基于局域定位场景聚类的虚拟应答器布局优化方法研究

Research on Optimized Virtual Balise Placement Method Based on Local Positioning Scenario Clustering

基于全球导航卫星系统的列车定位是下一代列车控制系统的重要技术方向之一,卫星定位驱动的虚拟应答器技术是实现卫星定位与既有列控系统兼容的有效途径。重点关注虚拟应答器的选址布局优化设计问题,结合虚拟应答器捕获与报文触发对卫星定位性能的需求,从候选虚拟应答器布设位置的局域卫星定位观测性能特征出发,运用动态时间规划和具有噪声的基于密度的聚类方法,提出一种铁路沿线局域定位场景聚类策略,基于聚类结果,重点针对卫星信号观测受限区段,构建一种基于麻雀搜索改进遗传算法的虚拟应答器布局优化方法。采用实际线路数据及现场试验数据,对所提出的局域定位场景聚类策略进行了验证,并通过与传统优化算法的对比,验证所提布局优化方法的能力。所得结果对于辅助实际线路虚拟应答器布设方案设计与实施具有参考意义。

Train positioning based on Global Navigation Satellite System(GNSS)is a major technical direction for the next generation train control systems.The Virtual Balise(VB)enabled by GNSS offers an effective way to achieve compatibility of satellite-based positioning with the existing train control systems.In this paper,the optimized design of VB placement was investigated.According to the requirements of VB capturing and VB telegram triggering for satellite positioning performance,the local satellite positioning observing performance characteristics of the candidate VB location were focused in the design.By using the Dynamic Time Warping(DTW)and Density-Based Spatial Clustering of Applications with Noise(DBSCAN)methods,a local positioning scenario clustering strategy for a specific railway line was presented.With the clustering results,a Sparrow Search Algorithm(SSA)enhanced Genetic Algorithm(GA)method was proposed for VB placement optimization in GNSS-constrained railway sections.Data sets of the practical railway line and field experiments were employed to validate the presented scenario clustering strategy.The results of comparisons with conventional optimization solutions validated the performance of the proposed VB placement optimization method.The contribution of this paper is of great reference significance for the design and implementation of VB placement schemes in practical railway engineering.