IEEE Consumer Electronics Magazine - September 2018 - 23

of vehicles can cause some issues with their identification. A
vehicle being viewed from the front contains a different set of
features than a vehicle from the side or back. Often vehicle
classifiers consider various classes of vehicles, such as cars,
trucks, and semis that are trained with many orientations.

SIGN DETECTION
Many ADASs are beginning to support traffic sign detection.
The most common use case is determining the speed limit on
the road by reading a speed sign (an ADAS would alert the
driver if the vehicle speed is over the limit). For instance,
color thresholds can be used to find the location of a sign and
optical character recognition to determine what that sign displays [as shown in Figure 5(b)]. Other methods include using
CNNs and hybrid techniques, such as [20].

LANE DETECTION
Another ADAS feature used in a few production vehicles is
the ability to keep the vehicle within the lane lines on the
road (illustrated in Figure 6). However, lane lines are one of
the hardest road features to detect because of their inconsistencies, such as being different colors, faded, and sometimes
not even present. Current methods to detect lane lines often
use a Canny transform to find the edges in the image. Once
the edges are found, a Hough transform is used to compare
the lines to a single slope to determine if they are indeed lane
lines [21]. The use of CNNs is also becoming popular for lane
line detection.

COLLISION AVOIDANCE
ADASs are beginning to incorporate automatic braking and
collision avoidance. This is done by combining many features
discussed earlier, such as object tracking, vehicle detection,
and distance estimation [14]. With this combination of data, a
vehicle can predict a collision and stop it from happening by
braking or even steering out of the way.

INDOOR MONITORING
In a study conducted by the National Highway Traffic
Security Administration [22], it was observed that driver
fatigue, drowsiness, or distraction are the causes of 80%
of vehicle accidents. As ADAS becomes prevalent in production vehicles, there has been an increase in focus on
monitoring the driver using a camera pointed at him or
her. If the driver accesses a phone or does not look at the
road for a specific time duration, an alert or attempt to get
off the road will be made [23]. Drowsiness-fatigue-detection systems have also included the ability to detect if the
driver has fallen asleep and can attempt to alert the driver
though a sequence of seatbelt vibrations and speaker
alerts [24].

One of the major problems
with today's ADASs is that the
performance of the system
is significantly impacted by
changing environmental
and weather conditions.
SENSOR FUSION
Sensor fusion refers to combining information from multiple
homogenous or heterogeneous sensors to find a single best
estimation of the state of the environment. Fusion helps sensors complement each other's limitations and offers greater
leverage to the system compared to a system with individual
sensors. Sensor fusion offers high precision, reliability,
robustness to uncertainty, extended spatial and temporal coverage, and improved resolution, which are crucial in safetycritical systems, such as vehicles. Although this comes at a
higher computation cost, the computation power available in
modern-day cars and the reducing cost of the sensors are
facilitating the widespread integration of these systems.
The classification of different levels of sensor fusion along
with the most commonly used techniques for fusing data are
discussed in [25]. The growing interest in deep learning and
other ML methods in recent years has driven researchers
toward exploring more efficient and intelligent techniques
that enhance ADASs with sensor fusion capabilities.

V2X COMMUNICATION
V2X communication represents a class of communication
systems that provides the vehicle with an ability to exchange
information with other systems in the environment. Examples
include vehicle-to-vehicle (V2V) for collision avoidance,
vehicle-to-infrastructure (V2I) for traffic signal timing, vehicleto-network for real-time traffic updates, and vehicle-topedestrian for pedestrian signaling. State-of-the-art V2X
communication is based on either dedicated short-range communications (DSRC) or cellular networks [26]. The IEEE
1609 family of standards for Wireless Access in Vehicular
Environment (WAVE), which is developed based on the IEEE
802.11p standard, defines an architecture and a set of services
and interfaces to enable DSRC-based secure V2V and V2I
communication [27].

NEXT-GENERATION ADASs
Next-generation ADAS solutions are beginning to use sensor
fusion and other advanced communication systems, such as
vehicle-to-everything (V2X).

FIGURE 6. The object (lane, vehicle, and sign) detection.

september 2018

IEEE Consumer Electronics Magazine

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