Automatically reconstructing car crashes from police reports for testing self-driving cars

Alessio Gambi, Tri Huynh, Gordon Fraser

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

Abstract

Autonomous driving carries the promise to drastically reduce the number of car accidents; however, recently reported fatal crashes involving self-driving cars show this important goal is not yet achieved, and call for better testing of the software controlling self-driving cars. To better test self-driving car software, we propose to specifically test critical scenarios. Since these are difficult to test in field operation, we create simulations of critical situations. These simulations are automatically derived from natural language police reports of actual car crashes, which are available in historical datasets. Our initial evaluation shows that we can generate accurate simulations in a matter of minutes.

OriginalspracheEnglisch
TitelProceedings - 2019 IEEE/ACM 41st International Conference on Software Engineering
UntertitelCompanion, ICSE-Companion 2019
Redakteure/-innenJoanne M. Atlee, Tevfik Bultan, Jon Whittle
Herausgeber (Verlag)IEEE / ACM
Seiten290-291
Seitenumfang2
ISBN (elektronisch)9781728117645
DOIs
PublikationsstatusVeröffentlicht - 31 Mai 2019
Extern publiziertJa

Publikationsreihe

NameProceedings - 2019 IEEE/ACM 41st International Conference on Software Engineering: Companion, ICSE-Companion 2019

Fingerprint

Untersuchen Sie die Forschungsthemen von „Automatically reconstructing car crashes from police reports for testing self-driving cars“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren