Jagatheesaperumal et al.

Computational Urban Science (2024) 4:10 Computational Urban

[Link]
Science

ORIGINAL PAPER Open Access

Artificial intelligence of things for smart

cities: advanced solutions for enhancing
transportation safety
Senthil Kumar Jagatheesaperumal1 , Simon Elias Bibri2*, Jeffrey Huang2, Jeyaranjani Rajapandian1 and
Bhavadharani Parthiban1

Abstract
In the context of smart cities, ensuring road safety is crucial due to increasing urbanization and the interconnected
nature of contemporary urban environments. Leveraging innovative technologies is essential to mitigate risks
and create safer communities. Thus, there is a compelling imperative to develop advanced solutions to enhance
road safety within smart city frameworks. In this article, we introduce a comprehensive vehicle safety framework
tailored specifically for smart cities in the realm of Artificial Intelligence of Things (AIoT). This framework seamlessly
integrates a variety of sensors, including eye blink, ultrasonic, and alcohol sensors, to bolster road safety. The utiliza-
tion of eye blink sensor serves to promptly detect potential hazards, alerting drivers through audible cues and thereby
enhancing safety on smart city roads. Moreover, ultrasonic sensors provide real time information about surrounding
vehicle speeds, thereby facilitating smoother traffic flow. To address concerns related to alcohol consumption and its
potential impact on road safety, our framework incorporates a specialized sensor that effectively monitors the driver’s
alcohol levels. In instances of high alcohol content, the system utilizes GPS and GSM technology to automatically
adjust the vehicle’s speed while simultaneously notifying pertinent authorities for prompt intervention. Additionally,
our proposed system optimizes inter-vehicle communication in smart cities by leveraging Li-Fi technology, enabling
faster and more efficient data transmission via visible light communication (VLC). The integration of Li-Fi enhances
connectivity among connected vehicles, contributing to a more cohesive and intelligent urban transportation
network. Through the structured integration of AIoT technologies, our framework lays a robust foundation for a safer,
smarter, and more sustainable future in smart city transportation. It offers significant advancements in road safety
and establishes the groundwork for further enhancement in intelligent urban transportation networks.
Keywords Smart cities, Artificial intelligence of things, Road safety, Transport networks, Ultrasonic sensor, MQ3
sensor, Li-Fi Technology, V2V communication

1 Introduction
The increasing number of fatal road accidents poses a
critical challenge in daily transportation within smart cit-
*Correspondence:
Simon Elias Bibri ies. This increase can be attributed to factors like reckless
[Link]@[Link] driving, distracted driving, driving under the influence
1
Department of Electronics & Communication Engineering, Mepco of drugs or alcohol, and poor road conditions, causing
Schlenk Engineering College, Sivakasi 626005, Tamilnadu, India
2
Institute of Computer and Communication Sciences (IINFCOM), School physical, emotional, and financial harm to those affected.
of Architecture, Civil and Environmental Engineering (ENAC), Media Prioritizing road safety in smart cities involves following
and Design Laboratory (LDM), Swiss Federal Institute of Technology traffic rules, maintaining vehicles, and driving responsi-
Lausanne (EPFL), 1015 Lausanne, Switzerland
bly (Yuvaraju et al., 2021). Studies have shown a growing

© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit [Link]
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 2 of 18

frequency of road incidents due to negligent driving, decision making and automation processes across vari-
distracted driving, driving while intoxicated, and inade- ous domains of smart cities (Bibri et al., 2023a, b) point
quate road maintenance, highlighting the critical need to to promising opportunities for developing innovative
adhere to traffic regulations and ensure responsible driv- vehicle safety frameworks, transforming urban mobil-
ing (Yuvaraju et al., 2021). The study conducted by (Singh ity and making the transportation ecosystem safer and
& Bera, 2018) points to driver errors, such as speeding, more efficient for all road users.
fatigue, and alcohol consumption, as major causes of In recent years, the urgent need to modernize and
road accidents, emphasizing the collective responsibil- enhance urban mobility systems has led to the adop-
ity of society to improve infrastructure, enforce stricter tion of Artificial Intelligence of Things (AIoT) in
laws, and raising awareness about the dangers of reckless smart city transportation (Cui & Lei, 2023; Gong
driving. To prevent accidents caused by drowsy driving, et al., 2023; Priya & Saranya, 2023). The integration
monitoring the driver’s level of fatigue is essential, and of AIoT emerges as a revolutionary force for enhanc-
technological advancements like driver monitoring sys- ing transportation safety. It addresses the multifaceted
tems with eye blink sensors are deployed to detect signs challenges of population growth, traffic congestion,
of drowsiness and provide timely warnings to drivers environmental sustainability, and safety concerns that
(Kamran et al., 2019; Selvaraj & Umakanth, 2021). cities worldwide are grappling with.
Expanding on the endeavors to enhance road safety Large volumes of real time data can be gathered from
and transportation within smart cities using the more IoT devices embedded in different components of the
advanced applications of the Internet of Things (IoT) transportation infrastructure thanks to AIoT. City plan-
(Bibri, 2018, 2020) and Artificial Intelligence of Things ners and authorities can optimize traffic flow, control
(AIoT) (Cui & Lei, 2023; Gong et al., 2023; Priya & congestion, and improve overall transportation effi-
Saranya, 2023)). Alam et al. (2019) propose a vehicle ciency, leveraging the actionable insights derived from
safety framework integrating various sensors to detect AI driven data analysis for informed decision making.
driver drowsiness, speeding, and alcohol consumption, AIoT makes it easier to analyze past and current data to
thus mitigating the risk of road accidents. The system forecast traffic trends, spot bottlenecks, and proactively
includes an eye blink sensor installed in the driver’s manage traffic flows by utilizing advanced AI algorithms,
temporary glasses to detect drowsiness and trigger especially Machine Learning (ML) and Deep Learn-
alarms when necessary (Alam et al., 2019). Moreover, ing (DL), and predictive modeling. This capability plays
ensuring road safety extends beyond individual vehi- a pivotal role in averting traffic jams and guaranteeing
cles. Calculating and displaying the speed of other a more efficient transportation network across the city
vehicles in front and behind is of equal importance to (Siddiqui et al., 2021).
avoid collisions. Ultrasonic sensors emitting ultrasonic The development of intelligent infrastructure that
waves are utilized to determine the distance between adapts dynamically to shifting traffic conditions is made
vehicles and facilitate a safer driving experience (Odat possible by AIoT. For example, smart traffic lights can
et al., 2017). To combat the prevalence of drunk driv- optimize signal control and reduce delays by adjusting
ing, smart city initiatives incorporate alcohol sensors, timings based on real time traffic data.
like the MQ3 alcohol sensor, coupled with an Arduino By providing predictive maintenance for vehicles, opti-
controller. The system ensures that a driver exceed- mizing routes based on demand patterns, and boosting
ing the legal alcohol limit is prevented from starting the general dependability and efficiency of public transit
the vehicle’s motor, while also alerting authorities for systems (Alahi et al., 2023), AIoT aids in improving the
timely assistance (Prasad et al., 2022; Siva et al., 2022). services provided by public transportation. A smooth
Emphasizing seamless communication among vehicles, link between various forms of transportation, including
smart cities leverage Vehicle to vehicle (V2V) com- bicycles, buses, trains, and ride sharing services, is made
munication and the revolutionary Light Fidelity (Li-Fi) possible by the integration of AIoT. A more comprehen-
technology. V2V communication reduces traffic conges- sive and linked transportation network is made possible
tion and prevents accidents, while Li-Fi, uses the visible by this integration, providing residents with a variety of
light spectrum for data transmission, and offers secure effective travel options (Zhang et al., 2023).
and efficient communication between connected cars Through the implementation of intelligent surveil-
(Mamikandan et al., 2022). In a nutshell, smart cities lance systems, predictive analysis of accident prone
harness technological advancements to revolutionize regions, and real time traffic situation monitoring,
road safety. In this context, the integration of emerg- AIoT plays a critical role in improving safety. This
ing AIoT and related sensors and devices for enhancing preventative measure helps to lower accident rates
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 3 of 18

and guarantee commuter safety (Wu et al., 2020). By Based on these objectives the following research ques-
encouraging electric and hybrid vehicles, advocating tions can be formu lated to guide this study.
traffic patterns that minimize emissions, and support-
ing projects for sustainable urban transportation, AIoT • How can eye blink sensors be effectively integrated
can help bring environmentally friendly practices into into a vehicle safety framework to detect driver
reality. AIoT can help bring environmentally friendly drowsiness and alert the driver promptly?
practices into reality. AI and AIoT offer a multitude of • How can ultrasonic sensors be deployed to measure
benefits in transportation services and infrastructure, the speed of surrounding vehicles and provide real
as evidenced by numerous studies. (e.g., Abduljabbar time information to enhance traffic flow within smart
et al., 2019; Iyer, 2021). city environments?
The justification for this study lies in the urgent need • How can an alcohol sensor be efficiently incorpo-
to address the escalating challenge of road accidents rated into the framework to detect the driver’s alco-
within smart cities. Despite advancements in technol- hol levels and limit the vehicle’s speed automatically
ogy and various safety measures, the frequency of fatal when high alcohol content is detected?
accidents continues to rise, posing significant risks to • How does the utilization of Li-Fi technology opti-
public safety and well-being. mize inter-vehicle communication within smart cit-
Understanding the potential of AIoT in enhancing ies, enabling faster and more efficient data transmis-
transportation safety offers a compelling avenue for sion via VLC using LEDs?
mitigating these risks and improving urban mobility. • How does the AIoT implementation of the proposed
Against the backdrop outlined above, this study aims vehicle safety framework enhance road safety within
to enhance next generation smart city transporta- smart cities, proactively mitigating road risks and
tion safety by developing a Li-Fi enabled vehicle safety improving transportation safety for all road users?
framework. This framework integrates various sensors,
such as eye blink, ultrasonic, and alcohol sensors, to In light of the research objectives drawn from existing
detect and address critical issues in daily transportation literature and the research questions arising from identi-
within smart cities, including driver drowsiness, speed- fied gaps, we are driven to address the challenges within
ing, and alcohol consumption, thereby mitigating the this domain. The significant contributions of this study
risk of fatal road accidents. To guide our investigation, are outlined as follows:
we undertake the following objectives:
• Development of an AIoT enabled comprehensive
• To design and implement an AIoT based vehicle vehicle safety framework that integrates multiple
safety framework tailored for smart cities, integrat- sensors to detect driver drowsiness, speeding, and
ing eye blink sensors for prompt detection of driver alcohol consumption, effectively mitigating the risk
drowsiness and alerting the driver using a buzzer. of road accidents.
• To deploy ultrasonic sensors to measure the speed • Implementation of an eye blink sensor in the driv-
of surrounding vehicles and provide real time infor- er’s temporary spectacles, allowing for the detec-
mation to the driver, contributing to the seamless tion of driver drowsiness and providing timely alerts
flow of traffic within smart city environments. through a buzzer to prevent accidents.
• To incorporate an alcohol sensor that efficiently • Integration and configuration of ultrasonic sensors
detects the driver’s alcohol levels and, when high to accurately measure the speed of surrounding vehi-
alcohol content is detected, automatically limits the cles, enhancing collision avoidance capabilities.
vehicle’s, ensuring responsible driving practices and • Detects the driver’s alcohol consumption limits the
alerting concerned authorities for swift action. vehicle’s speed automatically, and updates the infor-
• To optimize inter-vehicle communication within mation to authorities.
smart cities through the utilization of Li-Fi tech- • Application of innovative Li-Fi technology for com-
nology, facilitating faster and more efficient data munication between connected vehicles, enabling
transmission via visible light communication (VLC) faster and more efficient data transmission through
using light emitting diodes (LEDs), and fostering VLC, thereby improving overall road safety.
enhanced connectivity between connected vehicles.
• To proactively mitigate road risks, enhance overall The study is structured as follows: Section 2 shows
transportation safety and create a sustainable and the conceptual background definition of the key terms
intelligent urban environment where smart vehicles involved in the smart city research platforms. Section 4
coexist harmoniously within smart cities. provides an overview of related works, presenting the
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 4 of 18

existing literature and research relevant to the proposed • Technological integration: Smart cities heavily inte-
framework. It outlines the background information, iden- grate information and communication technologies
tifies the need for such a framework, and addresses gaps (ICT) into their infrastructure, services, and govern-
in the current literature. Section 5 describes the pro- ance processes.
posed autonomous speed detection and accident avoid- • AIoT: They apply this technological framework that
ance framework in detail. Section 6 presents the results seamlessly integrates AI models and algorithms into
of the framework’s implementation and its effectiveness IoT networked sensors and devices. This integration
in enhancing transportation safety. It outlines the meth- holds the transformative potential to fundamentally
odology used in the experiments and discusses the results reshape how we perceive, interact with, and manage
obtained from the developed hardware module. Finally, transportation systems in their entirety.
Section 7 concludes and summarizes the key findings and • Data analytics: They leverage the vast amount of data
contributions of the proposed framework. generated via IoT sensors and devices and the pro-
cessing and analytical capabilities of AI to generate
2 Conceptual definitions meaningful insights for enhanced decision making
This section delves into the foundational understand- processes.
ing of smart cities, encompassing their defining charac- • Efficiency: They aim to enhance the efficiency of urban
teristics, inception, and the driving forces behind their operations, infrastructure, and services, resulting in
emergence. It offers an insightful overview of the con- optimized resource utilization, improved service deliv-
ceptual landscape of smart cities, laying the groundwork ery, and reduced environmental impacts.
for a more in depth examination of their pivotal role in • Resilience: They prepare for and respond to challenges
advancing transportation safety through AIoT. such as natural disasters and climate change by build-
The concept of smart cities has attracted considerable ing resilient infrastructures and systems.
attention as a promising solution for addressing urbani- • Livability: They improve the quality of life for their
zation and sustainability challenges. Multiple efforts have residents by creating safe, comfortable, functional, and
been made to define smart cities, resulting in a variety community oriented urban environments.
of interpretations and developmental directions (e.g., • Smart Transportation: They integrate advanced tech-
(Singh & Singla, 2021; Toli & Murtagh, 2020)) 2020 has nologies and data driven solutions to optimize and
evolved over the last two decades, with its roots tracing enhance the efficiency, safety, and sustainability of
back to the 1970s, and it was not until 2010 that it gained transportation systems.
significant attention with the support of the European • Li-Fi enabled vehicles: They enable high speed wireless
Union for smart city projects (Bibri, 2019). In essence, a communication between vehicles and infrastructure
smart city incorporates state of the art technologies and through visible light signals that enhance vehicle to
data driven approaches to optimize urban operations, vehicle and vehicle to infrastructure connectivity.
enhance resilience, improve quality of life, and foster • Light emitting diodes (LEDs): They embed energy effi-
sustainability (Mishra et al., 2022). However, the inclina- cient and sustainable illumination in streets, traffic sig-
tion of smart city initiatives to prioritize economic gains nals, and public spaces. They also enable cost savings,
over environmental and social aspects (Ahvenniemi and longer lifespan, with the ability to integrate with
et al., 2017; Evans et al., 2019; Toli & Murtagh, 2020) has smart systems for adaptive lighting control, enhanc-
emphasized the necessity for greater focus on research ing safety and reducing energy consumption in urban
and practical endeavors aimed at achieving the environ- transportation.
mental and social objectives of sustainable development
(Sharifi et al., 2024). In this evolving landscape, smart Worth noting is that defining a smart city precisely
cities are increasingly embracing and leveraging AI and remains challenging, as there is no universally agreed
AIoT to tackle complex challenges (Bibri et al., 2023a, b; upon definition. Different perspectives and objectives
Efthymiou & Egleton, 2023; Gourisaria et al., 2022; Zaidi lead to varying definitions and approaches to smart
et al., 2023), with related solutions emerging as a pivotal city development (e.g., (Singh & Singla, 2021)). Regard-
force in shaping their development practices and trajec- less, smart cities are at the forefront of urban innova-
tories. Among the core features of a smart city relevant to tion and actively seeking ways to address the challenges
this study include: of urbanization while improving the well-being of their
residents.
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 5 of 18

3 A novel conceptual framework: artificial strategically positioned within the transportation net-
intelligence of smart city things work process sensor data in near real time.
in transportation
The integration of AI with IoT has given rise to a trans- 3.4 Cloud computing
formative paradigm known as AIoT. This convergence Cloud computing provides scalable and centralized
finds numerous applications and profound implica- data storage and processing capabilities for handling
tions, especially in smart city domains, including large datasets generated by smart transportation sys-
transportation and its safety (Alahi et al., 2023; Bibri tems. Cloud servers support intensive data analytics and
& Jagatheesaperumal, 2023; Bibri et al., 2023a; Puri machine learning tasks, contributing to the overall intel-
et al., 2020; Zhang & Tao, 2020). The conceptual frame- ligence of the system, including safety related predictive
work presented here aims to elucidate the technical analytics.
components of integrating AIoT within the realm of
smart cities, with a specific focus on enhancing smart 3.5 Data analytics and ML
transportation systems, including safety considera- Centralized data analytics platforms employ ML and DL
tions. It is crucial to note that this study is one of the algorithms to analyze vast datasets. Predictive analytics
initial endeavors to introduce the conceptual frame- models not only forecast traffic congestion but also iden-
work of Artificial Intelligence of Smart City Things in tify potential safety hazards and patterns, contributing to
the context of transportation safety. The underlying the efficiency and safety of smart transportation.
technical components are derived based on two studies
conducted by (Bibri & Jagatheesaperumal, 2023; Bibri 3.6 Data driven decision making systems
et al., 2024). AI driven decision making systems interpret the analyzed
Sensing devices and data acquisition: Smart transporta- data to make intelligent decisions. These systems can
tion begins with a dense network of sensing devices, such dynamically adjust traffic signal timings, reroute vehicles
as cameras, sensors, and GPS units, embedded in the based on real time conditions, optimize public transpor-
urban environment and transportation infrastructure. tation schedules, and prioritize safety measures.
These devices continuously collect real time data on traf-
fic patterns, vehicle movements, environmental condi- 3.7 Safety monitoring and response
tions, and safety related events. Dedicated AI algorithms monitor safety related events,
such as accidents or abnormal driving behavior, trigger-
ing immediate responses. Automated safety measures,
3.1 Communication infrastructure such as real time alerts to authorities or adaptive speed
A robust communication infrastructure, including low controls, enhance overall transportation safety.
latency networks and protocols, facilitates seamless data
transmission between edge devices and centralized sys- 3.8 User interfaces and applications
tems. Communication technologies like 5G enhance the User interfaces and applications provide a human centric
responsiveness and reliability of smart transportation interaction layer for both administrators and the public.
systems, ensuring timely safety related communication. Mobile apps, dashboards, and public displays offer real
time information on transportation status, alternative
routes, and safety advisories.
3.2 Edge computing The technical components of AIoT framework are dis-
Edge computing plays a pivotal role in processing data tilled based on the extant literature on AIoT architectures
closer to the source, reducing latency and enabling real (Bibri et al., 2024; Seng et al., 2022; Yang et al., 2020a,
time decision making. Edge nodes are strategically posi- 2020b; Zhang & Tao, 2020). These embody a powerful
tioned within the transportation network, process- framework that seamlessly integrates the computational
ing sensor data, extracting relevant information, and and analytical power of AI with the ubiquitous connec-
addressing safety related concerns. tivity of, and the abundant data generated by, IoT. In the
context of transportation systems and their safety, the
role of edge computing particularly becomes apparent
3.3 Fog computing within the framework of AIoT. Specifically, the impor-
Fog computing brings computation and data storage tance of edge computing lies in facilitating the relocation
closer to the edge, enhancing the efficiency of data pro- of data processing closer to the network edge, optimiz-
cessing for time sensitive applications in smart trans- ing proximity to the IoT devices associated with trans-
portation, including safety monitoring. Fog nodes portation systems. This strategic placement enhances the
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 6 of 18

efficiency of AIoT driven transportation systems by lev- and the engine will slow down. However, there may be
eraging edge intelligence, a crucial aspect for implement- errors in the optimization process of the high level com-
ing cutting edge AIoT solutions aimed at the planning of puter languages used to process the images for drowsi-
safe and intelligent transportation within smart cities. ness detection. Additionally, the use of sensors may cause
The conceptual framework outlined above positions variations in precision. To enhance the system’s effi-
AIoT as a pivotal enabler of smart transportation and ciency, incorporating speed detectors and other detec-
sustainable mobility, with a strong emphasis on enhanc- tors may be necessary.
ing safety. By integrating cutting edge technologies, such Gowri et al. (2019) have developed a system that
as AI, ML, DL, IoT, edge computing, fog computing, detects accidents and drowsiness. The system includes
and cloud computing, and advanced communication, four sensors: a vibration sensor, an eye blink sensor, and
this framework establishes a foundation for responsive two heart rate sensors. These sensors work together to
and proactive decision making based on real time data detect mishaps. In addition, the system incorporates
analysis, addressing not only traffic flow but also safety GSM to send a message and GPS to inform the driv-
concerns. The integration of the components of AIoT er’s relative in case of an accident or unusual situation.
optimizes transportation efficiency as well as enhances However, the heart rate sensor’s occasional inaccurate
the overall safety experience for drivers. measurements may affect the data’s average and cause
Moreover, the conceptual framework draws on exist- problems.
ing literature to validate its technical components. Kinage et al. (2019) have proposed a system that pre-
The integration of computing models aligns with the vents car collisions using various components, including
decentralized and distributed processing trend in AIoT an Arduino, MQ sensor, infrared sensor, accelerometer,
architectures, while the emphasis on communication and webcam. The MQ3 sensor detects alcohol intake,
infrastructure and 5G reflects current research on the while the infrared sensors monitor the driver’s eyes to
connectivity needs of advanced smart city applications. detect drowsiness. The accelerometer tracks head move-
Additionally, the application of predictive analytics in ments, and the OpenCV software counts the number of
transportation, including safety related events, mirrors mouth yawns. If the number of observed yawns exceeds
recent advancements in AI driven models for traffic pre- a threshold, the driver receives a voice message. Fur-
diction and safety monitoring. The conceptual framework thermore, the system uses an ultrasonic sensor to detect
presented lays the groundwork for the implementation speed bumps on the road. This approach provides a prac-
of AIoT in smart transportation beyond this study, con- tical solution to the increasing number of accidents.
tributing to the broader discourse on the groundbreaking Khan et al. (2019) have developed an Intelligent
convergence of AI and IoT in the context of smart cities, Autonomous Accident Prevention, Detection, and Vehi-
with a focus on enhancing transportation safety. cle Monitoring System to prevent accidents. The system
includes a black box and a post accident rescue system.
4 Related works Ultrasonic sensors are used to detect vehicles and obsta-
In recent years, several studies have been conducted to cles, and the system is notified accordingly. Gas sensors
investigate autonomous speed detection and accident detect engine performance, carbon monoxide, smoke,
avoidance frameworks for connected vehicles. In this and compressed gas leaks and transmit the information
section, we review the related works in this area, with a to a display. The accelerometer monitor detects acci-
focus on Li-Fi enabled vehicle frameworks. We highlight dents. GSM and GPS modules are used to locate the
the advantages and limitations of existing systems and precise location for emergency call instructions. Addi-
identify the research gaps that need to be addressed in tionally, the black box records the vehicle’s speed, posi-
future studies. tion, and time continuously, and the data is stored on a
Kabir et al. (2020) has developed a smart car manage- memory card.
ment system that aims to reduce traffic accidents. The Lim et al. (2018) have demonstrated and simulated
system consists of three key components: an overload the use of ultrasonic sensors to detect obstacles. These
detector, an alcohol detector, and a sleep detector. If the sensors are affordable and easy to replicate. However,
sleep detector detects drowsiness in the driver during to enhance detection precision, multi sensor imple-
the drive, the engine will slow down, and the buzzer will mentation is essential. Additionally, weaknesses in
beep. The alcohol monitor determines the driver’s intoxi- other instruments must be analyzed, and the vulner-
cation level, and if it exceeds the threshold, a message is abilities of autonomous features should be identified to
sent to the control center, and the vehicle will not start. strengthen the system. Suryana et al. (2019) developed
The overload detector senses the car’s weight, and if the an automatic braking device for motorcycles. The sys-
passenger weight exceeds the limit, an alert will sound, tem uses an ultrasonic sensor to detect objects, and the
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 7 of 18

braking system is implemented automatically using the An intelligent transportation system for accident iden-
principles of physical laws, specifically kinematic non tification and prevention is offered by (Selvathi et al.,
uniform slowing motion. The servo function moves the 2017). The protection system includes both helmet
brake handle, enabling the system to operate autono- and alcohol detection components. The relay is trig-
mously based on the variable speed and the distance of gered when the helmet is detected and the alcohol level
the motorcycle from the object within the range of 0 m is below the threshold. Air pressure is used to locate the
to 4 m. This provides the motorcycle with ample time helmet, while an MQ3 sensor monitors the alcohol level.
to stop before colliding with the object in front of it. The alcohol content is displayed on the LCD screen. The
The use of Arduino for seat belt management and vehicle’s accelerometer is utilized to monitor acceleration
alcohol detection was suggested by (Malathi et al. in all three axes and detect accidents. If an accident is
2017). This method prohibits drivers from operat- detected, the data is transmitted via Bluetooth module to
ing a vehicle while intoxicated and/or without a seat a connected smartphone, which then sends a notification
belt. The output from the infrared and alcohol sensors to a pre set emergency contact.
is compared by the comparator and then sent to the Janjua et al. (2021) developed a system for truly con-
microprocessor. If the result is valid, the relay is acti- nected vehicles with Li-Fi communications for smart cit-
vated, allowing the vehicle to start. If the comparator ies. They discussed the advantages and disadvantages of
output is false, the GSM modem connected to GPS Li-Fi technology. Although Li-Fi has some drawbacks,
sends information about the location and a message the authors provide a detailed analysis of its benefits as
stating that the driver has not fastened their seat belt a communication medium for vehicles and traffic infra-
or has consumed alcohol. Plans for this system include structure interactions. Table 1 summarizes the inferences
sending notifications to nearby police stations. and challenges observed from recent popular literature.
Alcohol detection for vehicle locking devices was created The study (Yuvaraju et al., 2021) presents an implemen-
by (Al-Youif et al., 2018). The device has demonstrated tation architecture for integrating Li-Fi communication
its capability to detect and measure the concentration between vehicles and traffic infrastructure, including a
of a gas. In case it detects that the concentration of a gas ”Transmitter” and ”Receiver” model that enables secure
exceeds a specific threshold that is not allowed, it can con- wireless communication between vehicles. Addition-
trol the ignition of the car. While this is only a prototype, it ally, the study illustrates a VLC system that uses Li-Fi
can be utilized as a basis for other innovative systems. The transmitters and receivers to transmit data. Here, a LI-FI
relay can be used to regulate the ignition of the car, and an based collision avoidance system uses a visible light
alarm system can be installed on the device. transmission to prevent collisions between vehicles, as
The device, which has features for speed tracking, alco- radio frequencies can cause electromagnetic interference
hol detection, over speeding, reckless driving, driving and are restricted in places like hospitals, airplanes, and
without a license, and seat belt checking, was created power plants. The VLC system utilizes Li-Fi transmitters
by (Jacob et al. 2020). If any of these conditions are vio- and receivers in cars and can be integrated with future
lated, the controller sends emergency data to the cloud, LED digital matrix headlamps for faster communication.
where the RTO receives updates about vehicles that are This technology is expected to produce the fastest and
not complying with the regulations. The next step is to most sophisticated wireless communication system for
establish a coordinate and mapping system for the driv- long distance communication, with a capacity of 10 Gbps
er’s location, which will enable GPS to determine the type at a low cost.
of road the vehicle is traveling on. The study in (Chen et al. 2021) presents a data driven
A controller for driving while intoxicated was devel- AI solution using a neural network to predict pre crash
oped for cars by (Manikandan et al., 2021). This control- conditions of cars in accidents. The approach utilizes
ler detects intoxication by analyzing gaseous substances plastic deformation patterns to accurately reconstruct
in the driver’s breath. If the sensor output detects that the initial collision parameters, demonstrating its potential
blood alcohol level has exceeded the legal limit, the vehi- for crash analysis and engineering failure assessment,
cle’s motor will not start. The alcohol sensors used in this with relevance to autonomous vehicles. In (Xie et al.,
system are highly durable and can detect alcohol from a 2022), the authors introduce a novel algorithm using
distance of up to two meters. In the event of a violation, variational Bayesian learning and a neural network to
the police will receive the car’s location through the GSM efficiently reconstruct a crashed car’s deformation field
module. Additionally, this system can remotely lock the and pre crash state from its damaged structure, aiding in
doors and identify the car that was being driven while 3D collision analysis and forensic assessment. It achieves
intoxicated. accurate results with fewer iterations, highlighting its
Table 1 Summary of Inferences and Challenges from Recent Popular Related Works
Ref Year Application Technique/ Inferences Challenges
Technology

Gowri et al. (2019) 2019 Drowsiness Eye flicker A vibration sensor Computational
and Accident Detection System sensor, vibration sensor, beat rate sensor and a heartbeat rate sensor are used, capacity to process raw input. For improved
Jagatheesaperumal et al. Computational Urban Science

as well as an eye flicker sensor to deter- accuracy, high frequency and continuous-
mine the driver’s sleepiness sensor readings are required
Kabir et al. (2020) 2020 Smart Raspberry Have provided safety Analyzes the images
Vehicle Management System Picam issues by forming a drowsiness detec- for the drowsiness detector, but opti-
eraEye Movement Detection, Pressure tor, an alcohol detector, and a load cell mization mistakes can still happen. The
(2024) 4:10

sensor, andGas Sensor was used to maintain a normal load for any accuracy of detection is heavily dependent
particular vehicle on the environment, and camera quality
Sathasiv et al. (2020) am2020 Image Eye The Raspberry Pi Drowsiness detec
detection drowsiness system Aspect Ratio (EAR) camera, Raspberry Pi 4, and GPS module tion can be improved further by calculating
technique are used to identify and analyze the state yawn detection, and face orientation rather
of eye closure in real time than just relying on the eye closure state
Mateen et al. (2022) 2022 Accident IR sensor, Following the detec Test results may
detection and alert system Microphone Sensor, Smoke Sensor tion of an accident, drivers of oncoming differ from those expected in the opera-
vehicles are alerted by blinking lights tional environment due to differences
and a siren between the lab environment and the actual
road environment
Perumal et al. (2022) 2022 Intelligent VNet and Capable of correctly Requires high pro
Overtaking Advice System TTCNet calculating the velocity of the lead car, cessing time and needs the training to oper-
neural networks generating advice for drivers ate
Page 8 of 18
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 9 of 18

potential for car crash investigation and crashworthiness accidents. Several sensors are integrated into the system
evaluation. to help avoid accidents.
These studies encompass a range of approaches, This section is divided into five subsections: Section 3.1
including the development of smart car management discusses the overview of the proposed system; Sec-
systems incorporating overload, alcohol, and sleep detec- tion 3.2 describes the drowsiness detection module that
tors (Kabir et al., 2020). Other systems feature sensors for prevents accidents caused by drowsiness; Section 3.3
accident detection (Gowri et al., 2019; Jacob et al., 2020; explains the speed indication module that displays the
Khan & Howlader, 2019; Manikandan et al., 2021; Sel- speed of nearby vehicles; Section 3.4 discusses the alco-
vathi et al., 2017), collision prevention using Arduino and hol detection module with motor locking system and the
various sensors (Kinage & Patil, 2019), obstacle detection vehicle to vehicle communication using Li-Fi technology
using ultrasonic sensors (Lim et al., 2018), and automatic is explained in Section 3.5.
braking for motorcycles (Suryana et al.). Additionally,
there are proposals for alcohol detection and ignition 5.1 Overview
control (Al-Youif et al., 2018; Malathi et al., 2017), as The proposed system comprises three distinct modules:
well as an intelligent transportation system for accident the drowsiness detection module, speed indication mod-
identification and prevention (Selvathi et al., 2017). The ule, and alcohol sensing module, as illustrated in Fig. 1.
authors also discuss the advantages and disadvantages The drowsiness detection module is equipped with a
of Li-Fi technology in the context of vehicle and traf- temporary glass mounted eye blink sensor that detects
fic infrastructure interactions (Janjua et al., 2021). These the driver’s level of fatigue. In case of drowsiness, the
prior works collectively inform the motivation and direc- module activates a buzzer to alert and wake up the driver.
tion of the present research, contributing to the devel- The speed indication module utilizes ultrasonic sensors
opment of innovative solutions for enhanced smart city that are mounted on the sides of the vehicle to measure
transportation safety. the distance and speed between the vehicle and other
objects. These measurements are then displayed on our
system. The alcohol sensing module continuously detects
5 System models and methods alcohol and alerts relevant personnel if alcohol is detected
In this section, autonomous speed detection and acci- in the driver’s breath. A GPS and GSM MODEM are used
dent avoidance frameworks for connected vehicles are to detect the driver’s location and inform the concerned
discussed. Dynamic transport systems have been devel- authorities if the driver is found to be under the influence
oped as a result of technological advancement, and of alcohol. The system then proceeds to lock the engine
they unquestionably have had a significant influence on to prevent the driver from operating the vehicle.
society and people’s daily lives. Despite having count-
less advantages, it has brought about a lot of negative 5.2 AIoT
situations. The most harmful aspect of vehicles, in addi- AIoT makes predictive maintenance possible by using
tion to contaminating and crowding the city, is vehicular IoT sensors to continuously monitor vehicle status. This
proactive maintenance strategy lowers downtime, helps

Fig. 1 Block diagram of the proposed AIoT enabled intelligent transportation system for Smart Cities
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 10 of 18

avoid malfunctions, and guarantees the dependability transportation, streamlining traffic, and assisting in the
of fleets of vehicles. By gathering data in real time from overall decrease of carbon emissions (Hammad et al.,
IoT devices integrated into infrastructure, cars, and 2023), AIoT helps lessen the negative environmental
roads, AIoT algorithms can forecast traffic jams, iden- effects of transportation. AIoT incorporates cutting edge
tify patterns in traffic, and dynamically modify traffic safety technologies like intelligent surveillance, real time
signals. Smoother traffic flow and shorter travel times vehicle status monitoring, and predictive analysis of acci-
are the results of this [Link] (Soomro et al., dent prone locations. As a result, both pedestrians and
2018). By offering real time data on passenger demand, commuters benefit from a safer transportation ecosystem
optimizing routes, and dynamically modifying sched- (Wu et al., 2020).
ules, AIoT improves public transportation systems. This
flexibility increases public transportation’s dependabil-
ity and efficiency, promoting usage and easing traffic 5.3 Drowsiness detection module
(Alahi et al., 2023). The proposed structure, as shown in Fig. 2, utilizes an
The development of connected and autonomous auto- Arduino mini for its implementation. The system consists
mobiles is greatly aided by AIoT. AIoT makes autono- of a temporary glass that a driver can wear while driving,
mous driving safer and more effective by facilitating with the necessary hardware attached to it. The hardware
constant communication and data sharing between vehi- includes an IR sensor, a buzzer, a switch, a 3.7 V battery,
cles, traffic signals, and infrastructure (Wu, 2022). AIoT and an Arduino Mini. We have taken great care to pre-
makes it possible to provide customized and situation vent any obstruction to the drivers’ line of sight while
specific services for travelers. AIoT improves the general driving. It ensures that it does not obstruct the drivers’
user experience and ease of transportation, from smart view, allowing them to have a clear and unobstructed
parking solutions to customized route recommendations view of the road ahead. The IR sensor is used to detect
(Darmawan et al., 2021). City planners are aided in their when the driver’s eyes are closed. The sensor is com-
decision making on infrastructure development, tran- prised of an emitter and a receiver that together form the
sit policies, and urban design by the abundance of data infrared sensor. The IR emitter transmits IR light toward
produced by AIoT devices. More sustainable and livable the driver’s eye, and the IR photodiode detects whether
cities are produced as a result of this data driven strat- the same wavelength light is reflected. When the driver’s
egy (Son et al., 2023). By offering real time visibility into eye is closed, the reflected IR light is stronger, which is
the movement of products, AIoT improves supply chain detected by the photodiode. Conversely, when the driv-
management and logistics in the field of freight transpor- er’s eye is open, the reflected IR light scatters and has a
tation. Better inventory control, fewer delays, and overall reduced intensity. Therefore, eye closure action can be
efficiency are the results (Nozari et al., 2023). By encour- detected by monitoring photodiodes.
aging the adoption of environmentally friendly means of

Fig. 2 AIoT enabled drowsiness detection module

Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 11 of 18

The configuration of the Arduino Mini requires the bounce back after striking nearby objects. By measur-
use of an Arduino Uno board, which is used to program ing the time taken for the waves to return, the distance
the Mini. To program the Mini, the IC in the Arduino between the vehicles can be calculated. The ultrasonic
Uno board is first removed. Once this is done, the con- sensors are placed on the sides of the car, and in the event
nections are established, and the Mini is updated with of a vehicle attempting to pass us on either side, the sen-
the appropriate program. The IR sensor that is used in sors detect the movement and notify us of the direction
the proposed system continuously emits infrared radia- from which the other vehicle is approaching. The sys-
tion, which is directed toward the driver’s eye. When tem is also capable of displaying a message if no vehicle
the driver blinks, the sensor detects the change in the is detected nearby. Moreover, the system is designed to
output, which is then sent to the Arduino Mini. The track the movement of every car in the vicinity of our
Arduino Mini processes this data and activates the vehicle, providing the driver with real time information
buzzer if the driver doesn’t blink for a certain prede- on the surrounding traffic conditions. This information
termined period. The buzzer is used to alert the driver can be displayed on the system’s screen, enabling the
and wake them up, preventing potential accidents due to driver to make informed decisions while driving.
drowsiness or fatigue.
5.5 Alcohol detection module
5.4 Speed indication module A system has been designed using an Arduino Mega
It is not sufficient to only display the speed of our vehi- microcontroller that incorporates an alcohol sensor or
cle while driving. It is equally important to determine MQ3 sensor for the detection of alcohol consumption,
the speed of the nearby vehicles both in front and behind as shown in Fig. 4. Drunk driving is a significant cause of
our vehicle, as this information can help prevent acci- accidents, and the system locks the motor when the alco-
dents. To accomplish this, the proposed system utilizes hol sensor detects alcohol in the breath of the driver. To
an ultrasonic sensor that is attached to the sides of the ensure the safety of the driver and others on the road, the
vehicle. The ultrasonic sensor is capable of measuring the system sends location messages to authorized individuals
distance between our vehicle and the neighboring cars, via a GPS receiver and GSM modem. The system com-
and based on this information, the system can calculate prises an Arduino Mega board, an MQ3 alcohol monitor,
the speed of these vehicles. By displaying this informa- a GSM/GPS module, and a notification device.
tion to the driver, they can adjust the speed of their vehi- The MQ3 alcohol sensor can detect the amount of alco-
cle accordingly to maintain a safe distance from other hol in the air starting from 0.4 mg/L. When the alcohol
vehicles. The system is composed of an LCD, an Arduino sensor detects alcohol in the breath, it sends a signal to
Mega, and an ultrasonic sensor, as shown in Fig. 3. This the microcontroller, which is connected to an ADC and
setup enables the system to accurately calculate and dis- transmits the data. The signal is then converted to digi-
play the speed of neighboring vehicles in real time, pro- tal code, and the value is compared with a predefined
viding the driver with critical information to help them threshold. If the blood alcohol content exceeds the
make informed driving decisions. 0.4 mg/L threshold, the car’s speed will be limited, if it
In the proposed system, ultrasonic sensors are used to detects during running condition or the engine will be
accurately measure the distance between the vehicles in disabled during the idle condition.
front, behind, and on the sides of our car. The ultrasonic During driving, the GPS constantly calculates the loca-
sensors emit high frequency sound waves, which then tion and speed of the vehicle, and this information is

Fig. 3 AIoT enabled speed indication module

Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 12 of 18

Fig. 4 AIoT enabled alcohol detection module

sent to the Arduino Mega microcontroller. If the vehi- enabling real time communication and collaboration
cle’s speed exceeds the predetermined limit, the Arduino among them.
Mega sends a command to the GSM module, which Li-Fi, being a relatively novel technology, does present
transmits a message to the appropriate authorities con- challenges in signal quality, especially in dynamic and
taining the vehicle’s latitude and longitude. This feature congested traffic environments. Li-Fi has evolved sig-
ensures that authorities can take appropriate action and nificantly in recent years, demonstrating its effectiveness,
prevent accidents caused by reckless driving. security, and ability to transmit data at very high rates
To effectively address concerns regarding the utiliza- (Memedi & Dressler, 2020). It offers several advantages
tion of Arduino and GSM technology within the pro- over traditional wireless communication technologies,
posed framework for v2V communication, it exhibits a such as WiFi and cellular networks. Li-Fi technology is
profound comprehension of the significance of indus- less prone to interference, more secure, and better suited
try standards, reliability, and safety in such applications. to high density environments, such as urban areas. It is
It offers a well justified rationale for their inclusion, a cost-effective, high data rate, and bandwidth efficient
strengthening the framework’s credibility. One notable solution for vehicular communication that has gained
strength lies in the rationale behind opting for Arduino considerable attention in recent years. Li-Fi technology
and GSM technology, as it provides distinct advantages employs light as a medium for communication, allowing
within the context of enhancing smart city transporta- for faster and more secure data transmission. VLC is the
tion safety. Moreover, the thorough consideration of reli- underlying technology that enables Li-Fi to eliminate the
ability and safety concerns not only acknowledges these complexity of cable communication.
apprehensions but actively endeavors to mitigate them. Researchers and experts in the field of optical commu-
It achieves this by elaborating on how Arduino and GSM nication and traffic management have conducted empiri-
technology align with industry standards and safety pre- cal studies that can validate the practicality of Li-Fi for
requisites. This commitment to transparency and detail traffic safety applications (Dibaei et al., 2021). This col-
serves to underscore the robustness and dependability of laborative effort will help us gather valuable data and
the chosen technologies. insights to substantiate our claims and ensure that Li-Fi
can indeed contribute significantly to traffic safety. It pro-
5.6 Li‑Fi vehicle to vehicle communication vides a well founded and evidence based assessment of
Vehicle to vehicle (V2V) communication technology has Li-Fi’s effectiveness in improving traffic safety within the
emerged as a game changer for intelligent transporta- smart city context.
tion systems, helping to prevent accidents and improve The degree of integration of the setup in real time
safety on the roads (Chhabra et al., 2022). This technol- determines the anticipated costs of each system compo-
ogy leverages various in-vehicle sensors to generate and nent, such as sensors, processors, communication mod-
transmit messages of actions between vehicles, thereby ules, and power management units. Researchers and
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 13 of 18

practitioners may better understand the financial impli- an alarm or alert system to notify the driver of their
cations of deploying comparable solutions and make drowsiness.
resource allocation decisions by quantifying the costs If the outcome from the eye flicker monitor is mini-
of each component based on the deployment scenario. mal, the indication will be adjusted to high. If the driver
Numerous aspects, like sensor precision, processor begins to nod off, the buzzer will sound as a warning, and
power, and communication range for real world applica- the indication can be turned off if necessary. To improve
tions, influence cost variability even with the prototype blink recognition and minimize the likelihood of false
model. For example, sophisticated signal processing positive blink events resulting from non blink lid move-
algorithms and higher resolution sensors may be more ment, it is possible to include one or more extra emit-
expensive, but they also provide better performance and ters and/or detectors. This additional equipment would
dependability. Comparably, the choice of transmission augment the capabilities of a single IR LED (emitter)
rates and communication protocols might affect hard- and phototransistor (detector), enhancing the system’s
ware prices, where providing faster V2V communication accuracy. An array of two or three emitter/detector pairs
technologies is more expensive. positioned parallel to the eye surface may provide an
The primary objective of V2V communication is to pre- alternative path for IR rays to travel when the lids move
vent accidents by having moving vehicles transmit their up and down simultaneously while remaining apart (for
positions and data rates via intelligent vehicular com- example, during a downward gaze). Furthermore, by
munication, as depicted in Fig. 5. The integration of Li-Fi incorporating a collection of emitter/detectors, the sens-
technology with V2V communication can significantly ing system’s resistance to changes in eyeglass position
enhance the safety, efficiency, and reliability of our trans- caused by head or facial movements that cause the skin
portation system, making it more intelligent and secure. on the nose’s bridge to move or by head movements may
be improved. The primary input source is the intercon-
6 Results and discussions nection between the wearable component, which tracks
A prototype instrument designed to detect sleepiness driver intoxication and sleepiness, and the vehicle’s
consists of a temporary glass with an installed IR sen- onboard systems. It is used to continually track impor-
sor and is powered by a 3.7 V battery that supports an tant physiological indicators including heart rate, eye
Arduino Pro Mini processor. Standard safety spectacles movements, and alcohol content of the breath. The cen-
are equipped with an infrared (IR) emitter and detec- tral processing unit of the vehicle receives these real time
tor pair, positioned laterally across the palpebral fissure data streams wirelessly, and using cutting edge machine
to generate a monitored IR stream that is interrupted learning techniques, it evaluates them to determine the
when the eyes close. This interruption is detected by the driver’s level of drunkenness and cognitive state. To
IR sensor and signals the processor, which then activates reduce possible safety concerns, relevant measures are

Fig. 5 AIoT enabled Vehicle to Vehicle Communication Module using Li-Fi

Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 14 of 18

initiated based on the analysis results, such as automated vehicles, we can interface three ultrasonic sensors on
speed decrease or audio alarms. the rear of our car. When the rightmost sensor detects
Subsequently, the collision avoidance system of the a vehicle overtaking on the right, the display will show
vehicle is integrated with the ultrasonic speed detector, the message”Alert, Vehicle overtakes at Right,” and when
which keeps track of the speed of another vehicle. Real the leftmost sensor is in use, the message”Alert, Vehicle
time situational awareness through ultrasonic sensors Overtakes at Left” will be displayed. If the moving vehicle
could be able to detect the relative velocities and dis- approaches our car too closely, a”Danger” warning will
tances from other surrounding vehicles. Moreover, the appear, while the message”No vehicle” will be displayed if
collision avoidance system uses V2V Li-Fi technology there are no nearby vehicles.
to connect with other cars, enabling cooperative moves To improve the accuracy of alcohol detection, we have
and improving traffic safety in smart city settings. These integrated an MQ3 sensor into the microcontroller. This
integrations could be made a reality by configuring the sensor is known to be reliable and widely used for alco-
decision making phase in the distributed AIoT integrated hol monitoring. Its detection range extends up to 2 m,
platforms. making it an efficient tool for detecting the presence of
The role of AIoT in the aforementioned scenarios facili- alcohol. Furthermore, the sensor’s sensitivity can be
tates smooth communication and coordination between adjusted to meet specific requirements, making it adapt-
various system components. AIoT ensures prompt reac- able to different scenarios. The MQ3 sensor’s conductiv-
tions to changing driving circumstances and new safety ity increases as the alcohol concentration rises, providing
concerns by enabling dynamic data exchange and deci- the Arduino with the corresponding reading.
sion making across heterogeneous devices and platforms The results of using an Alcohol Sensor to detect the
through a distributed network architecture. Additionally, presence of alcohol indicate that there is no alcohol
the system can continually adjust and enhance its perfor- detected in the driver’s exhaled air as the level is below
mance based on contextual information and real world 150. However, it should be noted that there may still be
input thanks to AIoT driven analytics and predictive an undetected alcohol level present. When the driver
modeling, which maximizes driving efficiency and safety. exhales, the sensor can detect a small amount of alcohol,
We may look to apply the suggested approach in a and if the level is above the legal limit, a”DRUNK” notifi-
variety of situations, including personal vehicles, public cation will appear on the screen. Conversely, if the level is
transit, and commercial fleets. For example, the technol- within legal limits, a message stating”Within legal limits”
ogy might be implemented in commercial fleets under will be displayed. The purpose of the system is to reduce
forced adoption situations to enforce safety standards the number of accidents caused by impaired driving by
and reduce the hazards related to driver fatigue and implementing an engine locking system. Table 2 pro-
alcohol impairment. On the other hand, in situations vides a comparative analysis of smart city transportation
where adoption is voluntary, the system may be sold as safety Systems, including drowsiness detection, alcohol
an optional accessory for private automobiles, attracting detection, and accident rates, for various safety systems,
customers who value safety and want to improve their including the proposed system and referenced systems.
driving experience. The DC motor will be turned off by Arduino when the
One can use an ultrasonic sensor to determine the reading goes beyond the threshold level. The L293D is
speed of a nearby vehicle. It consists of a transmitter and connected to Arduino and supplied with 5 V, which is
a receiver that emit and receive sound waves, respec- then linked to the DC motor. The SIM800A and NEO6M
tively. By transmitting high frequency sound waves and GPS modules are used by this model to send a mes-
monitoring the time it takes for them to return after hit- sage with the vehicle’s location to the registered cell-
ting an object, the ultrasonic sensor can measure the phone number after the engine is locked. GPS receivers
distance between the object and the sensor. This con- are capable of determining their distance from several
cept enables the measurement of an object’s motion as
it passes an ultrasonic sensor. When the item moves in Table 2 Comparative Analysis of Smart City Transportation
front of the sensor, the sound waves bounce back to the Safety Systems
sensor. The speed of the passing vehicles can be calcu-
Safety System Drowsiness Alcohol Accident
lated using the equation v = d / t, where v represents the Detection (%) Detection (%) Rate (per hour)
object’s velocity, d represents the distance between the
object and the sensor, and t represents the time it takes Proposed System 95 92 0.78
for the sound waves to reach the object and return. We Guede et al. (2019) 93 90 0.84
can determine and display the speed of passing vehi- Fujiwara et al. (2018) 94 91 0.81
cles by utilizing this technique. To detect overtaking Walizad et al. (2022) 92 93 0.79
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 15 of 18

satellites. The locations of the 32 GPS satellites have cell on the receiver side absorbs the light from the LED
already been pre programmed into them. The GPS mod- and converts it into electrical energy. The receiver then
ule is responsible for determining the vehicle’s speed decodes the binary data into real data and transmits it to
and location. Radio signals from the satellites are sent the Arduino board.
to Earth, carrying details about their positions and the
time at that moment. These signals identify the satel- 7 Conclusions and future research
lites and provide location information to the receiver. This study presented an innovative solution to enhance
The SIM800L module can send text SMS data to a host transportation safety in next generation smart cities. The
server after receiving serial data from radiation moni- Li-Fi enabled vehicle safety framework integrates eye
toring equipment like survey meters or area monitors. blink, ultrasonic, and alcohol sensors to address criti-
When the alcohol level crosses the threshold, the GSM cal issues like driver drowsiness, speeding, and alcohol
module’s function is to send an SMS to the appropriate consumption, major contributors to fatal road accidents.
authorities at the provided number. The GPS 8 M module The system detects driver drowsiness and alerts the
records and securely stores location data. The location is driver through a buzzer, preventing fatigue related acci-
stored on the Arduino Mega board as soon as it is known dents. Ultrasonic sensors provide real time information
because the initial fix takes longer to establish. on surrounding vehicle speeds, enhancing traffic flow
The complete system module developed in this work and reducing collision risks. The alcohol sensor detects
consists of two units: a transmitter unit (Fig. 6a) and alcohol consumption and restricts the vehicle’s speed,
a receiver unit (Fig. 6b). The two units act as transceiv- alerting authorities via GPS and GSM to prevent drunk
ers, each equipped with an LCD screen that displays its driving incidents. Li-Fi technology for vehicle to vehicle
state and any data acquired from the sensing modules communication via VLC improves data exchange, further
mounted in the vehicles. The transmitter transforms enhancing road safety.
digital data into visible light using an LED. The LED is Looking ahead, this adaptable system prototype can be
a suitable component because of its linear relationship customized and expanded to meet the diverse demands
between current and light output. The receiver uses a of various markets. It serves as a testament to technol-
photodiode to convert the incoming light into an elec- ogy driven solutions in making our urban transportation
trical charge. The photodiode is a semiconductor device systems safer, more efficient, and conducive to the overall
that converts light into an electric current. The first well being of smart city residents.
transmitter is connected to the Arduino device. Once the This study has significant implications as the proposed
transmitter receives the data from the Arduino board, it system can significantly reduce fatal road accidents by
converts it to binary and transmits it using the LED. If the addressing key factors like drowsy driving, speeding, and
binary value is 0, the LED does not blink, but if the binary alcohol consumption. Ultrasonic sensors enhance traffic
value is 1, the LED blinks. The blinking happens so fast flow, reducing congestion and accident risks. By automat-
that it is not visible to the human eye. The photovoltaic ically restricting vehicle speed when alcohol is detected,

Fig. 6 Li-Fi transmitter and receiver modules in the vehicles

Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 16 of 18

the system deters drunk driving. The adoption of Li-Fi and artificial intelligence (Ai) for smart city scenario: Recent advance-
ments and future trends. Sensors (Basel, Switzerland), 23(11), 5206.
for vehicle communication highlights VLC technology’s Alam, S. H., Tahsin, S. S., Tarannum, I., Chowdhury, T., & Sarony, K. (2019). A cost-
role in transportation safety and efficiency. The adaptable effective driver drowsiness recognition system. In: 2019 IEEE International
system can be customized to meet various market needs, Conference on Biomedical Engineering, Computer and Information
Technology for Health (BECITHCON), IEEE pp. 20–23
contributing to safer and more sustainable smart cities. Bibri, S. E. (2018). The iot for smart sustainable cities of the future: An analytical
Our forthcoming research will focus on enhancing the framework for sensor based big data applications for environmental
practicality of the drowsiness detection module, espe- sustainability. Sustainable Cities and Society, 38, 230–253.
Bibri, S. E. (2019). On the sustainability and unsustainability of smart and
cially when drivers wear recommended driving lenses. smarter urbanism and related big data technology, analytics, and
We will thoroughly investigate and test the system’s abil- application. Big data science and analytics for smart sustainable urban-
ity to consistently and accurately detect signs of tiredness ism: Unprecedented paradigmatic shifts and practical advancements,
183–220.
and drowsiness in such scenarios. This research aims to Bibri, S. E. (2020). The iot and big data analytics for smart sustainable cities:
make the system more user friendly and adaptable to real Enabling technologies and practical applications. Advances in the Lead-
world driving conditions. Additionally, our research has ing Paradigms of Urbanism and their Amalgamation: Compact Cities,
Eco–Cities, and Data Driven Smart Cities, 191–226.
paid significant attention to the rationale for selecting Bibri, S. E., & Jagatheesaperumal, S. K. (2023). Harnessing the potential of
Arduino and GSM technology in our proposed frame- the metaverse and artificial intelligence for the internet of city things:
work for smart city transportation safety. We understand Cost-effective xreality and synergistic aiot technologies. Smart Cities, 6(5),
2397–2429.
the importance of justifying these choices and ensuring Bibri, S. E., Alexandre, A., Sharifi, A., & Krogstie, J. (2023a). Environmentally
their suitability for our framework’s specific applica- sustainable smart cities and their converging Ai, iot, and big data tech-
tions. The integration of AIoT innovations for transpor- nologies and solutions: An integrated approach to an extensive literature
review. Energy Informatics, 6(1), 9.
tation safety represents a revolutionary change toward Bibri, S. E., Krogstie, J., Kaboli, A., & Alahi, A. (2023b). Smarter eco-cities and their
safer and smarter mobility systems. We are sculpting a leading-edge artificial intelligence of things solutions for environmental
future in which transportation is not only more effective sustainability: A comprehensive systematic review. Environmental Science
and Ecotechnology, 19, 100330.
but also fundamentally safe for all users of the road by Bibri, S. E., Huang, J., Yigitcanlar, T., & Krogstie (2024). Artificial intelligence of
fusing AI intelligence with IoT connection. sustainable smart city cyber-physical systems of systems: A pioneering
framework for data-driven management and planning. Sustainable Cities
and Society (in press).
Authors’ contributions Chen, Q., Xie, Y., Ao, Y., Li, T., Chen, G., Ren, S., Wang, C., & Li, S. (2021). A deep
All authors contributed equally to the study. SEB additionally proofread and neural network inverse solution to recover pre crash impact data of car
reviewed the manuscript. All authors read and agreed to the published ver- collisions. Transportation Research part C; Emerging Technologies 126,
sion of the manuscript. 103009.
Chhabra, R., Krishna, C. R., & Verma, S. (2022). Augmenting driver’s situational
Funding awareness using smartphones in vanets. Arabian Journal for Science and
This research received no external funding. Engineering, 1–18.
Cui, Y., & Lei, D. (2023). Design of highway intelligent transportation system
Availability of data and materials based on the internet of things and artificial intelligence. Ieee Access :
There is no data made available from this work for public access. Practical Innovations, Open Solutions.
Darmawan, A. K., Setyawan, M. B., Bakir, B., Walid, M., Hamzah, M. A., & Asir,
A. (2021). Assessing and enhancing an existing user experience (ux) of
Declarations smart regency mobile apps service with mecue 2.0 framework. In: 2021
9th International Conference on Cyber and IT Service Management
Competing interests (CITSM), pp. 1–6 IEEE.
The authors don’t have a conflict of interest over the contents of the Dibaei, M., Zheng, X., Xia, Y., Xu, X., Jolfaei, A., Bashir, A. K., Tariq, U., Yu, D., &
manuscript. Vasilakos, A. V. (2021). Investigating the prospect of leveraging blockchain
and machine learning to secure vehicular networks: A survey. IEEE Trans-
actions on Intelligent Transportation Systems, 23(2), 683–700.
Received: 3 January 2024 Revised: 22 February 2024 Accepted: 20 March Efthymiou, I. P., & Egleton, T. E. (2023). Artificial intelligence for sustainable
2024 smart cities. In Handbook of research on applications of ai, digital twin,
and internet of things for sustainable development p, 1–11.
Evans, J., Karvonen, A., Luque-Ayala, A., Martin, C., McCormick, K., Raven, R., &
Palgan, Y. V. (2019). Smart and sustainable cities? Pipedreams. practicalities
and Possibilities, 24(7), 557–564.
References Fujiwara, K., Abe, E., Kamata, K., Nakayama, C., Suzuki, Y., Yamakawa, T., Hiraoka,
Abduljabbar, R., Dia, H., Liyanage, S., & Bagloee, S. A. (2019). Applications of T., Kano, M., Sumi, Y., Masuda, F., et al. (2018). Heart rate variability based
artificial intelligence in transport: An overview. Sustainability, 11(1), 189. driver drowsiness detection and its validation with eeg. IEEE Transactions
Ahvenniemi, H., Huovila, A., Pinto-Seppa, I., & Airaksinen, M. (2017). R.: What are on Biomedical Engineering, 66(6), 1769–1778.
the differences between sustainable and smart cities? Valvon Takonsultit Gong, T., Zhu, L., Yu, F. R., & Tang, T. (2023). Edge intelligence in intelligent trans-
Oy (pp. 234–245). Cities. portation systems: A survey. IEEE Transactions on Intelligent Advancing
Al-Youif, S., Ali, M. A., & Mohammed, M. (2018). Alcohol detection for car Transportation Safety through AIoT 25 Transportation Systems.
locking system. In: 2018 IEEE Symposium on Computer Applications & Gourisaria, M. K., Jee, G., Harshvardhan, G., Konar, D., & Singh, P. K. (2022). Artifi-
Industrial Electronics (ISCAIE), pp. 230–233 IEEE Advancing Transportation cially intelligent and sustainable smart cities. In: Sustainable Smart Cities:
Safety through AIoT 29. Theoretical Foundations and Practical Considerations, pp. 237– 268.
Alahi, M. E. E., Sukkuea, A., Tina, F. W., Nag, A., Kurdthongmee, W., Suwannarat, Gowri, S. M., Anitha, P., Srivaishnavi, D., & Nithya, M. (2019). Internet of things
K., & Mukhopadhyay, S. C. (2023). Integration of iot enabled technologies based accident detection system. In: 2019 Third International Conference
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 17 of 18

on I-SMAC (IoT in Social, Mobile, Analytics and Cloud)(I-SMAC), pp. Priya, S., & Saranya, K. (2023). Significance of artificial intelligence in the
159–163 IEEE . development of sustainable transportation. The Scientific Temper, 14(02),
Guede-Fernandez, F., Fernandez-Chimeno, M., Fernandez-Chimeno, J., & 418–425.
Garcia- Gonzalez, M. .A. (2019). Driver drowsiness detection based on Puri, V., Van Le, C., Kumar, R., & Jagdev, S. S. (2020). Fruitful synergy model
respiratory signal analysis. IEEE Access, 7, 81826–81838. of artificial intelligence and internet of thing for smart transportation
Hammad, S. S., Iskandaryan, D., & Trilles, S. (2023). An unsupervised tinyml system. International Journal of Hyperconnectivity and the Internet of Things
approach applied to the detection of urban noise anomalies under the , 4(1), 43–57.
smart cities environment. Internet of Things, 23, 100848. Sathasivam, S., Saon, S., Sidek, A., Som, M. M., Ameen, H. A. (2020). : Drowsiness
Iyer, L. S. (2021). Ai enabled applications towards intelligent transportation. detection system using eye aspect ratio technique. In: 2020 IEEE Student
Transportation Engineering, 5, 100083. Conference on Research and Development (SCOReD),pp. 448–452 IEEE .
Jacob, C. M., George, N., Lal, A., George, R. J., Antony, M., & Joseph, J. (2020). An Selvaraj, S., & Umakanth, N. (2021). Providing safety and detecting accidents in
iot based smart monitoring system for vehicles. In: 2020 4th International bike transport with smart helmets using iot. Handbook of research on deci-
Conference on Trends in Electronics and Informatics (ICOEI)(48184), pp. sion Sciences and Applications in the Transportation Sector (pp. 280–303).
396–401 IEEE. IGI Global.
Janjua, J. I., Khan, T. A., Khan, M. S., & Nadeem, M. (2021). Li-fi communications Selvathi, D., Pavithra, P., & Preethi, T. (2017). Intelligent transportation system for
in smart cities for truly connected vehicles. In: 2021 2nd International accident prevention and detection. In: 2017 International Conference on
Conference On Smart Cities, Automation & Intelligent Computing Sys- Intelligent Computing and Control Systems (ICICCS), pp. 442–446 IEEE.
tems (ICON-SONICS),pp. 1–6 IEEE. Seng, K. P., Ang, L. M., & Ngharamike, E. (2022). Artificial intelligence internet
Kabir, A. T., Zinnurayen, N., Abrar, T., Ta-sin, A. J. (2020). The smart vehicle of things: A new paradigm of distributed sensor networks. International
management system for accident prevention by using drowsiness, Journal of Distributed Sensor Networks 18(3), 15501477211062835 28
alcohol, and overload detection. In: 2020 10th Electrical Power, Electron- Advancing Transportation Safety through AIoT.
ics, Communications, Controls and Informatics Seminar (EECCIS), pp. Sharifi, A., Allam, Z., Bibri, S. E., & Khavarian-Garmsir, A. R. (2024). Smart cities
173–177 IEEE. and sustainable development goals (sdgs): A systematic literature review
Kamran, M. A., Mannan, M. M. N., & Jeong, M. Y. (2019). Drowsiness, fatigue and of co-benefits and trade offs. Cities, 146(10465), 9.
poor sleep’s causes and detection: A comprehensive study. Ieee Access, 7, Siddiqui, S., Ahmad, I., Khan, M., Khan, B., Ali, M., Naseer, I., Parveen, K., & Usama,
167172–167186. H. (2021). Aiot enabled traffic congestion control system using deep
Khan, M. H. U., & Howlader, M. M. (2019). Design of an intelligent autonomous neural network. EAI Endorsed Transactions on Scalable Information Systems
accident prevention, detection and vehicle monitoring system. In: 2019 8(33).
IEEE International Conference on Robotics, Automation, Artificial intel- Singh, R., & Bera, T. (2018). Obstacle avoidance of bicycle vehicle model using
ligence and Internet of Things (RAAICON), IEEE pp. 40–42. overwhelming controller. Arabian Journal for Science and Engineering, 43,
Kinage, V., & Patil, P. (2019). Iot based intelligent system for vehicle accident 4821–4833.
prevention and detection at real time. In: 2019 Third International Confer- Singh, A., & Singla, A. (2021). Constructing definition of smart cities from
ence on I-SMAC (IoT in Social, Mobile, Analytics and Cloud)(I-SMAC), IEEE systems thinking view. Kybernetes, 50(6), 1919–1950.
pp. 409–413. Siva, A. S., Vishal, P., Subramanian, A. S., Ramesh, M., Sathishkumar, S., &
Lim, B. S., Keoh, S. L., & Thing, V. L. (2018). Autonomous vehicle ultrasonic sen- Vinothini, N. (2022). Highly enhanced safety & security system in helmet
sor vulnerability and impact assessment. In: 2018 IEEE 4th World Forum & bike using various sensors with gsm modules. In: 2022 3rd International
on Internet of Things (WF-IoT), IEEE pp. 231–236. Conference on Smart Electronics and Communication (ICOSEC), pp.
Malathi, M., Sujitha, R., & Revathy, M. (2017). Alcohol detection and seat belt 183–187 IEEE.
control system using arduino. In: 2017 International Conference on Son, T. H., Weedon, Z., Yigitcanlar, T., Sanchez, T., Corchado, J. M., & Mehmood,
Innovations in Information, Embedded and Communication Systems R. (2023). Algorithmic urban planning for smart and sustainable develop-
(ICIIECS), IEEE pp. 1–3. ment: Systematic review of the literature. Sustainable Cities and Society,
Mamikandan, P., Naveen, P., Ramkumar, K. (2022). : Li-fi technology for vehicle 104562.
to vehicle communication. In: 2022 International Conference on Edge Soomro, S., Miraz, M. H., Prasanth, A., & Abdullah, M. (2018). Artificial intel-
Computing and Applications (ICECAA), IEEE pp. 451–455. ligence enabled iot: Traffic congestion reduction in smart cities.
Manikandan, P., Muneeswaran, V., Ramesh, G., Rakesh, R. S., Chakraesh, P., Suryana, A., & Familiana, H. (2019). Automatic braking system on motorbikes
Reddy, N. S., & Sahul, N. (2021). Drunk and drive controller for vehicles. In: using the concept of kinematics non-uniform slowing down motion for
2021 International Conference on Advance Computing and Innovative safety of motorcycle riders on the highway. In: 2019 International Confer-
Technologies in Engineering (ICACITE), IEEE pp. 190–194. ence on ICT for Smart Society (ICISS), vol. 7, pp. 1–6 IEEE.
Mateen, A., Hanif, M. Z., Khatri, N., Lee, S., & Nam, S. Y. (2022). Smart roads for Toli, A. M., & Murtagh, N. (2020). The concept of sustainability in smart city
autonomous accident detection and warnings. Sensors (Basel, Switzer- definitions. Frontiers in Built Environment, 6, 77.
land), 22(6), 2077. Walizad, M. E., Hurroo, M., & Sethia, D. (2022). Driver drowsiness detection
Memedi, A., & Dressler, F. (2020). Vehicular visible light communications: A system using convolutional neural network. In: 2022 6th international
survey. IEEE Communications Surveys & Tutorials, 23(1), 161–181. conference on trends in Electronics and Informatics (ICOEI). Advancing
Mishra, R. K., Kumari, C. L., Janaki Krishna, P., & Dubey, A. (2022). Smart cities Transportation Safety through AIoT, 31, 1073–1080.
advancing Transportation Safety through AIoT 27 for sustainable develop- Wu, H. T. (2022). The internet of vehicle traffic condition system developed by
ment: An overview. Smart Cities for Sustainable Development, 1–12. artificial intelligence of things. The Journal of Supercomputing 78(2):30
Nozari, H., Tavakkoli-Moghaddam, R., Rohaninejad, M., & Hanzalek, Z. (2023). Advancing Transportation Safety through AIoT 2665–2680.
Artificial intelligence of things (aiot) strategies for a smart sustainable Wu, C. K., He, Y., Tsang, K. F., & Mozar, S. (2020). The idex case study on the safety
resilient supply chain. In: IFIP International Conference on Advances in measures of aiot based railway infrastructures. In: 2020 IEEE International
Production Management Systems, Springer pp. 805–816. Symposium on Product Compliance Engineering Asia (ISPCE-CN), pp.
Odat, E., Shamma, J. S., & Claudel, C. (2017). Vehicle classification and speed 1–4 IEEE.
estimation using combined passive infrared/ultrasonic sensors. IEEE Xie, Y., Wu, C., Li, B., Hu, X., & Li, S. (2022). A feed forwarded neural network
Transactions on Intelligent Transportation Systems, 19(5), 1593–1606. based variational bayesian learning approach for forensic analysis of
Perumal, P. S., Wang, Y., Sujasree, M., Mukthineni, V., & Shimgekar, S. R. (2022). traffic accident. Computer Methods in Applied Mechanics and Engineering,
Intelligent advice system for human drivers to prevent overtaking acci- 397, 115148.
dents in roads. Expert Systems with Applications, 199, 117178. Yang, L., Chen, X., Perlaza, S. M., & Zhang, J. (2020a). Special issue on artificial
Prasad, A., Pavankalyan, J., Ganesh, A. S., & Krishna, K. M. (2022). Drowsiness intelligence powered edge computing for internet of things. IEEE Internet
and alcohol detection with engine locking. In: 2022 3rd International of Things Journal, 7(10), 9224–9226.
Conference on Communication, Computing and Industry 4.0 (C2I4), IEEE Yang, S., Xu, K., Cui, L., Ming, Z., Chen, Z., & Ming, Z. (2020b). Ebi-pai: Toward an
pp. 1–5. efficient edge-based iot platform for artificial intelligence. IEEE Internet of
Things Journal, 8(12), 9580–9593.
Jagatheesaperumal et al. Computational Urban Science (2024) 4:10 Page 18 of 18

Yuvaraju, M., Mansingh, P. B., & Sekar, G. (2021). A li-fi based collision avoid-
ance system for vehicles using visible light communication. In: 2021
7th International Conference on Electrical Energy Systems (ICEES), pp.
114–116 IEEE.
Zaidi, A., Ajibade, S. S. M., Musa, M., & Bekun, F. V. (2023). New insights into the
research landscape on the application of artificial intelligence in sustain-
able smart cities: A bibliometric mapping and network analysis approach.
International Journal of Energy Economics and Policy 4, 287.
Zhang, J., & Tao, D. (2020). Empowering things with intelligence: A survey of
the progress, challenges, and opportunities in artificial intelligence of
things. IEEE Internet of Things Journal, 8(10), 7789–7817.
Zhang, F., Pan, Z., & Lu, Y. (2023). Aiot enabled smart surveillance for personal
data digitalization: Contextual personalization privacy paradox in smart
home. Information & Management, 60(2), 103736.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in pub-
lished maps and institutional affiliations.