Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
Handling the microscopic level of Vehicle-Pedestrian interactions is highly useful for analyzing patterns-based pedestrian accident exposure. Recently proposed systems are based on vehicle-to-pedestrian communication to analyze real-time situations and thus warning both sides of potential crashes. Two parts are composing those systems. The first concerns network technologies ensuring direct communication between vehicles and pedestrians and the second is related to information processing analyzing and predicting trajectories and dangers. One of the most challenges facing the last systems category is to provide them with precise and sufficient data to test and improve them. Precisely, to analyze vehicle-pedestrian interactions, we suggest detecting pedestrian behavior with respect to its surrounding vehicles as well as detecting vehicle behavior with respect to its surrounding pedestrians. In this work, we consider generating pedestrians and vehicles’ data using SUMO. Moreover, we design a handler for preparing Vehicle-Pedestrian interactions data for upper layers for deep analytic and patterns finding purposes. Mixed realistic traffic simulation is performed in Mohammedia city and the first results are shown.