Document - Deep learning in automotive: Challenges and opportunities

2017

Conference article Restricted

Deep learning in automotive: Challenges and opportunities

Falcini F., Lami G.

ADAS (advanced driver assistance systems) Deep learning Software development lifecycle Automotive SPICE W model

The interest of the automotive industry in deep-learning-based technology is growing and related applications are going to be pervasively used in the modern automobiles. Automotive is a domain where different standards addressing the software development process apply, as Automotive SPICE and, for functional safety relevant products, ISO 26262. So, in the automotive software engineering community, the awareness of the need to integrate deep-learning-based development with development approaches derived from these standards is growing, at the technical, methodological, and cultural levels. This paper starts from a lifecycle for deep-learning-based development defined by the authors, called W-model, and addresses the issue of the applicability of Automotive SPICE to deep-learning-based developments. A conceptual mapping between Automotive SPICE and the deep learning lifecycles phases is provided in this paper with the aim of highlighting the open issues related to the applicability of automotive software development standards to deep learning.

Source: Software Process Improvement and Capability Determination. 17th International Conference, pp. 279–288, Palma de Mallorca, Spain, 4-5/10/2017

Publisher: Springer, Heidelberg ;, Germania

Metrics

Back to previous page

Cite as

BibTeX entry

@inproceedings{oai:it.cnr:prodotti:380116,
	title = {Deep learning in automotive: Challenges and opportunities},
	author = {Falcini F. and Lami G.},
	publisher = {Springer, Heidelberg ;, Germania},
	doi = {10.1007/978-3-319-67383-7_21},
	booktitle = {Software Process Improvement and Capability Determination. 17th International Conference, pp. 279–288, Palma de Mallorca, Spain, 4-5/10/2017},
	year = {2017}
}