Autonomous Fire Extinguisher Rover with Self-Preservation and Automatic Emergency Alert System

Abstract

Fire accidents remain one of the most serious hazards threatening human life, property, and the environment, and traditional firefighting methods frequently expose responders to dangerous conditions while suffering from delayed response in con-fined or hazardous areas. This paper presents the design and development of an Autonomous Fire Extinguisher Rover with Self-Preservation and Automatic Emergency Alert System, a mobile robotic platform that detects and suppresses fire in hazardous environments while simultaneously alerting nearby individuals and emergency responders in real time. The rover integrates infrared flame sensors, an MQ2 smoke/gas sensor, and a camera-based vision module to monitor environmental conditions and confirm fire presence through a two-stage, multi-modal detection scheme. A dual-controller architecture comprising an ESP32 microcontroller and a Raspberry Pi 4 enables real-time sensor polling, mo-tor and pump actuation, OpenCV-based HSV colour-segmentation fire detection, MJPEG live video streaming, and Telegram-based emergency alerting. On confirming a fire, the rover autonomously approaches the source, activates a servo-directed water-pump sup-pression mechanism, and notifies designated users wirelessly. Experimental evaluation across five complete end-to-end missions showed a camera detection latency of approximately 300–400 ms, an average approach distance of 48.4 cm from the flame, an aver-age fire-suppression time of 5.9 s, Telegram alert delivery within 1–3 s, and zero false activations after threshold tuning, confirm-ing that the proposed system provides a practical, low-cost, and reliable solution for reducing human risk and improving emer-gency response in fire-prone environments such as warehouses, industrial facilities, and residential buildings.

Introduction

This paper presents the design and development of an Autonomous Fire Extinguisher Rover with Self-Preservation and Automatic Emergency Alert System, a low-cost robotic platform capable of detecting, approaching, and extinguishing fires while minimizing risks to human firefighters.

Introduction

Fire accidents pose serious threats to human life, infrastructure, and the environment. Traditional firefighting methods often expose responders to dangerous conditions such as toxic smoke, high temperatures, and confined spaces. Advances in robotics, embedded systems, and sensor technology have enabled the development of autonomous firefighting robots capable of operating safely in hazardous environments.

The proposed rover integrates:

• Multi-sensor fire detection
• Autonomous navigation
• Targeted water-spray fire suppression
• Live video monitoring
• Real-time wireless emergency alerts

The system uses infrared flame sensors, an MQ2 smoke sensor, and a camera-based vision system to accurately detect fires. An ESP32 microcontroller manages real-time sensing and motor control, while a Raspberry Pi 4 performs computer vision, video streaming, and Telegram-based notifications. The goal is to provide affordable and reliable fire protection for residential buildings, warehouses, and industrial facilities.

Literature Review

Previous studies have developed autonomous firefighting robots using flame and smoke sensors controlled by Arduino or similar platforms. While these systems successfully automated fire detection and suppression, most lacked features such as live monitoring, remote communication, and advanced vision capabilities.

Research has shown that:

• Sensor fusion significantly improves fire detection accuracy compared to single-sensor systems.
• The ESP32 is suitable for real-time embedded control but has limited image-processing capability.
• The Raspberry Pi effectively supports computer vision and network communication.
• AI-based fire detection and wireless IoT communication offer promising directions for future firefighting systems.

Compared with previous work, the proposed rover uniquely combines:

• Dual-controller architecture (ESP32 + Raspberry Pi)
• Multi-modal fire confirmation using flame, smoke, and vision sensors
• Servo-controlled targeted water spraying
• Live MJPEG video streaming
• Real-time Telegram alerts

Proposed System

The rover consists of six major modules:

1. Sensor Module
   • Three infrared flame sensors detect flame direction.
   • An MQ2 gas/smoke sensor identifies combustible gases and smoke.
2. Processing and Control Unit
   • ESP32 performs real-time sensor monitoring, navigation, and actuator control using a finite state machine.
   • Raspberry Pi 4 handles OpenCV-based vision processing, live video streaming, and Telegram communication.
3. Navigation and Fire Suppression
   • Four DC motors driven by an L298N motor driver enable autonomous movement.
   • A relay-controlled water pump and servo-mounted nozzle direct water accurately toward the detected fire.
4. Vision-Based Fire Detection
   • A Raspberry Pi camera captures live images.
   • HSV color segmentation identifies flame-colored regions.
   • Morphological filtering and contour analysis reduce noise.
   • Fire confirmation combines flame sensor, smoke sensor, and vision outputs using a weighted confidence model to minimize false detections.
5. Communication and Monitoring
   • Live MJPEG video streaming allows remote monitoring.
   • Telegram notifications inform users of system startup, fire detection, suppression status, and mission completion.
6. Self-Preservation Mechanism
   • The rover stops at a safe distance before activating the water spray.
   • This prevents excessive heat exposure and protects onboard electronics while maintaining effective firefighting performance.

Working Principle

The rover continuously monitors its surroundings during normal operation. When multiple sensors consistently confirm a fire:

1. The rover moves toward the fire.
2. It stops at a safe extinguishing distance.
3. The servo aligns the nozzle with the flame.
4. The water pump suppresses the fire.
5. Telegram alerts and live video are transmitted.
6. After confirming the fire has been extinguished, the rover automatically returns to standby mode, ready for future emergencies.

Experimental Results

A prototype was built and tested through both individual module evaluations and complete mission trials.

The experiments demonstrated:

• Reliable fire detection under different lighting conditions.
• Successful autonomous navigation and targeted fire suppression.
• Accurate sensor fusion with reduced false alarms.
• Effective live video streaming and real-time Telegram notifications.
• Improved robustness after adjusting the vision algorithm to eliminate false positives caused by sunlight reflections.

Conclusion

This paper presented the design, implementation, and experi-mental validation of an Autonomous Fire Extinguisher Rover with Self-Preservation and Automatic Emergency Alert Sys-tem. The dual-controller architecture, combining an ESP32 for low-latency sensing and actuation with a Raspberry Pi 4 for vi-sion processing and wireless communication, proved effective in balancing real-time response with computational capability. The two-stage, multi-modal detection scheme — vision-based HSV segmentation together with redundant IR flame sensing achieved zero false suppression activations after threshold tuning, while the finite state machine ensured predictable and fully autonomous operation across repeated fire events. Exper-imental results across five complete end-to-end missions con-firm that the system reliably detects fire within a few hundred milliseconds, approaches and extinguishes the flame within seconds, and delivers emergency alerts almost instantaneously, demonstrating that a cost-effective, fully autonomous fire de-tection and suppression platform can be realised using widely available embedded hardware and open-source software. The modular architecture provides a practical foundation for future enhancements such as AI-based fire classification, obstacle-avoidance navigation, thermal imaging, and multi-agent co-ordination, supporting deployment in residential, commercial, and industrial fire-safety applications.

References

[1] A. Ghosh, S. Guha, K. N. Singh, C. Misra, A. Samui, and A. Dutta, “Autonomous Fire Extinguisher Rover with Self-Preservation and Emer-gency Protocols Using Arduino,” Journal of Mobile Multimedia, vol. 21, no. 3-4, pp. 599–618, 2025. [2] Espressif Systems. (2024) ESP32 Technical Reference Manual. [Online]. Available: http://www.espressif.com/en/support/resources [3] Raspberry Pi Foundation. (2024) Raspberry Pi 4 Model B. [Online]. Available: http://www.raspberrypi.org/products/ [4] H. R. Chen, “Multi-sensor fusion for fire detection in autonomous sys-tems,” Journal of Fire Science and Technology, vol. 45, no. 3, pp. 234–247, 2023. [5] J. Smith, A. Kumar, and L. Brown, “AI-based fire recognition using Ten-sorFlow Lite on embedded devices,” IEEE Trans. Ind. Electron., vol. 69, no. 6, pp. 5678–5686, 2022. [6] M. Kaur and N. Singh, “Wireless communication protocols for industrial IoT applications,” Int. J. Embedded Syst., vol. 14, no. 2, pp. 112–123, 2024. [7] S. Patel and R. Singh, “Design and development of autonomous firefight-ing robots: A comprehensive review,” Int. J. Robot. Autom., vol. 36, no. 4, pp. 312–329, 2024. [8] L. Wang, X. Zhao, and Y. Chen, “Sensor fusion techniques for fire detec-tion in robotic systems,” Sensors, vol. 22, no. 3, p. 1150, 2022. [9] D. Kumar and H. Park, “Energy-efficient embedded systems for safety-critical applications,” IEEE Trans. Ind. Informat., vol. 18, no. 1, pp. 400–410, 2022. [10] R. Garcia, T. Alvarez, and P. Romero, “Multi-agent coordination for fire-fighting robotics,” in Proc. IEEE Int. Conf. Robot. Autom., 2023, pp. 540–547. [11] X. Ren, Y. Zhao, and Z. Wang, “Research on Accurate Fire Source Local-ization and Seconds-Level Autonomous Firefighting,” Scientific Reports, vol. 15, 2025. [12] A. Hasib, M. Rahman, and S. Islam, “An IoT-Integrated Autonomous Fire Extinguishing and Monitoring System,” Proceedings of the IEEE National Conference on Intelligent Machines, 2025. [13] H. G. Medewar and S. Patil, “A Review on Fire and Smoke Detection with Intelligent Control for Enhanced Safety Using Machine Learning and Internet of Things,” Cureus Journal of Engineering and Technology, vol. 2024, pp. 1–15, 2024. [14] S. Danish, M. Afghah, and A. Khan, “Vision-Based Fire Management System Using Autonomous Aerial Vehicles and Artificial Intelligence,” Artificial Intelligence Review, vol. 58, 2025. [15] K. Jindal, A. Wang, D. Thakur, A. Zhou, and T. Krajnik, “Design and Deployment of an Autonomous Unmanned Ground Vehicle for Urban Firefighting Scenarios,” Proceedings of MBZIRC, pp. 1–10, 2021.

Copyright

Copyright © 2026 Muhammed Hashim K, Shahnad Babu P P, Sinan Shihab, Fardhil Aboo, Shanooja K. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.