Design Automation for Intelligent Automotive Systems

With rapid advancement of advanced driver assistance systems (ADAS) and autonomous driving functions, modern vehicles have become ever more intelligent than before. Sophisticated machine learning techniques have being developed for vehicle perception, planning and control. However, this also brings...

Full description

Saved in:
Bibliographic Details
Main Authors: Lan, Shuyue, Huang, Chao, Wang, Zhilu, Liang, Hengyi, Su, Wenhao, Zhu, Qi
Format: Conference Proceeding
Language:English
Subjects:
Automotive engineering
Autonomous vehicles
Cameras
Neural networks
Safety
Sensors
Testing
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With rapid advancement of advanced driver assistance systems (ADAS) and autonomous driving functions, modern vehicles have become ever more intelligent than before. Sophisticated machine learning techniques have being developed for vehicle perception, planning and control. However, this also brings significant challenges to the design, implementation and validation of automotive systems, stemming from the fast-growing functional complexity, the adoption of advanced architectural components such as multicore CPUs and GPUs, the dynamic and uncertain physical environment, and the stringent requirements on various system metrics such as safety, security, reliability, performance, fault tolerance, extensibility, and cost. To address these challenges, new design methodologies, algorithms and tools are greatly needed. This paper will discuss the challenges in designing next-generation connected and autonomous vehicles, and the need of design automation techniques to tackle them.
ISSN:2378-2250
DOI:10.1109/TEST.2018.8624723