International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)

Volume 8, Issue 6, June 2019, ISSN: 2278 – 1323

Advanced Driver Assistance System (ADAS): A

Review of Systems and Technologies
Dr. Bassim Abdulbaqi Jumaa , Anwaar Mousa Abdulhassan, Ammar Mousa Abdulhassan

 Different sensors have different observation capabilities and

Abstract— The driving is art, elegance and ethics. The human various detection properties. Many vehicle manufacturers
acts as the key element in this process and the weakest link in and automobile device companies have attempted as shown
the same time, where 90% of traffic accidents are caused by in Fig.1. For example, laser-based approach, radar-based
human error and carelessness.
approach or vision sensing to develop speed control systems
Every year many accidents are reported due to high speed
and wrong decision. A logical principle is that each 1% increase to maintain a vehicle safe distance, which represents one of
in velocity will result in a 4% potential increase in the risk of the ADAS functions [2], [3]. In critical driving situations, this
fatal collision and an actual increase of 3% in the risk of a system warns and actively supports the driver and, if
serious collision. Modern day cars represent a symbiosis of necessary, intervenes automatically in an effort to avoid a
several electronic subsystems that collaboratively give a safe collision [1], [4].
and sound driving experience. One of the car development
technologies takes the form of the Advanced Driver Assistant
System development (ADAS). It is a system that necessary to
monitor various parameters associated with the vehicle, vehicle
surroundings to detect potentially dangerous situations at an
early stage.
Sensors with sophisticated devices, which are known as
Sensor Network (SN), were included already in many cars
nowadays to achieve (ADAS) technologies. A well-known ADAS
is an Adaptive Speed Control system (ASC), Automatic Brake
System (ABS), Warning Collision System (WCS) and Legal
Restriction System to avoid high-speed dangerous. In addition,
Lane Keeping System (LKS) and Lane Change System (LCS).

Index Terms— Advanced Driver Assistant System

development, (ADAS), Car safety, Car sensors technologies.
Fig.1. Current technologies for Assisted Driving [2].

I. INTRODUCTION
II. SELF-DRIVING (AUTOMATION) LEVELS
Automotive companies are continuously improving
Given that the vast majority of accidents can be attributed
vehicle safety, whereas safety systems developed in vehicles
to human error, taking the driver out of the loop may reduce
can be divided into two categories: passive safety systems
or even eliminate driver error, which in turn, may lead to
and active safety systems. The passive safety system reduces safer roads. Maturation, integration and affordability of
the injuries sustained by passengers when an accident occurs. enabling technologies have turned self-driving cars from
For example, airbags and seatbelts have saved thousands of science fiction into reality.Google’s self-driving car
lives and became a milestone in the automotive industry. famously has been clocking up thousands of accident-free
Active safety systems refer to systems that try to keep a miles and several countries are now preparing themselves to
vehicle under control and avoid accidents [1]. In this work, adapt laws permitting self-driving cars on public roads [5].
the active safety systems have been proposed and Automated vehicle technologies have a range of
implemented. capabilities, from anti-lock brakes and forward collision
It is expected that active safety systems will play an warning, to adaptive cruise control and lane keeping, to fully
increasing role in collision avoidance in the future. Each automated driving [5].
application supported by ADAS requires its private sensor(s); Following the Society for Automotive Engineers
therefore adding new applications will require more sensors. taxonomy (SAE, 2018), Five levels of vehicle automation
have been defined as below [6]:
Based on the capabilities of ADAS, already in use or
Manuscript received May, 2019.
implementation, the automated vehicles in general and
Bassim Abdulbaqi Jumaa, Computer Engineering Department, University particularly the most popular concept of automated driving
of Technology, Baghdad,, Iraq (e-mail: [email protected]).
become not just a vision of a remote future – rather closed
Anwaar Mousa Abdulhassan, Computer Engineering Department, systems like metros and similar rail systems, and air traffic.
University of Technology, Baghdad,, Iraq (e-mail:
[email protected]).
The levels of automated driving are shown in Fig.2 [7].
Ammar Mousa Abdulhassan, Electric Power and Electrical Engineering
Department, Tambov State Technical University, Tambov, Russia (e-mail:
[email protected]).

All Rights Reserved © 2019 IJARCET 231

 International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 6, June 2019, ISSN: 2278 – 1323

front without using the throttle or brake pedals. It can mean

that a driver does not concentrate enough and when some
unexpected hazard occurs, the driver can react late or panic
and easily cause an accident. The first versions of ACC
systems were based on laser but bad weather conditions
significantly affected reliability. There were also problems
recognizing the speed and positioning of the car in front when
it was wet or non-reflective. For these reasons, nowadays
common ACC systems use Long Range Radar. This radar
works in a wider range of weather conditions and can
recognize non-reflective cars. Typical Long Range Radar
sees up to 200 m but in a narrow range. Car manufacturers
Fig.2. Levels of Automated Driving [7]. often use a combination of Long Range radar with Short
Range Radar or with an optical system to cover a wider range.
III. ADVANCED DRIVER ASSISTANCE SYSTEM BASED ON A wider range is necessary at low speeds because the gaps are
SENSORS smaller. There are other types of ACC system, including
cheaper systems which do not support stop and go, meaning
To achieve the ADAS, the sensing process should be that a car cannot automatically fully stop and then continue
accomplished as a technique of collecting necessary its journey. It works only within a range of speeds e.g. from
information about the physical parameters related to the car, 50 km/h to 200 km/h. More sophisticated ACC systems also
what surrounds it and the driver [4], [2]. In the following have ultrasonic sensors which have a short range and are used
points an overview of the specific ADAS functions based on at low speed to get more precise information. Those systems
sensors technologies is given: can automatically stop and continue during driving e.g. slow
A. Forward Collision Warning driving in traffic jams [11], [12] and [13].
Forward Collision Warning (FCW) system is one intended D. Lane Keeping System and Lane Change Assistant
to assist the driver in avoiding or mitigating a rear-end Lane Keeping System (LKS) and Lane Change Assistant
collision via presentation of audible, visual, and/or haptic (LCA) systems are designed to reduce high-speed collisions
alerts, or any combination thereof. FCW systems have especially on highways and to eliminate run-off-road
forward-looking vehicle detection capability, provided by accidents. These systems use optical recognition of markers
technologies such as radar, lidar (laser), cameras, etc. By on roads, usually white lines. Because they rely on optical
using the information provided by these sensors, an FCW recognition, those systems are sensitive to the quality of
system alerts the driver that a collision with another vehicle roads markings and the effects of the weather. That means
may be imminent unless a corrective action is taken [8]. there is a high possibility of a mistake occurs in the situation
B. Automatic Emergency Brake of heavy rain, snow or if there is excessive glare from the sun.
LKS is a proactive system that not only warns the driver
Automatic Emergency Brake (AEB) is an intelligent when to start moving out of the lane but it can also turn the
and complex system. Sometimes AEB system is called steering to control the car direction and therefore decrease
Collision Warning with Auto Brake or Automatic Emergency chances of an accident when the driver does not pay full
Brake, the specific name depends on the car manufacturer but attention. Sensors that are used in common cars are
the principle the same. This system can apply braking power Multi-Function Mono Camera or Multi-Function Stereo
automatically without driver intervention. AEB continuously Camera. Stereo camera has the advantage of being able to
monitors the area in front of the car and when the systems recognize 3D objects, lanes and possibly obstacles [1], [14].
recognize a serious possibility of a collision occurring then Fig.3 illustrates LKS and LCA [15].
the driver is alerted to start braking and brakes are
pre-activated. Sometimes small braking power is applied to
save as much braking distance as possible. If the driver, at the
critical distance, does not start braking, the system
automatically applies as much braking power as possible to
stop the vehicle or to reduce speed and the possibility of the
accident [9].
Most rear-end collisions occur in inter-urban areas at
low speed. To monitor the rear-end of a car, manufacturers
use short-range lidar (light detecting and ranging) or Fig.3. LKS and LCA [15]
long-range radar (radio detecting and ranging) sensor.
Short-range lidar is cheaper and works at speeds of up to 50 IV. SPECIFICATIONS OF VEHICLES SENSORS
km/h. The long-range radar sensor is necessary for adaptive The vehicles, which are fitted with ADAS, have many
cruise control as it can also recognize critical situations and sensors to provide all the necessary information to the
work at speeds of up to 200 km/h. The data from these systems. An overview of some sensors technologies used in
sensors are interpreted in the control unit to take the ADAS are given below:
appropriates actions. [9], [10]. A. RaDAR (Radio Detection And Ranging)
C. Adaptive Cruise Control RaDAR is a sensor that uses electromagnetic (radio)
Adaptive Cruise Control (ACC) is a simple system which waves for detecting the distance of fixed objects and the
allows drivers to adjust their speed and distance from a car in

www.ijarcet.org 232
 International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 6, June 2019, ISSN: 2278 – 1323

distance and speed of moving objects. It uses 22-29 GHz and determine the distance of an object in relatively close
77-81 GHz radar. There are: proximity. It measures the distance by sending out a sound
- Short-range and wide-angle (typically up to 50 m, wave at a specific frequency and listening for that sound
130 degree field of view). wave to return. It is possible to calculate the distance between
- Mid-range and wide-angle (typically up to 100 m the sensor and the detected object, where (distance = speed x
and more than 30 degree field of view). time) and by recording the elapsed time between the
- Long-range and narrow-angle (typically more than generated sound wave and the returned-back sound wave.
100 m and less than 20 degree field of view). Types Fig.6 shows the ultrasonic sensor that is used in cars [16].
of radars are shown in Fig.4 [17]. All automotive
radars are of Doppler type. Radar sensors for ACC
or AEB usually contain two transmitters and four
receivers or one transmitter and two receivers as a
cheaper option, which is used for detection of rear
vehicles [15], [16].

Fig.6. Ultrasonic Sensor [16]

D. Optical Cameras
Optical or vision systems, such as mono and stereo
cameras, as shown in Fig.7 [17], can be used in some
conditions to effectively map the environment of potential
hazards. The strengths and weaknesses of camera systems for
quantifying the environment are similar to those of human
vision. Camera systems cannot directly sense the relative
speed of a potential hazard and can be sensitive to
environmental conditions such as low sun angle, heavy
precipitation, and fog that reduce visibility. Therefore, they
are often used in concert with radar sensors, where they
complement radar’s limited angular resolution and limited
field of view. Stereo cameras allow the distance of a potential
hazard to be calculated. The usable distance for cameras is
Fig.4. Type of radars on vehicles [16] short-range detection, typically less than 100m. Modern
image processing methods and dedicated image processing
B. LiDAR (Light Detection And Ranging) hardware have allowed the development of camera systems
Fundamentally, LiDAR is similar to radar, except that it that are very effective at identifying objects in the
utilizes a transmitted laser beam that is then sent as a reflected environment around the vehicle in real time. A vision system
light. It can be similarly used to determine the speed and achieves excellent lateral estimation but can fall short in
distance of potential hazards but from 10 to 20m at low speed. longitudinal parameter estimation. Another usage of cameras
It typically has a shorter effective range than radar. Its is for driver monitoring. These cameras monitor the speed of
wavelength is 850 or 900 nm. LiDARs were first applied as eyes winking and head movements. From such information,
sensors for ACC because they were smaller. Nowadays they the system can recognize that the driver does not pay full
are often replaced by radars or cameras due to high cost and attention and should stop [16].
low resolution capabilities [17] and [18]. Fig.5 shows the
LiDAR look [17].

Fig.7. (A) Mono camera, (B) Stereo camera [17]

About the main technologies of RaDAR, LiDAR, US and
Cameras, a summarization of their main properties for the
environment perception is in Table 1. The camera is a
Fig.5. LIDAR look [17]
particular case, as by itself cannot detect objects, but it
depends on image processing algorithms to do so. As can be
C. US (Ultrasonic Sensor)
seen, none of these technologies is ideal and an attractive idea
Ultrasonic sensors are also similar to radar in concept, is to complement two or more to obtain the best results [14].
except that they utilize the reflection of ultrasound waves to

All Rights Reserved © 2019 IJARCET 233

 International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
Volume 8, Issue 6, June 2019, ISSN: 2278 – 1323

Table 1 Comparision of main technologies for

enveronment perception

REFERENCES
[1] M. Lu, K. Wevers, and R. Van Der Heijden, “Transportation Planning
and Technology Technical Feasibility of Advanced Driver Assistance
Systems ( ADAS ) for Road Traffic Safety,” Transp. Plan. Technol.,
vol. 28, p. 167?/ 187, 20115.
[2] M. Julia Carrillo, “Robotic Cars Test Platform for Connected and
Automated Vehicles,” UNIVERSITY OF CALIFORNIA, IRVINE
Robotic, 2015.
[3] A. Shaout, D. Colella, and S. Awad, “Advanced Driver Assistance
Systems Past , Present and Future,” in Seventh International Computer
Engineering Conference (ICENCO 2011), 2011, pp. 72–82.
[4] D. Sunehra and K. Jhansi, “Implementation of Microcontroller Based
Driver Assistance and Vehicle Safety Monitoring System,” in 2015
International Conference on Information Processing (ICIP), 2015, pp.
423–428.
[5] C. Diels and J. E. Bos, “Self-driving carsickness,” Appl. Ergon., vol.
53, pp. 374–382, 2016.
[6] “SAE Levels of Driving Automation,” American National Standards
Institute, 2018.
[7] E. Schoitsch, “AUTONOMOUS VEHICLES AND AUTOMATED
DRIVING STATUeS, PERSPECTIVES AND SOCIETAL
IMPACT,” in IDIMT 2016 Proceedings, 2016, no. September, pp.
405–423.
[8] G. J. Forkenbrock and B. C. O. Harra, “A FORWARD COLLISION
WARNING (FCW) PERFORMANCE EVALUATION,” in In Proc.
21st Int. Technical Conf. Enhanced Safety of Vehicles, 2009, pp. 1–12.
[9] E. Coelingh, A. Eidehall, and M. Bengtsson, “Collision Warning with
Full Auto Brake and Pedestrian Detection - a practical example of
Automatic Emergency Braking,” in 2010 13th International IEEE
Annual Conference on Intelligent Transportation Systems, 2010, pp.
155–160.
[10] E. M. Shakshuki, W. Alghamdi, and T. Sheltami, “Integrated
context-aware driver assistance system architecture,” in Procedia
Computer Science, 2013, vol. 9, pp. 225–240.
[11] L. Viktorová and M. Šucha, “Learning about advanced driver
assistance systems – The case of ACC and FCW in a sample of Czech
drivers,” Transp. Res. Part F J., pp. 1–8, 2018.
[12] U. Hofmann, A. Rieder, and E. D. Dickmanns, “Radar and vision data
fusion for hybrid adaptive cruise control on highways,” Mach. Vis.
Appl., pp. 42–49, 2003.
[13] A. Kesting, M. Treiber, M. Scho, and D. Helbing, “Adaptive cruise
control design for active congestion avoidance,” Transp. Res. Part C,
vol. 16, pp. 668–683, 2008.
[14] C. Ilas, “Electronic Sensing Technologies for Autonomous Ground
Vehicles : A Review,” 8th Int. Symp. Adv. Top. Electr. Eng., pp. 0–5,
2013.
[15] J. Liu, Q. S. Team, and C. Yan, Can You Trust Autonomous Vehicles :
Contactless Attacks against Sensors of Self-Driving Vehicles. 2016.
[16] J. Zolock, C. Senatore, R. Yee, R. Larson, and B. Curry, “The Use of
Stationary Object Radar Sensor Data from Advanced Driver
Assistance Systems ( ADAS ) in Accident Reconstruction,” SAE Tech.
Pap., pp. 1–23, 2016.
[17] F. Harashima, M.Parent, and L. Vlacic, Intelligent Vehicle
Technologies. Elsevier, 2001.
[18] G. Lu and M. Tomizuka, “LIDAR Sensing for Vehicle Lateral
Guidance : Algorithm and Experimental Study,” ASME Trans.
MECHATRONICS, vol. 11, pp. 653–660, 2006.

www.ijarcet.org 234