Autonomous collision-free navigation of UAV in unknown tunnel-like environments

Abstract

Unmanned Aerial Vehicles (UAVs) have emerged as essential tools for performing inspections and explorations in challenging underground environments, including mines, drainage systems, and subterranean infrastructures. These complex tunnel-like settings introduce significant operational challenges such as GNSS-denied localization, limited visibility, sparse visual features, electromagnetic interference (EMI), and dynamic obstacles. Consequently, there is a critical need for sophisticated autonomous navigation solutions combining advanced Simultaneous Localization and Mapping (SLAM), multi-modal sensor fusion, and AI-based path planning strategies to ensure safe and efficient UAV operations.