AI-Assisted LiDAR Mapping and Biometric Telemetry for Real-Time Risk Identification in Underground Exploration
Lizi Shamugia
12/08/2026
This research investigates the efficacy of an integrated risk-focused framework designed for safety assessment in underground systems where GPS (Global Positioning System) and vision-based sensors are unreliable. Traditional exploration often relies on static mapping that fails to account for dynamic hazards and physiological stress. This study proposes a dual-node architecture that fuses wave-based 2D LiDAR (Light Detection and Ranging) mapping with wearable biometric monitoring to provide real-time risk identification. Utilizing an RPLIDAR sensor, the system employs Simultaneous Localization and Mapping to reconstruct 3D environments while identifying spatial risks such as restricted clearance and obstacle density. Due to current hardware deployment stages, the methodology utilizes simulated LiDAR datasets to verify the Artificial intelligence’s ability to categorize environmental hazards and provide emergency pathfinding guidance. Results indicate that integrating human health indicators with spatial geometry allows for a more comprehensive safety assessment than geometric mapping alone. This research contributes a scalable, human-centric approach to improving the reliability of search and rescue operations in complex, GPS-denied underground environments.