Assessing safety-critical systems from operational testing: A study on autonomous vehicles
- Submitting institution
-
City, University of London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1250
- Type
- D - Journal article
- DOI
-
10.1016/j.infsof.2020.106393
- Title of journal
- Information and Software Technology
- Article number
- 106393
- First page
- -
- Volume
- 128
- Issue
- -
- ISSN
- 0950-5849
- Open access status
- Compliant
- Month of publication
- August
- Year of publication
- 2020
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
4
- Research group(s)
-
-
- Citation count
- 1
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Output demonstrates how to free Bayesian reliability assessment (favoured now by regulators like US NRC) from spurious assumptions that may make it dangerously optimistic. Producing sound arguments that autonomous cars will cause acceptably few accidents is a major challenge for the industry. This output presents maths and proposes the extra work needed for claims that are trustworthy though possibly more modest than hoped for. Work conducted as part of ICRI-SAVe (the international Intel Collaborative Research Institute – Safe Automated Vehicles). Preliminary version of output published at ISSRE2019, and shortlisted for best paper (best 3 out of 134 submissions).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
