基于轨道不平顺匹配的多源融合列车定位方法
Onboard multi-source fusion train positioning method based on railway track irregularity matching
周禹昆 1陈起金 1牛小骥1
作者信息
- 1. 武汉大学 卫星导航定位技术研究中心,湖北 武汉 430000
- 折叠
摘要
列车定位是保障铁路安全高效运营的基础,面向先进列车运行控制系统高精度、连续、可靠的位置服务需求,研究高精度、高可靠、低成本的匹配定位源提升列车定位系统的稳健性和准确度具有重要的现实意义.本文提出一种无布设轨道特征信号-轨道不平顺,利用惯性导航系统感知和测量轨道不平顺信号实现匹配定位,结合列车运动约束,实现了无布设情况下的全自主、全天候、全地域的连续可靠的组合定位.为了验证该方案的可行性,本文在真实铁路线路上进行了列车车载试验.试验结果表明:一个典型的战术级IMU可以实现 0.6 m(98.46%置信度)的列车定位精度.该方法为多传感器融合的列车定位系统提供了一种新的定位信号源,可以用于提高车载列车定位系统的准确性和稳健性.
Abstract
Train positioning is essential for ensuring the safe and efficient operation of railways.To meet the stringent requirements of advanced rail transit control systems,it is crucial to research the high-precision,reliable,and cost-effective matching positioning signal source that improves the robustness and accuracy of train positioning systems.This paper proposes a novel railway track feature signal-railway track irregularity,which does not need the auxiliary trackside equipment.An inertial navigation system(INS)is used to sense and measure these track irregularity signals to achieve matching positioning.The proposed tracking irregularity matching/INS integrated navigation scheme,com-bined with train motion constraints,enables autonomous and reliable positioning through the entire region without the need for trackside equipment.Train-borne experiments were carried out to validate the feasibility of the proposed method.The results demonstrate that a positioning accuracy of 0.6 m positioning accuracy with 98.46%confidence can be achieved using a typical tactical-grade inertial measurement unit(IMU).This method provides a new signal source to enhance the robustness and accuracy of onboard multi-sensor fusion train positioning systems.
关键词
列车定位/轨道不平顺/特征匹配/车载传感器定位系统/惯性导航系统/组合导航/多源融合/列车运动约束Key words
train positioning/track irregularity/feature matching/onboard sensor positioning system/inertial navi-gation system/integrated navigation/multi-source fusion/train motion constraint引用本文复制引用
基金项目
国家自然科学基金项目(41904019)
国家自然科学基金项目(41674038)
出版年
2024
国家科技期刊平台
NSTL
万方数据