AI Based Fire Detection and Control System For An Indoor Positioning System
AI Based Fire Detection and Control System For An Indoor Positioning System
Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr269
AI Based Fire Detection and Control System for an
Indoor Positioning System
1
Abhishek Andil; 2Pranav Bhalerao; 3Gaurav Raygade;
4
Saurav Bhosale; 5Pradnya Kasture
Department of Computer Engineering,
RMD Sinhgad School of Engineering,
Pune, India
Department of Computer Engineering,
RMD Sinhgad School of Engineering,
Pune, India
Department of Computer Engineering,
RMD Sinhgad School of Engineering,
Pune, India
Department of Computer Engineering,
RMD Sinhgad School of Engineering,
Pune, India
Prof., Department of Computer Engineering,
RMD Sinhgad School of Engineering,
Pune, India
Publication Date: 2025/04/19
Abstract: With the rapid advancement of the internet and the growth of the Internet of Things (IoT), buildings are becoming
increasingly intelligent, utilizing advanced technological solutions for safety and efficiency. One significant application in this
domain is emergency navigation through wireless sensor networks (WSNs), which can effectively respond to crises. In
emergency situations, these networks play a crucial role by guiding occupants to the safest and quickest evacuation routes
based on real-time threat detection. By continuously monitoring environmental conditions, sensors detect potential hazards
and promptly relay evacuation instructions to prevent congestion and ensure a swift and organized response. This system is
designed not only to direct people to safety but also to dynamically select the most efficient paths, mitigating the risk of
crowding and enhancing the safety of occupants during emergencies.
The growing improvements in intelligent building systems, driven by advancements in AI and IoT, have opened new
possibilities to make safety measures better. By using these technologies, emergency systems can now include smart tools to
predict fire risks and ensure more effective safety actions. Additionally, the system's flexibility allows it to fit into different
types of buildings, such as homes, offices, or factories, making it a practical solution for handling fire emergencies.
Keywords: Internet, IoT, Wireless Sensor Network, Emergency Navigation, Intelligent Buildings, Evacuation System.
IJISRT25APR269 www.ijisrt.com 583
Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr269
How to Cite: Abhishek Andil; Pranav Bhalerao; Gaurav Raygade; Saurav Bhosale; Pradnya Kasture (2025) AI Based Fire Detection
and Control System for an Indoor Positioning System. International Journal of Innovative Science and Research Technology, 10(4),
583-588. https://doi.org/10.38124/ijisrt/25apr269
I. INTRODUCTION represents a significant step forward in emergency navigation,
offering a more intelligent, responsive, and adaptive approach
With the rapid advancement of technology, particularly to evacuation in critical situations.
the Internet of Things (IoT), buildings are becoming more
intelligent, incorporating systems that enhance safety, security, Adding IoT to smart buildings has brought a major
and operational efficiency. One of the most critical change in how emergencies are handled by creating systems
applications in intelligent buildings is the use of Wireless that react quickly to changing threats. These systems use
Sensor Networks (WSNs) for emergency navigation. These advanced sensors and learning tools to improve safety and
networks can monitor real-time environmental conditions, provide useful tips for managing buildings better. For example,
detect threats, and provide dynamic evacuation instructions to they reduce false alarms, help locate threats faster, and guide
guide occupants to safety during emergencies such as fires. By people on the best way to leave the building based on the
continuously assessing hazards and adjusting escape routes, situation and their location.
WSNs can reduce congestion, prevent confusion, and ensure
an organized, efficient response. II. MOTIVATION
However, despite these advancements, traditional Indoor fire incidents present significant threats to both
evacuation systems often rely on static routes and generalized life and property, often leading to devastating consequences if
instructions, which are not responsive to the constantly not managed swiftly and effectively. Timely and accurate fire
changing conditions of an emergency. During a recent visit to a detection is crucial for initiating emergency response systems
mall, we found our-self trapped in a fire, unable to find a safe, that guide individuals safely out of hazardous areas. With
efficient route to escape amid the confusion and chaos. The advancements in AI and machine learning, we have the
lack of real-time, personalized evacuation guidance made the opportunity to significantly improve fire detection and
situation more perilous, as the predefined routes failed to evacuation strategies. These technologies enable the real-time
account for the rapidly changing environment and individual analysis of various environmental factors, allowing for faster,
needs. This experience highlighted the gaps in current more precise identification of fire hazards and dynamic,
evacuation systems, which do not dynamically adjust to the personalized evacuation routes. By leveraging AI’s ability to
real-time situation or consider individual factors like location, process vast amounts of data and predict patterns, we can not
mobility, or the severity of the threat. only detect fires more accurately but also enhance the overall
safety and efficiency of evacuation efforts, minimizing risk
Motivated by this experience, we developed Smart and ensuring quicker responses in critical situations. Fires
Escape, a dynamic evacuation system that integrates mobile often happen quickly, giving people little time to act safely.
technology and wireless sensor networks to provide According to reports, building fires in recent years have caused
personalized, real-time escape routes based on an individual's large amounts of damage and loss of life, showing the need for
location, physical condition, and the evolving fire conditions. smarter systems. AI-powered tools can predict how fires will
Smart Escape aims to overcome the limitations of traditional grow, test evacuation plans in advance, and give clear
evacuation systems by ensuring that every person can be directions during an emergency. These tools make evacuations
guided to the safest and most efficient exit, reducing the risks faster and safer while helping responders make better choices.
associated with overcrowding and confusion. This project
IJISRT25APR269 www.ijisrt.com 584
Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr269
III. SYSTEM ARCHITECTURE
Fig 1 System Architecture
This system is built with parts that are designed to work "LEDPOS: Indoor Visible Light Positioning Based on
together smoothly and reliably. Sensors like heat cameras and LED as Sensor and Machine Learning"** by Christian Fragner
smoke detectors collect real-time data. This data is then and Christian Krutzler, also published in 2024, investigates an
processed by small computers that use AI to figure out where innovative indoor positioning system that combines visible
the fire is and suggest the safest routes. Wireless light communication (VLC) technology with machine learning
communication devices, such as Bluetooth beacons, ensure a techniques. In this system, LEDs are used as sensors to
strong connection so the system can guide people even if some determine the location of users or objects by analyzing the
parts fail. Backup power and extra safety measures are added intensity and characteristics of the light signals. This approach
to keep everything working during emergencies. offers a novel solution for indoor positioning using visible
light, enhanced by machine learning for accuracy.
IV. REVIEW OF LITERATURE
"Indoor Positioning System: A Review" by N. Syazwani
“WhereArtThou: A WiFi-RTT-Based Indoor Positioning C.J, Keng Yinn Wong, and Yichiet Aun, published in 2022,
System" by Rebal Jurdi, Hao Chen, and Yuming Zhu, provides a comprehensive review of various indoor positioning
published in 2024, discusses an indoor positioning system that technologies, including WiFi, Bluetooth, RFID, and VLC. The
utilizes WiFi Round-Trip Time (RTT) measurements to paper highlights the advantages and limitations of various
estimate distances accurately. This approach is particularly technologies used in indoor positioning systems and identifies
beneficial in environments where GPS signals are ineffective, potential directions for future research in this domain, offering
as it provides precise location information indoors by valuable insights into the evolving landscape of indoor
calculating distances based on RTT data. navigation solutions.
IJISRT25APR269 www.ijisrt.com 585
Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr269
"Simulation Research on Fire Evacuation of Large Public examining security protocols such as BACNet, KNX, and
Buildings Based on Building Information Modeling" by Fuyu ZigBee. It categorizes research into threats, privacy issues, and
Wang, Xiao Xu, and Mengkai Chen, published in 2021, defenses, while focusing on vulnerabilities introduced by IoT
explores fire evacuation strategies in large public buildings. integration. The review covers over 90 papers and evaluates
The study utilizes Building Information Modeling (BIM) along security threats, protocol resilience, and study robustness,
with Fire Dynamics Simulator (FDS) software to simulate identifying gaps in empirical testing, non-technical factors, and
different evacuation strategies. It compares the effectiveness of the need for protocol standardization.
various approaches, such as using only stairs versus combining
stairs and fire elevators, in ensuring safe evacuation during fire "Enhancing Security of Smart Buildings using Internet of
emergencies in complex indoor environments. Things" by Zahra Alisha and Scott Gordon, proposes an agent-
based IoT system for smart building security, using a multi-
"Development of Fire Detection Systems in Intelligent layered approach to manage tasks and coordinate security
Buildings" by Z.G. Liu, J.M. Makar, and A.K. Kim, introduces responses. The system operates with decentralized control,
a smart building fire detection system that leverages IoT tested with simulated intrusion data. Performance metrics
sensors, machine learning models, and image processing include response time, reliability, and flexibility. Research
methods. The proposed system incorporates smoke, gaps include real-world validation, integration challenges,
temperature, and gas sensors for real-time monitoring and uses advanced threat detection, and scalability.
both supervised and unsupervised machine learning
algorithms, including Support Vector Machines (SVM) and V. OPEN ISSUES
Neural Networks, to detect anomalies. Additionally, image
processing methods such as edge detection and color The primary goal of this thesis is to enhance the
segmentation are applied to recognize smoke and flames from performance of object detection and tracking by contributing
a dataset containing 10,000 images of various fire scenarios. novel improvements to two key areas: (a) motion segmentation
The system’s performance is assessed through key metrics like and (b) object tracking. Automated object tracking has many
LPIPS (Learned Perceptual Image Patch Similarity), FID practical applications, and having an accurate, reliable tracking
(Fréchet Inception Distance), PSNR (Peak Signal-to-Noise system is crucial for developing advanced vision-based
Ratio), SSIM (Structural Similarity Index), and IS (Inception systems. However, object tracking presents several challenges
Score). due to the unpredictable behavior of objects, variations in their
movement, and the complexities involved in capturing images
"Efficient and Secure IoT Based Smart Home or video. The aim of video tracking is to maintain continuity
Automation Using Multi-Model Learning and Blockchain by linking target objects across consecutive frames. This
Technology" by Nazik Alturki et al., the authors propose a process involves identifying and tracking objects as they move
smart home automation system that leverages multi-model independently against a stationary background. Four key
learning, blockchain, and IoT integration. The system scenarios are considered in this work: a single camera tracking
combines machine learning models, such as decision trees and a single object, a single camera tracking multiple objects,
neural networks, for predictive analytics, and uses blockchain multiple cameras tracking a single object, and multiple
to ensure data integrity and security. IoT devices provide real- cameras tracking multiple objects. As reviewed earlier, many
time monitoring and control of smart home systems. Evaluated difficulties arise in object detection, tracking, and recognition
with a dataset of 5,000 smart home device records, the when using fixed camera networks. Therefore, the objective of
system's performance is measured by accuracy, latency, this work is to create an automatic system for segmenting and
security metrics, and throughput. tracking moving objects in videos recorded by stationary
cameras. This system can provide a foundation for higher-level
"SMART HOME AUTOMATION SYSTEM BASED analysis and applications, while also improving commonly
ON IoT" by Mandeep Singh et al., proposes an IoT-based used algorithms. The final objective is to demonstrate a
automation system that integrates sensors, actuators, and method for detecting and tracking moving objects based on
internet connectivity for remote management of home devices. their motion in footage from fixed cameras. Specifically, the
The system features a mobile/web interface for controlling study focuses on understanding how segmentation techniques
lighting, HVAC, and security, using Wi-Fi for communication. can help identify objects and examining different tracking
Though no specific dataset was used, the paper discusses a approaches for following one or more objects. Even with
general IoT setup framework. The performance metrics focus recent progress, there are still important challenges to solve.
on energy efficiency, security, and control latency, with One issue is the heavy workload on the system when
identified gaps in security, scalability, real-world testing, and processing large amounts of data quickly, which can slow it
interoperability among devices. down. Another problem is making sure different IoT devices
work together without trouble. There are also privacy
"The Security of Smart Buildings: A Systematic concerns, as monitoring systems collect a lot of data, raising
Literature Review"** by Pierre Ciholas et al., conducts a questions about how to protect people's information. Future
systematic literature review on smart building security, work should focus on creating simpler tools and setting
IJISRT25APR269 www.ijisrt.com 586
Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr269
common rules to ensure these systems are easy to use and [5]. G. S, P. Umaeswari, M. Muktasandhu, M. Vignesh, R.
secure. Ramyamaranan and S. Dhondiyal, "Internet of Things
(IoT) for Remote Earthquake and Fire Detection
VI. CONCLUSION Monitoring: Linking Safety," 2024 International
Conference on Trends in Quantum Computing and
AI-Based Fire Detection and Control System for Indoor Emerging Business Technologies, Pune, India, 2024, pp.
Positioning Systems enhances emergency response by 1-5, doi: 10.1109/TQCEBT59414.2024.10545307.
integrating real-time fire detection with indoor positioning. keywords: {Wireless communication;Wireless sensor
Utilizing sensors, BLE beacons, and efficient pathfinding networks;Earthquakes;Real-time systems;Sensor
algorithms, the system provides users with safe, dynamic systems;Public security;Sensors;IoT;IVS;remote
evacuation routes, minimizing exposure to hazardous areas. monitoring;real-time safety system;seismic and fire
This solution significantly improves indoor safety, enabling detection;data analytics;ML;rescuing from disasters;and
quicker, more informed evacuation decisions and helping optimising evacuations},
protect lives in fire emergencies. [6]. M. Abdat, T. -C. Wan and S. Supramaniam, "Survey on
indoor wireless positioning techniques: Towards adaptive
REFERENCES systems," 2010 International Conference on Distributed
Frameworks for Multimedia Applications, Jogjakarta,
[1]. H. Jiang, "Mobile Fire Evacuation System for Large Indonesia, 2010, pp. 1-5. keywords: {Wireless
Public Buildings Based on Artificial Intelligence and communication;Fingerprint
IoT," in IEEE Access, vol. 7, pp. 64101-64109, 2019, doi: recognition;Accuracy;Radiofrequency
10.1109/ACCESS.2019.2915241.keywords: {Heuristic identification;Wireless LAN;IEEE 802.11
algorithms;Fires;Artificial Standards;Servers;Indoor Positioning Systems;Adaptive
intelligence;Carbonmonoxide;Elevators;Artificial Systems;Location Fingerprinting;Signal Strength},
intelligence;IoT;fire evacuation;mobile terminal}, [7]. Kumareson P, Rajasekar R, Prakasam P. Accurate
[2]. W. Wang et al., "Dijkstra Algorithm Based Building Location Identification Using Bluetooth in Android
Evacuation Recognition Computing and IoT System Smartphone. Discovery, 2015, 30(120), 81-86 “Accurate
Design and Implementation," 2019 IEEE 13th Location Identification Using Bluetooth in Android
International Conference on Anti-counterfeiting, Security, Smartphone”
and Identification (ASID), Xiamen, China, 2019, pp. [8]. Yang-Hang Lee , Kuo-Wei Ho, Hsien-Wei Tseng,”
229-233, doi: 10.1109/ICASID.2019.8925068. keywords: Accurate Bluetooth Positioning Using Large Number of
{Fires;Buildings;Image recognition;Field programmable Devices Measurements”, IMCES ,Vol.2,No 4, PP.381-
gate arrays;Wireless communication;Light emitting 394 March 12 - 14, 2014,
diodes;Wireless sensor networks;Dijkstra's [9]. "Indoor Positioning Systems based on BLE Beacons,"
Algorithm;building fire;intelligent evacuation;image Locatify, [Online]. Available:
recognition;Internet of Things}, https://locatify.com/blog/indoor-positioningsystems-ble-
[3]. Y. Gao, "IoT and Lightbend based Intelligent Platform for beacons/. [Accessed 1 December 2021].C. Wojek, P.
Fire Monitoring System," 2020 Second International Dollar, B. Schiele, and P. Perona, “Pedestrian detection:
Conference on Inventive Research in Computing An evaluation of the state of the art,” IEEE Trans. Pattern
Applications (ICIRCA), Coimbatore, India, 2020, pp. 22- Anal. Mach. Intell., vol. 34, no. 4, p. 743, 2012.
25, doi: 10.1109/ICIRCA48905.2020.9182960. [10]. Myeongcheol Kwak, Youngmong Park, Junyoung kim,
keywords: {Buildings;Fires;Monitoring;Heat Jinyoung Han, and Taekyoung Kwon, "An Energy-
transfer;Heating systems;Combustion;Intelligent efficient and Lightweight Indoor Localization System for
monitoring;mechanism study;complicated Internet-of-Things (IoT) Environment," Proceedings of
conditions;intelligent buildings} the ACM on Interactive Mobile Wearable and Ubiquitous
[4]. A. Ojha, A. Jindal and P. Chanak, "An Intelligent Indoor Technologies, vol. 2, no. 1, pp. 1-28, 2018.
Emergency Evacuation System Using IoT-Enabled [11]. Nur Haliza Abdul Wahab, N. Syazwani C. J, Sharifah
WSNs for Smart Buildings," in IEEE Internet of Things Hafizah Syed Ariffin, Nuraini Huda Abdul Kadir, and
Journal, vol. 11, no. 5, pp. 8838-8847, 1 March1, 2024, Noorhazirah Sunar, "Three Dimension (3D) Indoor
doi: 10.1109/JIOT.2023.3321646. keywords: {Wireless Positioning via Received Signal Strength Indicator in
sensor networks;Navigation;Real-time systems;Internet Internet of Things," Lecture Notes in Electrical
of Things;Heuristic algorithms;Computational Engineering, vol. 842, pp. 1081-1092, 2021.
modeling;Three-dimensional displays;Dynamic [12]. Zhang Yong and Wu Chengbin, "An indoor positioning
emergency evacuation;Internet of Things (IoT);shortest- system using Channel State Information based on
safe path;wireless sensor networks (WSNs)}, TrAdaBoost Tranfer Learning," 2021 4th International
Conference on Advanced Electronic Materials,
Computers and Software Engineering (AEMCSE), pp.
1286-1293, 2021.
IJISRT25APR269 www.ijisrt.com 587
Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr269
[13]. Siye Wang , Chang Ding, Weiqing Huang , Yanfang [23]. https://chatgpt.com/
Zhang , Jianguo Jiang , Shaoyi Zhu , Yue Cui, and Jun-yu [24]. https://www.wikipedia.org/
Lin , "Determination of an Indoor Target Position: An [25]. https://www.researchgate.net/figure/System-architecture-
Accurate and Adaptable Approach Based on Differential for-an-IoT-based-fire-evacuation-
Positioning," International Journal of Antennas and system_fig2_339096717
Propagation, vol. 2019, p. 19, 2019. [26]. https://en.wikipedia.org/wiki/Indoor_positioning_system
[14]. Khan. A. A. Aesha. J.S. Aka. S.M. F. Rahman and M. J.
Rahman. "An IoT Based Intelligent Fire Evacuation
System." 2018 21st International Conference of
Computer and Information Technology (ICCIT). 2018.
pp. 1-6. doi:10.1 109/ICCITECHN.2018.8631945.
[15]. Jiang. Huixian (2019). Mobile Fire Evacuation System
for Large Public Buildings Based on Artificial
Intelligence and loT. IEEE Access. PP. 1-1. 10.1
109/ACCESS.2019.2915241
[16]. Jiang. Q.. Song. G.. Gao, C.. Wang. Y.. Si. W.. Xie. K.
(2011). Simulated Annealing Based Influence
Maximization in Social Networks. Proceedings of the
AAAI Conference on Artificial Intelligence, 25(1), 127-
132. Retrieved from
https://ojs.aaai.org/index.php/AAAI/article/view/7838
[17]. Inoue. Yutaka Sashima, Akio Ikeda, Takeshi K urumatani,
Koichi. (2008). Indoor Emergency Evacuation Service on
Autonomous Navigation System using Mobile Phone.
Proceedings of the 2nd International Symposium on
Universal Communication. ISUC 2008. 79-85. 10.1
109/1SUC.2008.49,
[18]. Y. Inoue. A. Sashima T. Ikeda and K.Kurumatani, "Indoor
Emergency Evacuation Service on Autonomous
Navigation System using Mobile Phone,' 2008 Second
International Symposium on Universal Communication,
2008, pp. 79- 85, doi: 10.1 109/1SUC.2008.49
[19]. Y. Inoue. A. Sashima T. Ikeda and K. Kurumatani,
"Iindoor Emergency Evacuation Service on Autonomous
Navigation System using Mobile Phone,' 2008 Second
International Symposium on Universal Communication,
2008, pp. 79- 85, doi: 10.1 109/1SUC.2008.49
[20]. S Majumder, S. O'Neil and R. Kennedy, "Smart apparatus
for fire evacuation- An IoT based fire emergency
monitoring and evacuation system." 2017 IEEE MIT
Undergraduate Research Technology Conference
(URTC), 2017, pp. 1-4, doi:
10.1109/URTC.2017.8284186
[21]. Yuanyuan Zeng. C. Sreenan and L. Sitanayah, "A real-
time and robust routing protocol for building fire
emergency applications using wireless sensor networks."'
2010 8th IEEE International Conference on Pervasive
Computing and Communications Workshops (PERCOM
Workshops), 2010, pp. 358-363,
doi:10.1109/PERCOMW.2010.5470643.
[22]. Saeed, Faisal, Anand Paul. Abdul Rchman, Won H.
Hong. and Hyuncheol Sco, 2018. "IoT-Based Intelligent
Modeling of Smart Home Environment for Fire
Prevention and Safety" Journal of Sensor and Actuator
Networks 7, no. 1: 11. https://doi.org/10.
.3390/jsan7010011.
