From Detection to Traversal: a Probabilistic Framework for UAS-Assisted Landmine Mapping and Circumvention

This research presents a robust probabilistic framework for minefield localization, mapping, and avoidance, addressing a technological gap in the field of aerial countermine intelligence, while bypassing the well-established techniques of landmine detection. Our approach propagates the pose uncertainty matrix delivered by a drone's flight controller's Kalman filter to probabilistically estimate the location of detected mines. This probability map then seeds an artificial potential field path generator which creates a safe path for ground traversal by producing waypoints through the minefield. The system's performance is evaluated in simulations and validated through flight trials, demonstrating its potential to improve the efficiency and safety of UAV-assisted minefield navigation and threat avoidance.