MineGuard-Net: A Context-Driven Embedded Intelligence Framework for Real-Time Subsurface Hazard Cognition

Abstract

Coal mining is one of the most hazardous industrial activities, where workers are constantly exposed to risks such as toxic gas leaks and unexpected seismic vibrations. Historically, mine safety relied on manual inspection methods and basic mechanical detectors, which were often slow, inaccurate, and unable to provide real-time alerts. With increasing mining depth and complexity, these traditional systems have become insufficient to ensure worker safety. The major problem lies in the lack of continuous monitoring and immediate warning mechanisms, which can lead to severe accidents, environmental damage, and loss of human lives. Conventional safety systems typically operate in isolation, lack remote monitoring capabilities, and require significant human intervention. Their limitations include delayed response, limited coverage, and inability to store or analyse data for preventive measures. Hence, there is a growing need for an intelligent, automated, and reliable monitoring system that can detect hazards in real time and communicate effectively. This research proposes a Zigbee-based IoT embedded system for earthquake and hazardous gas detection in coal mines. The system integrates sensors for gas and vibration detection with a microcontroller, enabling continuous monitoring of environmental conditions. Zigbee technology is used for efficient wireless communication between transmitter and receiver units, while IoT connectivity allows data to be uploaded to a remote server for real-time access and analysis. Additionally, alerts are provided through buzzers and display units to ensure immediate response. The proposed system enhances safety by providing early warning, reducing human dependency, and enabling remote monitoring. It offers a cost-effective, reliable, and scalable solution, significantly improving coal mine safety and helping prevent potential disasters.