The results show that our approach is capable of simulating real conflicts of various types with multiple participants and predicting their probability of collision.


    Zugriff

    Download

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Monte Carlo-Based Microsimulation Approach for Estimating the Collision Probability of Real Traffic Conflicts


    Weitere Titelangaben:

    Transportation Research Record: Journal of the Transportation Research Board


    Beteiligte:


    Erscheinungsdatum :

    03.03.2023




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch