Dimitrakopoulos G. The Future of Intelligent Transport Systems 2020

(National Geographic (Little) Kids) #1
31

The Future of Intelligent Transport Systems. http://dx.doi.org/10.1016/B978-0-12-818281-9.00003-6
Copyright © 2020 Elsevier Inc. All rights reserved.


Chapter 3


Computing technologies:


platforms, processors, and


controllers


3.1 Introduction to modern automotive high-performance
computing platforms


Hard real-time applications in the automotive field, such as autonomous driving
(AD), are particularly demanding from a technical and environmental perspec-
tive since they have high complexity and require a lot of computational power
and energy in order to operate. Such innovative applications emerge in auto-
motive but are equally suitable for other domains such as machinery, medical,
and avionics as well as in other cyber-physical domains. The challenges for
such applications for the automotive domain is that they have to be fully opera-
tional in difficult environmental conditions, for example, in temperature ranges
between −40 and 120°C, exposed in high humidity levels and in mechanical
and chemical stress, and must have a life span that exceeds the 15 years. The
functions to be implemented have to be available in real time and fulfill several
safety and security requirements. Therefore automotive could be treated as the
appropriate touchstone for the deployment of computation technology.
The following main trends are currently driving innovations in the automo-
tive industry:


• The transition to electric vehicles, with mechanisms that will completely
replace gas motors.


• Automated driving and supporting systems, such as the advanced driver as-
sistance system, which includes emergency braking, cruise control, lane
keeping, and other assistive functions.


• The Connected Car concept that implements connectivity between the car
and other vehicles and the car and the road network infrastructure.
These trends will have a clear impact on the future E/E (electrical/electronic)
architecture, related hardware, and software leading to a dramatically higher
overall system complexity and required computational performance [electronic
control units (ECU) as the “System,” a car as the “System,” or fleets as the
“System”].

Free download pdf