Reliability and Safety Assessment of Automated Driving Systems: Review and Preview

In 2018, SAE International released a revised version of ADS (Automated Driving System) classification standard-SAE J3016 and divided it into six different classification levels. Many people doubt about the reliability of AV (Autonomous Vehicles with ADS), ADS, and HAD (Highly Automated Driving). Th...

Full description

Saved in:
Bibliographic Details
Published in2020 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM) pp. 390 - 394
Main Authors Wu, Kuo-Wei, Liao, Chung-Chih, Wu, Wen-Fang
Format Conference Proceeding
LanguageEnglish
Published IEEE 14.12.2020
Subjects
Artificial intelligence
automated driving system
autonomous vehicle
Hardware
Monte Carlo methods
Reliability
Roads
Safety
safety and reliability assessment
Software reliability
Vehicles
Online AccessGet full text

Cover

More Information
Summary:In 2018, SAE International released a revised version of ADS (Automated Driving System) classification standard-SAE J3016 and divided it into six different classification levels. Many people doubt about the reliability of AV (Autonomous Vehicles with ADS), ADS, and HAD (Highly Automated Driving). They may wonder if and when the public is ready to enter different SAE levels. To illustrate and resolve some of their questions, this paper divides "ADS reliability and safety" into the following four segments: (1) AV hardware reliability, (2) HAD reliability, (3) Integration reliability by road tests, and (4) resilience & CPS (CyberPhysical System) reliability. The paper tries to answer the following RQs. RQ1: Are the reliability and safety of vehicle hardware sufficient for the current SAE level 0-2 vehicles? RQ2: Are HAD's decisions reliable in all current workable scenarios? RQ3: Is the failure rate of ADS significantly better than that of human-driven vehicles in public testing? RQ4: Do human drivers have to intervene or participate in while driving AVs?
DOI:10.1109/IEEM45057.2020.9309972