FoMo: Multi-Season Robot Navigation Data from a Boreal Forest

A multi-season dataset for robot navigation recorded in the boreal forest environment of Forêt Montmorency. It provides synchronized multi-modal sensor data from two lidars, an FMCW radar, stereo and monocular cameras, dual IMUs, wheel odometry, power data, and precise ground-truth trajectories via GNSS-PPK fusion. The dataset was created by Norlab at Université Laval and is hosted on AWS Open Data.

Use Cases

• Evaluate long-term SLAM and odometry performance based on repeated traversals of six trajectories.
• Develop traversability analysis models based on multi-season lidar, radar, and camera data in challenging terrain.
• Benchmark multi-sensor fusion for localization based on synchronized lidar, radar, camera, and IMU streams.
• Study the impact of variable weather on robot navigation based on integrated one-minute weather station measurements.

Strengths

• Includes synchronized data from multiple sensor types: two lidars, an FMCW radar, stereo and monocular cameras, and dual IMUs.
• Provides precise ground-truth trajectories generated via GNSS-PPK fusion.
• Contains repeated traversals of six trajectories of varying complexity for longitudinal evaluation.
• Features rich metadata including one-minute weather station measurements.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and total dataset size are unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Norlab, Université Laval

Collection Method

Recorded via robot platform in Forêt Montmorency.

Time Range

Multi-season coverage, but specific dates are unknown.

Freshness

Last updated date is unknown.

Geography

Forêt Montmorency, a boreal forest environment.