who DRIvEs thE CaR?
5.1 Introduction
Can a car read a trac sign?
Can it shut its grill vents to save fuel?
Can it stop itself to help prevent an accident or park itself in a tight spot?
With a host of technologically advanced features, this one can.
e all-new Ford Focus: Start more than a car.*
Making the car more intelligent—in other words, car robotization—
has taken o in recent years. e development of smart driver assis-
tance systems in cars is currently rapidly progressing. In part, this is
caused by the decrease in the price of the required components, such as
video cameras, microprocessors, and sensors. erefore, such systems
are no longer just built into expensive car models but are also increas-
ingly tted in middle-class models. In addition, car manufacturers
especially compete with each other in terms of comfort and safety,
because there is actually not much more to improve in the quality of
cars (Gusikhin, Filev, & Rychtyckyj, 2008). Intelligence, therefore,
becomes the unique selling point for a new car, as we can see from
the advertising text of the Ford Focus. Future systems will evolve
from “driver assistance” to “fully automated (autonomous) driving,
completely piloting a car along highways and through urban environ-
ments. Although the idea of driverless cars on the road may seem
futuristic, industry leaders anticipate that autonomous cars will hit the
road within the next decade. is projection is due to the fact that the
majority of the technologies necessary to build a fully autonomous car
already exist (Pawsey & Nath, 2013). is vision of the path toward
fully autonomous cars assumes that car robotization is a continuum
between conventional, fully human-driven vehicles and vehicles that
186 Just ordinAry robots
require no driver at all. Actually, it is an ongoing automation and
interconnection of single vehicles and trac’s infrastructure that
aims at fully self-driving or autonomous cars. e National Highway
Trac Safety Administration (2013) developed a ve-level hierarchy
to conceptualize this continuum as guidance “to help states imple-
ment this technology safely so that its full benets can be realized:
Level 0 (no automation): e human driver is in complete con-
trol of all functions of the vehicle.
Level 1 ( function-specic automation): One control function is
Level 2 (combined-function automation): More than one con-
trol function is automated at the same time (e.g., steering and
acceleration), but the driver must remain constantly attentive.
Level 3 (limited self-driving automation): e driver functions
are suciently automated so that the driver can safely engage
in other activities (e.g., reading a paper).
Level 4 (fully self-driving automation): e car can drive itself
without a human driver.
e automation of control functions started with the features of con-
venience and safety, such as cruise control in 1958* and anti-lock
braking system (ABS) in 1972
(level 1). Automation of multiple and
integrated control functions, such as adaptive cruise control with lane
centering, has become more common recently. ey are classied as
level 2, because they still allow the driver to override them: the driver
is responsible for monitoring the roadway and is expected to be avail-
able for control all the time. e next step in this continuum is taking
over driving tasks through cooperative systems—in conjunction with
trac management (level 3). Ultimately, it is expected that this will
lead to autonomous vehicles (level 4).
In this chapter, we dene car robotization as a combination of
developments in the following technologies: driver assistance systems,
trac management, and cooperative systems. In fact, these technolo-
gies are the building blocks for the fully autonomous car, which we

Get Just Ordinary Robots now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.