WildNav: GNSS-Denied Navigation Strategies for Long-Term Autonomy of Drones in the Wild

Autonomous unmanned aerial vehicles (UAVs) are becoming ubiquitous. The current autonomy stacks for outdoors flight rely on GNSS sensors. The WildNav project research is focused towards resilient GNSS-denied navigation in the wild for long-term flight even when GNSS signals are jammed or interferred with. We propose a multi-modal environment segmentation approach based entirely on onboard sensors for localizing the UAV and planning paths with low risk of getting lost.


WildNav effectively addresses current challenges in long-term aerial autonomy in the wild based only on onboard sensors. The project also delivers novel environment-aware, weather-aware and risk-aware planning and navigation algorithms. These will significantly raise the level of autonomy and degree of intelligence of aerial vehicles and robots.

Funded by: Finnish Ministry of Defence's Scientific Advisory Board for Defence (MATINE)




Related Publications

[1] Li Qingqing, Yu Xianjia, Jorge Peña Queralta, Tomi Westerlund, "Robust Multi-Modal Multi-LiDAR-Inertial Odometry and Mapping for Indoor Environments", arXiv preprint , arXiv (2023) . (View) (Download)

[2] Marius-Mihail Gurgu, Jorge Peña Queralta, Tomi Westerlund, "Vision-based GNSS-Free Localization for UAVs in the Wild", 7th International Conference on Mechanical Engineering and Robotics Research , IEEE (2022) . (View) (Download) (arXiv)