Print Email Facebook Twitter A method to assess safety implications during authority transitions in automated driving Title A method to assess safety implications during authority transitions in automated driving Author Athanasiadis, Omiros (TU Delft Civil Engineering and Geosciences) Contributor van Arem, Bart (mentor) Papadimitriou, Eleonora (graduation committee) Calvert, Simeon (graduation committee) Klunder, Gerdien (graduation committee) Degree granting institution Delft University of Technology Programme Transport, Infrastructure and Logistics Date 2020-07-07 Abstract The question of how well in terms of safety can a driver take over control of an automated vehicle in response to an emergency situation is of crucial importance. Most of the studies performed so far focus on the drivers’ reaction times and the mechanisms behind the transition. In this study, an effort is made to incorporate the braking times that are required in order to finalize a safety maneuver, with the aim to assess the safety implications of the entire transition in control. For this purpose, a new methodology was developed and a simulation model was used in order to simulate platoons of CACC equipped vehicles. Two new KPIs were defined: the Time to Control and the Safe Time Budget. The results suggest that higher number of critical events and crashes are associated with higher market penetration rates. This reveals that despite the fact that AV can in general increase traffic efficiency and safety, when it comes to emergency situations where safety is inextricably linked to the combination of AV and driver performance, overall safety may be compromised under certain conditions. In addition, the results revealed a strong connection of the above-mentioned action times with the initial speed of the vehicles involved in a conflict. The findings of this research, point to new directions particularly in concern to the extension of the operational design domain of automated vehicles in order to minimize system deactivations, and also with regard to the need for better prediction models and safety assessment tools. Subject Authority transitionsCACCSafety Assessmentdriver behaviourSimulationAVtake-over requests To reference this document use: http://resolver.tudelft.nl/uuid:02dc4999-7050-4cc2-9ac9-bced5e4cbf5f Part of collection Student theses Document type master thesis Rights © 2020 Omiros Athanasiadis Files PDF thesis_Omiros_Athanasiadis.pdf 3.59 MB Close viewer /islandora/object/uuid:02dc4999-7050-4cc2-9ac9-bced5e4cbf5f/datastream/OBJ/view