CU-Multi
收藏资源简介:
CU-Multi是一个多机器人协作感知数据集,由科罗拉多大学波尔得分校的自主机器人与感知小组收集。该数据集包含在两个大型户外场地进行的四次同步运行,具有对齐的起始时间和受控的轨迹重叠,复制了机器人团队的独特视角。数据集包括RGB-D传感、RTK GPS、语义LiDAR和精确的地面真实里程计。通过结合重叠变化和密集的语义注释,CU-Multi为多机器人协作感知任务的可重复评估提供了坚实的基础。
CU-Multi is a multi-robot collaborative perception dataset collected by the Autonomous Robotics and Perception Group at the University of Colorado Boulder. This dataset comprises four synchronized runs conducted at two large-scale outdoor venues, with aligned start times and controlled trajectory overlaps to replicate the unique viewpoints of robot teams. It includes RGB-D sensing data, RTK GPS, semantic LiDAR, and accurate ground-truth odometry. By combining varying overlaps and dense semantic annotations, CU-Multi provides a solid foundation for the reproducible evaluation of multi-robot collaborative perception tasks.




