Modeling Traffic Accidents Caused by Random Misperception

verfasst von

Volker Berkhahn, Marcel Kleiber, Chris Schiermeyer, Stefan Weber

Abstract

Understanding the formation of accidents is of major importance to the automotive industry, its related businesses and policymakers. This is not a trivial task considering the current stream of innovations driven by the development of autonomous vehicles. Historical accident data are inadequate for gauging the safety of future traffic systems. To cope with this challenge, we propose a microscopic traffic model that introduces small errors due to random misperception as an omnipresent cause for accidents - an issue affecting both human drivers and control systems of autonomous vehicles. We model errors dynamically by stochastic processes and investigate their impact on the safety and the efficiency of traffic systems by Monte Carlo simulations. We focus on two case studies: a simple one-lane road segment and a t-junction with turning vehicles.

Organisationseinheit(en)

Institut für Risiko und Zuverlässigkeit
Institut für Versicherungs- und Finanzmathematik

Typ

Sonstige Publikation

Anzahl der Seiten

7

Publikationsdatum

11.2018

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

ASJC Scopus Sachgebiete

Maschinenbau, Fahrzeugbau, Angewandte Informatik

Ziele für nachhaltige Entwicklung

SDG 3 – Gute Gesundheit und Wohlergehen

Elektronische Version(en)

https://doi.org/10.1109/itsc.2018.8569483 (Zugang: Geschlossen)