Paper Info
Title
Challenges in Certification of Autonomous Driving Systems
Abstract
Market readiness of on-board automotive software-intensive systems is tightly linked to the availability of appropriate certification schemes aimed at keeping the car makers confident and the consumers safe - especially in the context of Autonomous Driving, which is the next frontier of the automotive industry. Advanced driver assistance systems (ADAS) are going to be pervasively used in modern automobiles. New ADAS are principally based on Artificial Intelligence (AI) technology, and in particular on deep learning. While the automotive community is aware of the important changes such a technology demands in terms of technical skills, development paradigms, and cultural approach, there is still an important gap to be filled in the availability of technical standards and, consequently, in terms of certification capability. Currently, the global automotive industry is subject to a series of standards that are more or less explicitly referring to a traditional way of developing software and systems, so they not suitable at all to be applied to ADAS. In this paper the open issues in certification of AI technologies in automotive are addressed by providing an overview of the existing standards and the related applicability issues.
Year
DOI
Venue
2017
10.1109/ISSREW.2017.45
2017 IEEE International Symposium on Software Reliability Engineering Workshops (ISSREW)
Keywords
Field
DocType
ADAS (advanced driver assistance system),Automotive Standards,Deep Learning,ISO 26262,Automotive SPICE
Engineering management,Computer science,Advanced driver assistance systems,Software,Artificial intelligence,Deep learning,Certification,Reliability engineering,Technical standard,Automotive industry
Conference
ISSN
ISBN
Citations 
2375-821X
978-1-5386-2388-6
1
PageRank 
References 
Authors
0.34
5
2
Name
Order
Citations
PageRank
Fabio Falcini1223.73
Giuseppe Lami219522.98