FoMo (Forêt Montmorency) 数据集
收藏资源简介:
FoMo数据集是由加拿大魁北克市的拉瓦尔大学北方机器人实验室和多伦多大学机器人研究所联合创建,旨在为移动机器人导航提供一个多季节、多样化的数据集。该数据集位于蒙莫朗西森林,覆盖6公里的六条不同轨迹,通过不同季节的重复记录,累计42公里的数据。数据集包含多种传感器数据,如激光雷达、雷达和导航级惯性测量单元,特别强调季节性变化,如树冠变化和高达2米的积雪深度,为机器人导航算法带来新的挑战。此外,数据集提供厘米级精确的地面实况,通过后处理动态(PPK)全球导航卫星系统(GNSS)校正获得,旨在推动自主导航领域的进步。数据集的应用领域包括自主导航、地图构建和语义分割等,旨在解决极端环境变化下的机器人导航问题。
The FoMo dataset was jointly created by the Northern Robotics Laboratory at Laval University in Quebec City, Canada, and the University of Toronto Robotics Institute, designed to provide a multi-seasonal, diverse dataset for mobile robot navigation. Located in Montmorency Forest, the dataset covers six distinct trajectories spanning 6 kilometers, with repeated recordings across different seasons resulting in a cumulative 42 kilometers of valid data. It includes various types of sensor data such as LiDAR, radar, and navigation-grade inertial measurement units (IMUs), with a particular emphasis on seasonal changes including canopy variations and snow depths up to 2 meters, which poses novel challenges for robot navigation algorithms. Additionally, the dataset offers centimeter-accurate ground truth obtained via post-processed kinematic (PPK) Global Navigation Satellite System (GNSS) corrections, aiming to promote advancements in the field of autonomous navigation. Application areas of the dataset include autonomous navigation, mapping, and semantic segmentation, among others, with the objective of addressing robot navigation problems under extreme environmental variations.




