Artificial Intelligence in Autonomous Vehicles

International Journal of Trend in Scientific Research and Development (IJTSRD)
Volume 5 Issue 2, January-February 2021 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470
@ IJTSRD | Unique Paper ID – IJTSRD38514 | Volume – 5 | Issue – 2 | January-February 2021 Page 715
Artificial Intelligence in Autonomous Vehicles
Matthew N. O. Sadiku1, Sarhan M. Musa1, Abayomi Ajayi-Majebi2
1Roy G. Perry College of Engineering, Prairie View A&M University, Prairie View, Texas
2Department of Manufacturing Engineering, Central State University, Wilberforce, Ohio
ABSTRACT
An autonomous vehicle is one that uses a combination of sensors, cameras,
radar, and artificial intelligence (AI) to travel between destinations without a
human operator. It is designed to be able to detect objects on the road,
maneuver through the traffic without human intervention, and get to the
destination safely. It is fitted with AI-based functional systems such as voice
and speech recognition, gesture controls, eye tracking, and other driving
monitoring systems. Several companies have announced their plan to get
involved in autonomous or driverless and electric vehicle technology. This
paper presents uses of AI technology in autonomous vehicles.
KEYWORDS: autonomous vehicles, driverless vehicles, artificial intelligence,
artificial intelligence in autonomous vehicles, Unmanned Ariel Vehicles
How to cite this paper: Matthew N. O.
Sadiku | Sarhan M. Musa | Abayomi Ajayi-
Majebi "Artificial Intelligence in
Autonomous
Vehicles" Published
in International
Journal of Trend in
Scientific Research
and Development
(ijtsrd), ISSN: 2456-
6470, Volume-5 |
Issue-2, February 2021, pp.715-720,URL:
www.ijtsrd.com/papers/ijtsrd38514.pdf
Copyright © 2021 by author(s) and
International Journal ofTrendinScientific
Research and Development Journal. This
is an Open Access article distributed
under the terms of
the Creative
CommonsAttribution
License (CC BY 4.0)
(http://creativecommons.org/licenses/by/4.0)
INTRODUCTION
Engineers and scientists all over the globe are struggling to
make human life more comfortable. One of their most
spectacular recent developments is autonomous vehicle
(AV). Autonomous vehicles have been hitting the headlines
lately. They usually refer to self-driving vehicles that can
fulfill transportation capabilities of a traditional vehicle.
They are not designed to obey the law but for the purposes
of promoting safe and efficient trafficanddrivingfromAtoB
[1]. They are regarded as a post-Uber disruption to public
commute and transportation of goods. They are starting to
become a real possibility in main economic sectors such as
transportation, agriculture, and military. AV technology is
regarded as a significant market disruptor for multiple
industries [2].
A host of auto heavyweights are investing in autonomous
R&D and developing road-going self-driving vehicles. Such
companies include Amazon, Audi, Apple, Waymo, Tesla,
BMW, Google, General Motors, Volkswagen, Volvo, Ford
Motor, IBM, Microsoft, Mercedes-Benz, Bosch, Nissan,
Alphabet, Honda Motor, and Uber Technologies. The United
States is presently the largest automobile market in the
world. Vehicles with crash-warning systems, lane-keeping
framework, adaptivecruisecontrol system,anti-lock braking
systems, automatically activated safety mechanisms, and
self-parking technology are already in use in the United
States.
Autonomous vehicle (AV) is one of the key application areas
of artificial intelligence (AI). In other words, AI is the most
important component of autonomous and connected
vehicles. This is due to the humongous amount of
environmental data generated by the vehicle. AVs depend on
AI to interpret the environment, understand its conditions,
and make driving-related decisions. Autonomous vehicles
employ numerous cameras, sensors, radars, communication
systems, and AI technology to enable the vehicle to generate
massive amounts of data which can be processed infractions
of a second to help the vehicle to see, hear, think and make
decisions just like human drivers do. A typical autonomous
vehicle is shown on Figure 1 [3].
Autonomous vehicles have been in development for almost
thirty years. For a vehicle to be autonomous, it must be
continuously aware of its surroundings, by perceiving and
then acting on the information. SAE International has
suggested the following levels of driving automation,
illustrated in Figure 2 [4]:
Automation for driver assistance: This is a
preliminary level or starting point of car automation
where the system assists the driver but does not take
control of the car.
Partially automated driving: The system takes partial
control, but the driver is primarily responsible for the
operation of the vehicle.
Highly automated driving: This allows users to let the
system take control of the vehicle for a longer duration
of time (e.g., on the highway).
Fully automated driving: Thesystemisresponsiblefor
driving the vehicle without interference from any
human. However, the human presence is still needed.
IJTSRD38514

International Journal of Trend in Scientific Research and Development (IJTSRD) @ www.ijtsrd.com eISSN: 2456-6470
Completelyautomated car: Thevehiclecancompletely
navigate its way through from one point to another
without any assistance from a driver.
OVERVIEW ON ARTIFICIAL INTELLIGENCE
The term “artificial intelligence” (AI) was first used at a
Dartmouth College conference in 1956. AI is now one of
the most important global issues of the 21st century. AI is
the branch of computer science that deals with designing
intelligent computer systemsthatmimichumanintelligence,
e.g. visual perception, speech recognition, decision-making,
and language translation. The ability of machines to process
natural language, to learn, to plan makes it possible for new
tasks to be performed by intelligent systems. The main
purpose of AI is to mimic the cognitive function of human
beings and perform activities that would typically be
performed by a human being. Without being taught by
humans, machines use their own experience to solve a
problem.
AI is stand-alone independent electronic entity that
functions much like human expert. Today, AI is integrated
into our daily lives in several forms, such as personal
assistants, automated mass transportation, aviation,
computer gaming, facial recognition at passport control,
voice recognition on virtual assistants, driverless cars,
companion robots, etc. AI is not a single technology but a
range of computational models and algorithms.
Some forms of AI that are most commonly used in different
applications include the following [5,6]:
Expert systems: They solve problemswithaninference
engine that drawsfroma knowledgebase equippedwith
information about a specialized domain, mainly in the
form of if-then rules. Expert systems are the earliest,
most extensive, the most active, and most fruitful area.
Fuzzy logic: This makes it possible to create rules for
how machines respond to inputs that account for a
continuum of possible conditions, rather than
straightforward binary.
Neural networks: These are specific types of machine
learning systems that consist of artificial synapses
designed to imitate the structure and function of brains.
They are similar to the human brain. They are made up
of artificial neurons, take in multipleinputs,andproduce
a single output. The network observes and learns as the
synapses transmit data to one another, processing
information as it passes through multiple layers.
Machine learning: This includes a broad range of
algorithms and statistical models that make it possible
for systems to find patterns, draw inferences, and learn
to perform tasks without specific instructions. Machine
learning is a process that involves the application of AI
to automatically perform a specific task without
explicitly programming it. ML techniques may result in
data insights that increase production efficiency.Today,
artificial intelligence is narrow and mainly based on
machine learning.
Deep learning: This is a formofmachinelearningbased
on artificial neural networks. Deep learning
architectures are able to process hierarchies of
increasingly abstract features, making them especially
useful for purposes like speech and image recognition
and natural language processing. Deep learning
networks can deal with complex non-linear problems.
Deep learning is expected to be the largest and the
fastest-growing technology in the automotive AI
market.
Natural Language Processors: For AI to beuseful tous
humans, it needs to be able to communicate with us in
our language. Human language is complex, but AIcanbe
trained to slowly pick up the language. Computer
programs can translate or interpret language as it is
spoken by normal people.
Robots: These are computer-based programmable
machines that have physical manipulators and sensors.
Sensors can monitor temperature, humidity, pressure,
time, record data, and make critical decisions in some
cases. Robots have moved from science fiction to your
local hospital. In jobs with repetitive and monotonous
functions they might even completely replace humans.
Robotics and autonomous systems are regarded as the
fourth industrial revolution.
These AI tools are illustrated in Figure 3 [7]. Each AI tool has
its own advantages. Using a combination of these models,
rather than a single model, is recommended. AI systems are
designed to make decisions using real-time data. They have
the ability to learn and adapt as they make decisions.
AI IN AUTONOMOUS VEHICLES
Artificial intelligence (AI) is a branch of computer science
that deals anything related to making machines smart. AI
uses data and algorithms to replicate human decision or
thinking ability. AI performs several tasks in a self-driving
automobile. One of the main tasks is path planning, which
involves the navigation system of the vehicle. Path planning
is needed to optimize the trajectory ofthevehicleandtolead
to better traffic patterns. AI helps vehicles to navigate
through the traffic and handle complex situations. AI can
respond quickly to data points generated from hundreds of
different sensors. Sensors provide information for the road,
other vehicles on the road, and other impediments.
AI in the auto industry can be classified into four segments
[8]:
Autonomous driving: Autonomous or self-driving
vehicles are becoming more and more desirable.
Autonomous vehicles can bring safety because they are
much more alert and will not be overcome by
distractions.
Connected vehicles: Vehicles are quickly transforming
into connected devices. Cars now use cellular and Wi-Fi
connections to upload and download entertainment,
navigation, and operational data. AI is an essential
technology for connected vehicles.
Mobility as a service: Mobility is the lifeblood of any
city. It is what makes a city livable and an attractive
place to live. Car companies are becoming mobility
companies to address changing consumer demand. Car
ownership in urban areas may decline in favor of
various forms of public transportation and ride sharing.
Smart manufacturing: Industrial Internet of Things
(IIoT) and Industry 4.0 technologies can be used to
automate and optimize manufacturing processes. The
use of AI in vehicles helps reduce manufacturing costs
while ensuring safer and more innovative vehicle
production. Car manufacturers are using AI tools in
every aspect of the car-making process.

APPLICATIONS OF AI IN AUTONOMOUS VEHICLES
AI is revolutionizing the transportation sector. Successful
applications of AV technologies could fundamentally
transform various industries such as automotive,
transportation, energy, agriculture, battlefield, space, the
deep sea, and other dangerous environments. Autonomous
vehicles include deep space probes, spacecraft, unmanned
aerial vehicles (UAV), unmanned ground vehicles (UGV),
unmanned sea/surface vehicles (USV), and unmanned
underwater vehicles (UUV) [9].
Autonomous cars and other highway vehicles: The
automotive industry is focusing on the integration of AI
in autonomous cars, while other application areas
include R&D, procurement, supply chain management,
manufacturing, mobility services, and customer
experience. Cars are at the top of the list of autonomous
vehicles. Figure 4 illustrates a man in an autonomous
vehicle [10]. In 2015, Amazon filed for a patent for a
system that helps autonomous cars navigate roadways.
Traditionally,automotivemanufacturing wasdominated
by internal combustion engine. Today, car
manufacturers are implementing a range of AI
technologies to mimic, augment, and support actions of
humans, including voice controls, telematics, interior-
facing cameras, touch-sensitive surfaces, and
personalized platforms. Autonomous trucks have been
tested in the U.S. and Europe to enable drivers to use
autopilot over long distances.
Autonomous flight vehicle: The integration of AI with
the aircraft that could automate thetask ofmanual flight
and navigation would be considered a milestone in the
field of aeronautics. The idea of autonomous flight
vehicle incorporating AI on a large scale involves the
commercial and military aircrafts and drones
autonomously operating with little or no harm
interaction. Such autonomous vehicles will execute
shortest and fastest flight routes with more efficient
path planning, which in turn will reduce the fuel
consumption [11]. Large aircraft have been using
automatic landing systems for several years.
Autonomous undcrsea vehicle: They have also been
called autonomous surface craft. As global positioning
systems (GPS) have become more compact, effective,
and affordable, AUVs have become more capable.
Autonomous agricultural vehicles: These are
characterized as the autonomous guidance system for
agricultural vehicles including navigation sensors (GPS,
machine vision, sensors, etc.), computational methods
for features and fuse extraction, navigation planners to
supply control algorithms, and steering controllers.
Artificial intelligence (AI) and robotictechnologywill be
key tools for such vehicles [12].
Other applications of AI in the automotive industry include
predictive maintenance and driver assistance.
BENEFITS
Autonomous vehicles promise to reduce traffic congestion
and collisions, reduce accident and pollution problems,
improve traffic flow, mobility, relieve individuals from
driving, decrease fuel consumption, and facilitate
transportation and businesses operations.Basedonthedata
collected over time, AI can speculate and provide
preferences such as seat position adjustment, screen
adjustment, controls air input, and songs to be played.
Other benefits include [13]:
Safety and privacy: Security and privacy are always
being the major issue associated with the electronic
systems. The main benefit of autonomous vehicle is
safety. Autonomous vehicles are fully capableofsensing
the surrounding environment to enhance roadway
safety. AVs are designed to choose the course of action
that causes the less damage to everybody. Autonomous
cars can remove risk factors from the equation and
significantly reduce the number of accidents or crashes.
AI techniques are also being implemented in vehicles to
provide more safety by simply removing the need of a
human driver. To ensure safety, automakers are
introducing many exciting features such as automatic
emergency braking, lane crossing alerts, driver and
passenger monitoring, blind spot elimination, side
collision warning, and optional self-driving capabilities.
Risk of cyber attack will increase since the AVs are a
moving computer.
Consumer behavior: To be successful, AVs has to cope
with individual idiosyncrasies and variation in needs.
Deployment of autonomous vehicles on public roads
requires understanding the intent of humandriversand
adapting to their driving styles. Interacting with human
drivers is a major challenge of autonomous driving.
Changes in consumer behavior and technology are
disrupting traditional modes of operation. AI and
perception technologies promise to provide a safer and
more deterministic behavior which will lead to benefits
such as fuel efficiency, comfort, and convenience.
Smart mobility: Autonomous vehicles are an essential
part of smart mobility in smart cities. They will increase
the mobility of individuals who do not drive as a result
of age or disability. AI promotes advanced driving so
that people can experience easy navigation. AI can help
provide customized entertainment while the vehicle is
traveling.
Ease of claim: Data from the vehicle can be used for
faster processing of claims in case of accidents. This can
contribute to decrease in prices for insurance, since the
safety is more deterministic and guaranteed.
Less equipmentfailure: Ifa machinefailsunexpectedly
on an automotive assembly line, the costs can be
catastrophic. AI-based algorithms can digest masses of
data from sources, detect anomalies, separate errors
from background noise, diagnose the problem, and
predict if a breakdown is likely or imminent.
Quality control: Once the parts are put intoproduction,
it is necessary to carefully control their quality. Sensor-
based AI can assess the quality of every part on the
production line and detect defects more accurately than
humans.
Leaner supply chain: Automotive supply chains are
complex networks. AI-powered supply chains are being
used to analyze a massive amount of data to be able to
forecastaccurately. AI-poweredsupplychainswill allow
fully automated self-adjusting systems to make supply-
chain management decisions autonomously, adjusting
routes and volumes to meet predicted demand spikes.
Predictive maintenance: Predictive analytics is one of
the strongest capabilities of AI and machine learning.
The essence of predictive maintenance is that the

system analyzes the equipment,comparesitsspecswith
industry and safety standards. The essence of reactive
maintenance is to replace a critical partbeforeitcrashes
the production system.
CHALLENGES
There are still several aspects of autonomous vehicles that
are presenting significant challenges and some of the
challenges are due to the intrinsic nature of AI. This section
presents specific challenges for developing, testing, and
deploying AI technologies in the autonomous vehicle. Some
serious obstacles such as adequate regulations and
international standards for digital infrastructure remain.
Other challenges include [14]:
Complexity:Autonomousdrivingsystemsarebecoming
increasingly complex and must be tested effectively
before deployment. This necessitates long and
numerous rounds of testing. Intensive AI algorithms
consume more power, especially for electric vehicles.
Accidents with testing vehicles often happen because
the simulation environment is different from the real
world conditions.
Cost: Manufacturers of autonomousvehicleshavespent
heavily in building those vehicles. For example, Google
paid about $80,000 for the AV module which is way out
of a normal man’s reach. It is hoped that in the future,
the prices of AVs will come close to the cost of
traditrional vehicles.
No regulation and standard: There are no clear
regulations on data collection and governing the new
methods of autonomoustransportation.Lawmakersand
regulators are yet to determine who is liable when an
autonomous vehicle is involved in an accident. AV
technology presents a hugechallengetostandardization
and legal bodies.
No user’s trust: Automakers are facing the need of
gaining the user’s trust. The willingness of people to
trust AI is increasing at a very slow rate. Modern users
want a car to be functional, comfortable, and safe.
Amount of data: AI techniques need to interact with
numerous sensors and use data in real time. This is
related to achieving safety while driving the vehicle.
Great amount of data and expertise is required to build
and leverage simulators. To mimic human behaviors is
difficult. The autonomous driving task as a whole
becomes hugely complex due the massive amount of
data involved.
Social dilemma: Autonomous vehicles are meant to
reduce traffic accidents, but they will sometimeshaveto
choose between two evils, such as running over
pedestrians or sacrificing themselves and their
passenger to save the pedestrians. Developing the AI
algorithms that will help AVs to handle this social
dilemma is formidable.
All of these challenges explain the slower adoption of AI the
automotive industry.
THE FUTURE OF AI IN AUTONOMOUS VEHICLES
Cars are being manufactured all over the world, with each
manufacturer in intense competition with one another to
produce the best vehicle. Autonomous cars are the future
smart cars which are expected to be driverless, efficient,
crash avoiding, and ideal urban car of the future. Some are
working tirelessly to create their very own self-driving
vehicle from scratch. Car manufacturers around the globe
are using AI in just about every facet of the car
manufacturing process. AI is changing the way cars are
manufactured globally. Due to the various challengesofAI in
autonomous vehicles, barriers to widespread adoption
remain. In the near future, AI will enable autonomous
vehicles to become mainstream. Technology companies are
at the forefront, leveraging theirAIexperiencetocapture the
autonomous vehicle market.
Connected and automatedvehiclehasbecomethefocal point
of current transportation studies (covering topics like
automation, car visions systems, and AI) and has a crucial
role to play in the future of transportation. The demand and
the need for autonomous vehicle technology is almostthere.
As the autonomous vehicle technology matures, personal
and public transportation will be greatly transformed.Aday
is fast approaching when you can commute to work with
driverless car, without needing to watch the road.
Figure 5 shows seven trends shaping the automotive
industry by 2030 [15].
CONCLUSION
Artificial intelligence is making inroads in the automotive
industry. It is the backbone of self-driving, autonomous or
connected vehicles. It is being harnessed to bring
autonomous driving from in reality. AVsmustbeinstalledon
a mobile platform. Leading automakers use the AI
technologies in their operationsfromdesigndevelopmentto
the sale of a car. To stay on top of the developments in AI in
autonomous vehicles, oneshouldconsultthebooksin[16-21]
and two related journals: Artificial Intelligence Review and
Artificial Intelligence and Law.
REFERENCES
[1] H. Prakken, “On the problem of making autonomous
vehicles conform to traffic law,” Artificial Intelligence
and Law, vol 25, no. 3, pp. 341–363.
[2] C. Medrano-Berumen and M. I. Akbaş, “Validation of
decision-making in artificial intelligence-based
autonomous vehicles,” Journal of Information and
Telecommunication, 2020.
[3] A. Hristozov, “The role of artificial intelligence in
autonomous vehicles,” July 2020,
https://www.embedded.com/the-role-of-artificial-
intelligence-in-autonomous-vehicles/
[4] R. Elezaj, “How AI is paving the way for autonomous
cars,” October 2019,
https://www.machinedesign.com/mechanical-
motion-systems/article/21838234/how-ai-is-paving-
the-way-for-autonomous-cars
[5] M. N. O. Sadiku, Y. Zhou, and S. M. Musa, “Natural
language processing in healthcare,” International
Journal of Advanced Research in ComputerScienceand
Software Engineering, vol. 8, no. 5, May 2018, pp. 39-
42.
[6] “Applications of AI and machine learning in electrical
and computer engineering,” July, 2020,
https://online.egr.msu.edu/articles/ai-machine-
learning-electrical-computer-engineering-
applications/#:~:text=Machine%20learning%20and
%20electrical%20engineering,can%20%E2%80%9C
see%E2%80%9D%20the%20environment.

[7] S. Greengard, “What is artificial intelligence?” May
2019, https://www.datamation.com/artificial-
intelligence/what-is-artificial-intelligence.html
[8] S. Rao, “Artificial intelligence in the automotive
industry,” June 2010,
https://blog.netapp.com/artificial-intelligence-in-the-
automotive-industry/
[9] E. Yagdereli, C. Gemci, and A. Z. Aktas, “A study on
cyber-security of autonomous and unmanned
vehicles,” Journal of Defense Modeling and Simulation:
Applications, Methodology, Technology, 2015, vol. 12,
no. 4, 2015, pp. 369–381.
[10] “A man in a autonomous driving test vehicle,”
https://www.123rf.com/stock-
photo/autonomous_car.html?sti=njm1zmuighvkuev2l
e|&mediapopup=36488489
[11] J. Makadia et al, “Autonomous flight vehicle
incorporating artificial intelligence,” Proceedings of
the International Conference on Computational
Performance Evaluation, Meghalaya, India. July 2-4,
2020.
[12] A. Roshanianfard et al., “A review of autonomous
agricultural vehicles,” Journal of Terramechanics, vol.
91, 2020, pp. 155–183.
[13] M. Anirudh, “AI & sutomotive — 8 Disruptive use-
cases,” March 2020,
https://unfoldlabs.medium.com/ai-automotive-8-
disruptive-use-cases-fd079926aea9
[14] O. Odukha, “Adoption of AI in the automotive
industry: is it worth the effort?” February 2020,
https://www.intellias.com/adoption-of-ai-in-the-
automotive-industry-is-it-worth-the-
effort/#:~:text=Also%2C%20artificial%20intelligenc
e%20in%20car,algorithms%20is%20worth%20the%
20money.
[15] O. Wyman, “Future automotive industry structure —
FAST 2030,” Unknown source
[16] L. B. Eliot, New Advances in AI Autonomous Driverless
Self-Driving Cars: Artificial Intelligence and Machine
Learning. LBE Press Publishing, 2017.
[17] S. Liu et al., Creating Autonomous Vehicle
Systems. Morgan & Claypool; 2nd edition, 2020.
[18] A. Raymond, How Autonomous Vehicles Will Change
the World: Why Self-Driving CarTechnologyWill Usher
in a New Age of Prosperity and Disruption. Clever
Books, 2020.
[19] M. E. McGrath, Autonomous Vehicles: Opportunities,
Strategies and Disruptions. Independently published,
2019.
[20] P. Lin, K. Abney, and R.Jenkins (eds.),RobotEthics2.0:
From Autonomous Cars To Artificial Intelligence. New
York: Oxford University Press, 2017.
[21] M. Maurer et al. (eds.),AutonomousDriving:Technical,
Legal and Social Aspects. Springer, 2016.
Figure 1 Autonomous/connected vehicle [3].
Figure 2 The levels of vehicle autonomy [4].

Figure 3 Different components of Artificial intelligence [7].
Figure 4 A man in an autonomous vehicle [10].
Figure 5 Seven trends shaping the automotive industry by 2030 [15]

Artificial Intelligence in Autonomous Vehicles

More Related Content

What's hot (20)

Similar to Artificial Intelligence in Autonomous Vehicles (20)

More from ijtsrd (20)

Recently uploaded (20)

Artificial Intelligence in Autonomous Vehicles