LuSNAR
收藏资源简介:
LuSNAR数据集由中国科学院空间应用工程与技术中心和中国科学院空间利用重点实验室创建,是一个多任务、多场景的月球表面数据集。该数据集包含9个基于Unreal Engine的模拟月球场景,每个场景根据地形起伏和物体密度划分,提供了高分辨率的立体图像对、全景语义标签、深度图、点云等数据。数据集的创建旨在支持环境感知和导航算法的开发与验证,适用于2D和3D语义分割、视觉SLAM、LiDAR SLAM、立体匹配和3D重建等任务。LuSNAR数据集的应用领域主要集中在月球探索的自主感知和导航技术,旨在提高月球探测任务的安全性和效率。
The LuSNAR dataset, developed by the Center for Space Applied Engineering and Technology of the Chinese Academy of Sciences and the Key Laboratory of Space Utilization of the Chinese Academy of Sciences, is a multi-task, multi-scenario lunar surface dataset. It includes 9 Unreal Engine-based simulated lunar scenes, each classified according to terrain undulation and object density, and provides high-resolution stereo image pairs, panoramic semantic labels, depth maps, point clouds and other data. The dataset is designed to support the development and validation of environment perception and navigation algorithms, and is suitable for tasks such as 2D and 3D semantic segmentation, visual SLAM, LiDAR SLAM, stereo matching and 3D reconstruction. The application fields of the LuSNAR dataset mainly focus on autonomous perception and navigation technologies for lunar exploration, aiming to improve the safety and efficiency of lunar exploration missions.
LuSNAR 数据集
简介
LuSNAR 数据集是一个基于多传感器(立体相机、激光雷达、IMU)的月球分割、导航和重建数据集,用于自主探索。该数据集包含 9 个基于 Unreal Engine 的月球模拟场景,每个场景根据地形起伏和物体密度进行划分。
数据内容
LuSNAR 数据集包括:
- 高分辨率立体图像对
- 全景语义标签
- 密集深度图
- 激光雷达点云
- IMU 数据
- 漫游车姿态数据
应用场景
该数据集可用于全面评估自主感知和导航系统,包括:
- 2D/3D 语义分割
- 视觉/激光雷达 SLAM
- 3D 重建
可用性
LuSNAR 数据集总大小为 108GB,包含:
- 42GB 的立体图像对
- 50GB 的深度图
- 356MB 的语义分割标签
- 14GB 的单帧点云数据(带语义信息)
数据集结构如下: plaintext ├── image1 │ ├── RGB │ │ ├── timestamp1.png │ │ ├── timestamp2.png │ │ └── ... │ ├── Depth │ │ ├── timestamp1.png │ │ ├── timestamp2.png │ │ └── ... │ └── Label │ ├── timestamp1.png │ ├── timestamp2.png │ └── ... ├── image2 │ ├── RGB │ │ ├── timestamp1.png │ │ ├── timestamp2.png │ │ └── ... │ ├── Depth │ │ ├── timestamp1.png │ │ ├── timestamp2.png │ │ └── ... │ └── Label │ ├── timestamp1.png │ ├── timestamp2.png │ └── ... ├── LiDAR │ ├── timestamp1.txt │ ├── timestamp2.txt │ └── ... ├── Rover_pose.txt └── IMU.txt
语义图像中颜色与类别编号的对应关系如下:
| 类别编号 | 类别 | 颜色 |
|---|---|---|
| 0 | 月球风化层 | BB469C |
| 1 | 撞击坑 | 7800C8 |
| 2 | 岩石 | E8FA50 |
| 3 | 山脉 | AD451F |
| 4 | 天空 | 22C9F8 |
激光雷达点云数据中类别编号与类别的对应关系如下:
| 类别编号 | 类别 |
|---|---|
| -1 | 月球风化层 |
| 0 | 撞击坑 |
| 174 | 岩石 |
文件格式
LiDAR/timestamp.txt
plaintext | x [m] | y [m] | z [m] | 类别编号 |
Rover_pose.txt
plaintext | 时间戳 [ns] | p_RS_R_x [m] | p_RS_R_y [m] | p_RS_R_z [m] | q_RS_w [] | q_RS_x [] | q_RS_y [] | q_RS_z [] | v_RS_R_x [m s^-1] | v_RS_R_y [m s^-1] | v_RS_R_z [m s^-1] | b_w_RS_S_x [rad s^-1] | b_w_RS_S_y [rad s^-1] | b_w_RS_S_z [rad s^-1] | b_a_RS_S_x [m s^-2] | b_a_RS_S_y [m s^-2] | b_a_RS_S_z [m s^-2] |
IMU.txt
plaintext | 时间戳 [ns] | w_RS_S_x [rad s^-1] | w_RS_S_y [rad s^-1] | w_RS_S_z [rad s^-1] | a_RS_S_x [m s^-2] | a_RS_S_y [m s^-2] | a_RS_S_z [m s^-2] |

- 1LuSNAR:A Lunar Segmentation, Navigation and Reconstruction Dataset based on Muti-sensor for Autonomous Exploration中国科学院空间应用工程与技术中心,中国科学院空间利用重点实验室 · 2024年



