Danger level of Intersections is particularly high because numerous vehicle trajectories converge here. The commonly used parameter TTC (Time to Collision) does not evaluate crossroad scenario exactly. EPET (Estimating Post Encroachment Time) and TTC can more effectively assess the danger degree in this scenario. In order to promote the development of active safety products at crossings, the trajectories of the two vehicles in crossroad scenario in the natural driving data are reproduced through image processing method. Features of monocular camera used in the data collecting process were considered and the picture distortion was corrected. The ‘EPET-TTC’ method is applied to divide hazard area into pieces of this kind of scenario. The evaluation method can be an advanced reference for automated vehicle products at intersections. Moreover, this article describes the procedure of utilizing Naturalistic Driving Data.
Danger Degree Evaluation Method of Crossroad Scenario Based on Naturalistic Driving Analysis
01.11.2018
418022 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
SceNDD: A Scenario-based Naturalistic Driving Dataset
IEEE | 2022
|CROSSROAD DANGER WARNING SYSTEM BASED ON SITUATION PREDICTION
Europäisches Patentamt | 2022
|Research on Evaluation Model of Danger Degree in Driving
British Library Conference Proceedings | 2019
|