相关实验视频
Updated: May 17, 2025

07:36
High-Throughput Analysis of Optical Mapping Data Using ElectroMap
Published on: June 4, 2019
9.2K
使用机器学习增强实时全球电离层地图
Marcel Iten1, Shuyin Mao1, Yuanxin Pan1
1ETH Zurich, Zurich, Switzerland.
概括
机器学习改进了全球导航卫星系统 (GNSS) 的实时电离层地图,大大减少了错误. 这提高了高精度应用的准确性,并提高了单频GNSS定位性能.
科学领域:
- 地质物理学 地质物理学
- 太空科学 太空科学
- 卫星导航 卫星导航 卫星导航 卫星导航
背景情况:
- 全球电离层地图 (GIM) 对于高精度的全球导航卫星系统 (GNSS) 应用至关重要.
- 由于延迟,国际GNSS服务 (IGS) 的实时GIM (RT GIM) 的准确性低于最终GIM.
- 与最终的GIM相比,当前的IGS RT GIM显示了3.5-5.5总电子含量单位 (TECU) 的根平均平方误差 (RMSE).
研究的目的:
- 通过机器学习 (ML) 提高现有的实时GIM的准确性.
- 调查卷积神经网络 (CNN) 和条件生成对抗网络 (cGAN) 对GIM改进的有效性.
- 评估ML增强的GIM对单频GNSS定位的影响.
主要方法:
- 将CNN和cGAN模型应用于IGS组合的RT GIM和加泰罗尼亚理工大学 (UPC) GIM.
- 利用了超过13万对实时和最终GIM用于培训和验证.
- 在3.5个月的测试期内评估性能.
主要成果:
- 实现了RT GIM中平均绝对误差的30%以上的降低.
- 对于具有高垂直总电子含量 (VTEC) 值的区域,准确度明显提高了近50%.
- 在单频GNSS用户的3D定位误差中观察到高达21厘米的减少.
结论:
- 机器学习方法,特别是CNN和cGAN,显示了提高实时GIM准确性的巨大潜力.
- 开发的方法为生成更准确和精细的实时电离层产品提供了可行的解决方案.
- 增强的RT GIM可以在实时GNSS应用中提高性能.
相关概念视频
Errors in Global Positioning System
24
Global Positioning System (GPS) technology has revolutionized navigation and positioning, but its accuracy is often compromised by various errors. These errors, stemming from environmental, satellite, and receiver-related factors, require careful mitigation to ensure reliable performance across applications.Atmospheric ErrorsGPS signals travel through the Earth’s ionosphere and troposphere, introducing delays which affect accuracy. The ionosphere is strongly influenced by charged particles,...
24
Doppler Effect - II
3.3K
The Doppler effect has several practical, real-world applications. For instance, meteorologists use Doppler radars to interpret weather events based on the Doppler effect. Typically, a transmitter emits radio waves at a specific frequency toward the sky from a weather station. The radio waves bounce off the clouds and precipitation and travel back to the weather station. The radio frequency of the waves reflected back to the station appears to decrease if the clouds or precipitation are moving...
3.3K

