autonomous vehicles benchmark computer vision environmental extreme weather geospatial GNSS IMU lidar localization mapping meteorological perception radar RINEX robotics signal processing
The FoMo dataset is a multi-season collection recorded in a boreal forest environment, featuring deep snow, off-road terrain, steep slopes, and highly variable weather. It provides synchronized multi-modal sensor data—including two lidars (RoboSense and Leishen), an FMCW radar (Navtech), stereo and monocular cameras, dual IMUs, wheel odometry, power data, calibration sequences, and precise ground-truth trajectories via GNSS-PPK fusion. Designed to support research on robust robot autonomy under adverse conditions, FoMo includes repeated traversals of six trajectories of varying complexity for long-term SLAM and odometry evaluation, as well as rich metadata such as one-minute weather station measurements. The dataset is intended to challenge state-of-the-art SLAM, localization, traversability analysis, and multi-season robotics research.
The dataset is considered complete and stable. Minor updates or corrections may occur, but they are expected to be infrequent.
Creative Commons Attribution 4.0 International (CC BY 4.0). See https://creativecommons.org/licenses/by/4.0/
https://fomo.norlab.ulaval.ca/overview
See all datasets managed by Norlab, Université Laval.
FoMo - A Multi-Season Dataset for Robot Navigation in Forêt Montmorency was accessed on DATE from https://registry.opendata.aws/fomo-norlab.
arn:aws:s3:::fomo-datasetus-west-2aws s3 ls --no-sign-request s3://fomo-dataset/