Core Concepts
The FoMo dataset aims to provide a comprehensive, multi-season dataset for evaluating robot navigation algorithms in the challenging boreal forest environment, featuring diverse sensor modalities and centimeter-accurate ground truth.
Abstract
The FoMo (Forêt Montmorency) dataset is a proposal for a comprehensive, multi-season data collection in the Montmorency Forest, Quebec, Canada. The dataset will capture a rich variety of sensory data, including lidar, radar, and navigation-grade IMU, over six diverse trajectories totaling 6 kilometers, repeated through different seasons to accumulate 42 kilometers of recorded data.
The key highlights of the FoMo dataset include:
Seasonal variations, such as changes in tree canopy and snow depth up to 2 meters, presenting new challenges for robot navigation algorithms.
A wide range of sensor modalities, featuring a navigation-grade Inertial Measurement Unit (IMU) and a radar, in addition to lidar and stereo cameras.
Centimeter-level accurate ground truth, obtained through Post Processed Kinematic (PPK) Global Navigation Satellite System (GNSS) correction, facilitating precise evaluation of odometry and localization algorithms.
A public odometry and localization leaderboard and a dedicated software suite to facilitate data manipulation and algorithm benchmarking.
The proposed dataset aims to spur advancements in autonomous navigation, enabling the development of robust algorithms capable of handling the dynamic, unstructured environments characteristic of boreal forests. The authors seek feedback from the community to make the dataset as useful as possible for the robotics research community.
Stats
The dataset will cover a total distance of 42 km, with 6 km of trajectories repeated across 7 recording sessions.
Snow accumulation height reaches up to 200 cm in open areas and 100 cm under tree canopy.
The Montmorency Forest receives about 900 mm of rainfall and 600 cm of snowfall each year.
Quotes
"The boreal forest environment increases the diversity of datasets for mobile robot navigation."
"Notably, the FoMo dataset will be distinguished by its inclusion of seasonal variations, such as changes in tree canopy and snow depth up to 2 meters, presenting new challenges for robot navigation algorithms."
"With a public odometry and localization leaderboard and a dedicated software suite, we invite the robotics community to engage with the FoMo dataset by exploring new frontiers in robot navigation under extreme environmental variations."