IEEE Consumer Electronics Magazine - September 2018 - 9

his article critically analyzes the trends and limits of integrated technologies for sensing,
computing, and data storage when devices and systems originally developed for consumer electronics (CE)
are used for self-driving cars. Some hints, supported by theoretical analysis and experimental work, are
provided to overcome the issues of inertial sensors, micromirrors for lidar scanners, car data/program memory,
and computing platforms.
CE is changing the way of designing, producing, and driving cars, including human-vehicle interfaces and vehicle-toeverything (V2X) networking. By 2030 [1]-[4], cars will be connected, automated, and electrified to increase safety, energy efficiency, inclusivity, and comfort. The borders between automotive and consumer industries will blur, and cars will
be consumer centric. The market of 80 million vehicles sold worldwide per year is suited for microelectromechanical systems (MEMS) and micro-optoelectromechanical-system (MOEMS) technologies. The value in cars is shifting from the
chassis/powertrain to electronics/sensors (35% of car costs in 2025). Approximately 90% of all innovations in new cars
are driven by electronics (see Figure 1).
Plenty of CE components can be used onboard cars: inertial sensors (e.g., accelerometers and gyroscopes) to estimate the vehicle dynamics and stability and to improve position/navigation accuracy (see Figure 2); micromirrors for
heads-up display projectors or lidar scanning in assisted driving; and microcontrollers
(MCU) with nonvolatile memory (NVM) to implement sensor fusion or control algorithms. Vehicles require high-performance components, since they operate in harsh environments, to ensure security and functional safety [5]. For most of CE devices, the use for
self driving is not straightforward. The issues are not only the cost, although the threedimensional (3-D) lidar price is a bottleneck [6], but mainly performance and reliability.
To this aim, this article reviews the challenges of integrated components for sensing, computing, and storage in the automotive field. Suggestions to overcome such issues are provided and supported by theoretical and experimental analysis.

INERTIAL MEMS FOR SAFETY-CRITICAL APPLICATIONS
Concerning MEMS inertial sensors, the first commercial accelerometers were available in
the mid 1990s, and commercial gyroscopes appeared a few years later. The first automotive applications, such as air-bag crash detection, were not highly demanding, a fact that
allowed the development of low-cost devices in high volumes. By the mid 2000s, lowcost, single-package, three-axis accelerometers enabled many consumer applications,
from videogame controllers to smartphone gesture recognition. Today, consumer MEMS
inertial measurement units (IMUs), with three-axis acceleration, angle rate, and magnetic
field measures, are available for a few U.S. dollars. However, the exploitation of inertial

Automated
Driving Interurban
Remote City Stop and Go Automated Parking
Parking
Highway
Highw Assist
AEB >60 km/h
Highway
Highwa Automated

2000

A
Autonomous
Driving
D

202
5

ACC

200

2020

Blind Spot
LaneDeparture
Warning

201

5
99

Side
Airbag

1980

ESC

19
9

©istockphoto.com/creativaimages

AEB City
ity
Park Assist
ssist

Airbag
ABS

FIGURE 1. Some electronics-enabled car innovations. ABS: antiskid braking system; ACC: adaptive
cruise control; AEB: autonomous emergency braking; ESC: electronic stability control.

september 2018

IEEE Consumer Electronics Magazine

http://www.istockphoto.com/creativaimages

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - September 2018