SmartPNT-POS

Raw GNSS and INS measurements for research
The integration of GNSS and SINS (GNSS/SINS) stands as the cornerstone of current positioning and navigation technology. Raw GNSS and SINS measurements are vital to facilitate the process of GNSS/SINS research. This has also been proved in other public datasets, such as ImageNet, the perfect or dominate dataset in computer vision (CV), and KITTI, the pioneer in autonomous driving (AD). In the field of GNSS research, the importance of datasets has been recognized from the early stages. The International GNSS Service (IGS) has deployed tracking stations worldwide and released datasets for geoscience exploration and research. In addition, there are dense regional networks of Continuously Operating Reference Stations (CORS) around the world. Notable examples include CORS in America, the EUREF Permanent GNSS Network (EPN) in Europe, and the GNSS Earth Observation Network System (GEONET) in Japan, which have significantly contributed to seismology, PPP-RTK, and modeling the regional ionosphere and troposphere.
However, lack of high-quality public dataset in GNSS/SINS research would slow down the progress of algorithm research, especially for the complex urban scenes in real applications. Therefore, we release here the dataset collected from multiple scenes (open sky, urban area), moving platforms (vehicle, UAV, ship), and high precision IMUs (ISA-100C, HG 4930, SPAN FSAS), accompanied with high accuracy ground truth obtained by NovAtel's Inertial Explorer (IE) post processing software version 8.90.

Content
Data is organized into folders with names in this format: {Data Number}YYYYMMDD{IMU Name}{Platform}{Scene}. For example, Data01_20230612_ISA-100C_Vehicle_Opensky indicates that the data was the first set, collected in the open sky environment in June 12, 2023, using the ground vehicle with ISA-100C equipped. Each folder contains one or more sets of the following files.
• BASE.yyO: RINEX observation file for the base station,
• ROVE.yyO: RINEX observation file for the rover station,
• {IMU Name}.IMR: Raw IMU files in IMR formats (the C++ code to read IMR files are provided),
• *.yyP: Broadcast ephemeris,
• *.SP3: Precise ephemeris product, downloaded from the IGS center of Wuhan University,
• *.CLK: Precise clock product, downloaded from the IGS center of Wuhan University,
• ROVE_GroundTruth.TXT: Ground truth for the antenna phase center of GNSS,
• {IMUName}_GroundTruth.TXT: Ground truth for the IMU center,
• Trajectory.kml: the trajectory shown in Google Earth,
• README.xml: including the base station coordinates, the lever arm offsets from IMU to GNSS antenna, and the IMU rot angles.
• Figures (a folder): including the trajectory of the data and the solution quality of the ground truth.

Disclaimer
This dataset, including sensor data and ground truth data, is collected in public areas.

链接：https://pan.baidu.com/s/1OWMI1mslzkSXmu8InhrmbA
提取码：dy8s

面向科研的原始全球导航卫星系统（GNSS）与捷联惯导系统（SINS）测量数据
GNSS与SINS的组合技术是当前定位与导航技术的基石。原始GNSS与SINS测量数据对于推进GNSS/SINS相关研究至关重要，这一点已在诸多公开数据集上得到验证——例如计算机视觉（CV）领域的标杆数据集ImageNet，以及自动驾驶（AD）领域的先驱数据集KITTI。在GNSS研究领域，数据集的重要性从早期便已得到认可：国际GNSS服务（IGS）已在全球布设跟踪站，并发布用于地球科学探索与研究的数据集；此外，全球范围内还部署了密集的连续运行参考站（CORS）区域网络，典型代表包括美国的CORS、欧洲的EUREF永久GNSS网络（EPN）以及日本的GNSS地球观测网络系统（GEONET），这些网络在地震学、PPP-RTK以及区域电离层与对流层建模领域均作出了重要贡献。

然而，GNSS/SINS研究领域缺乏高质量公开数据集，这会延缓算法研究的推进速度，尤其是针对实际应用中的复杂城市场景。为此，我们发布本数据集，其采集自多种场景（开阔空域、城区）、多种移动平台（地面车辆、无人机（UAV）、船舶），搭载了高精度惯性测量单元（IMU）：ISA-100C、HG 4930与SPAN FSAS，并配套由诺瓦泰（NovAtel）公司版本为8.90的Inertial Explorer（IE）后处理软件生成的高精度真值数据。

数据组织与内容
数据以如下格式命名的文件夹进行组织：{数据编号}YYYYMMDD{IMU名称}{平台}{场景}。例如，Data01_20230612_ISA-100C_Vehicle_Opensky 代表该数据集为第一组数据，于2023年6月12日在开阔空域环境下采集，由搭载ISA-100C惯性测量单元的地面车辆获取。每个文件夹包含一个或多个以下类型的文件：
• BASE.yyO：基站的RINEX观测文件
• ROVE.yyO：流动站的RINEX观测文件
• {IMU名称}.IMR：IMR格式的原始IMU文件（附带用于读取IMR文件的C++代码）
• *.yyP：广播星历文件
• *.SP3：精密星历产品，从武汉大学IGS中心下载获取
• *.CLK：精密钟差产品，从武汉大学IGS中心下载获取
• ROVE_GroundTruth.TXT：GNSS天线相位中心的真值数据
• {IMU名称}_GroundTruth.TXT：IMU中心的真值数据
• Trajectory.kml：可在Google Earth中查看的轨迹文件
• README.xml：包含基站坐标、IMU至GNSS天线的杆臂偏移量以及IMU旋转角参数
• Figures（文件夹）：包含数据集轨迹与真值解算质量相关的可视化图表

免责声明
本数据集（含传感器数据与真值数据）采集自公共区域。

链接：https://pan.baidu.com/s/1OWMI1mslzkSXmu8InhrmbA
提取码：dy8s

搜集汇总

背景与挑战

背景概述

SmartPNT-POS数据集提供了GNSS和SINS的原始测量数据，支持定位和导航技术的研究。数据集包含多种场景和移动平台的数据，并配有高精度的地面真实值，适用于复杂城市环境下的算法研究。