The present study aims to provide answers and explanations resulting from statistical analyses to critical questions related to the reliability of autonomous driving systems. The objective of the paper is to define for how long and what distances they need to be tested, until their reliable operation is proven. While the technology itself has the potential to significantly reduce the frequency of accidents in road traffic, the obtained results demonstrate that fully self-driving cars necessitate covering hundreds of thousands, or even tens of millions of kilometers to prove their reliability. This means that current fleets would require decades, or even centuries to reach these milestones – an unfeasible task if the objective is to prove their performance before they are allowed to drive on public roads. The findings of this study support the need for alternative methods to supplement real-world tests in assessing their functional safety, while also developing appropriate policies and regulations.
Failure Rate Analysis of Driverless Car Reliability Using Different Confidence Intervals
Lect. Notes in Networks, Syst.
IEEE International Conference on Cognitive Mobility ; 2023 ; Budapest, Hungary October 19, 2023 - October 20, 2023
28.04.2025
14 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
DRIVERLESS VEHICLE, DRIVERLESS VEHICLE CONTROL METHOD AND DRIVERLESS VEHICLE CONTROL APPARATUS
Europäisches Patentamt | 2017
|CONTROLLER FOR DRIVERLESS VEHICLE, AND DRIVERLESS VEHICLE
Europäisches Patentamt | 2020
|Automatic obstacle avoidance driverless vehicle driverless system
Europäisches Patentamt | 2021
|IET Digital Library Archive | 2007
|