CN-122019516-A - Navigation track data processing method and system

CN122019516ACN 122019516 ACN122019516 ACN 122019516ACN-122019516-A

Abstract

The invention relates to the technical field of ground letter data processing, in particular to a navigation track data processing method and system. The method comprises the steps of obtaining track data uploaded by a user, enabling each track to comprise a time stamp, a position coordinate sequence and user identification information, conducting superposition analysis on the tracks and a preset physical obstacle layer, identifying obstacle crossing track segments, calculating physical violation scores, constructing a user credibility model based on user historical track compliance and data quality, generating user credibility scores, calculating path rationality scores by combining road network, topography, land utilization and traffic rules, conducting clustering analysis on the track data, fusing the three scores according to clustering types and user group characteristics, generating comprehensive rationality scores, setting a threshold according to score statistics results, eliminating abnormal tracks, and obtaining an effective track data set. The invention can realize multi-dimensional evaluation and screening of track data and improve the reliability of navigation path recommendation and map data updating.

Inventors

WANG HAICHAO
LIU JIAOHONG
GAO YANQIU

Assignees

艾迪普科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20251224

Claims (9)

1. A navigation track data processing method, characterized by comprising: acquiring a track data set uploaded by a user, wherein each track data set comprises a time stamp sequence, a position coordinate sequence and corresponding user identification information; carrying out superposition analysis on the track data and a preset physical obstacle layer, identifying track segments crossing physical obstacles in the track, and calculating the physical violation score of each track according to the crossing times, the crossing distance and the obstacle severity; based on the historical track compliance and track data quality of each user, constructing a user credibility model, and generating a corresponding user credibility score for reflecting the reliability of the uploading track of the user; Calculating a path rationality index for each track based on a preset road network, terrain data, land utilization information and traffic rule constraint, and generating a path rationality score of each track; clustering the track data according to the user credibility score, and identifying track groups with similar physical violation features or similar path behavior features so as to mine abnormal track modes or suspicious path segments; The physical violation degree score, the user credibility score and the path rationality score are subjected to weighted fusion to generate a comprehensive rationality score of each track, wherein the weighting coefficient is adaptively adjusted according to the track clustering category; and setting a threshold according to the comprehensive rationality score, and rejecting or marking the track data lower than the threshold as abnormal so as to obtain a quality-screened effective track data set.
2. The method of claim 1, wherein the step of calculating a physical violation score for each track comprises: carrying out space superposition analysis on the track and the physical barrier layer, and identifying a track section crossing the barrier; Extracting characteristic parameters of each track segment, wherein the characteristic parameters at least comprise the number of times of crossing, the crossing distance, the type of obstacle and the crossing speed; and inputting the characteristic parameters into a pre-trained physical violation evaluation model, and outputting corresponding physical violation scores by the model.
3. The method of claim 1, wherein the step of constructing a user confidence model comprises: Performing compliance analysis on the track data, and calculating the physical violation rate and the data integrity of the track; taking the physical violation rate, the data integrity and the track uploading frequency as user characteristic parameters; And inputting the user characteristic parameters into a user credibility evaluation model, and outputting a user credibility score by the model.
4. The method of claim 1, wherein the step of generating a path rationality score for each track comprises: Matching the position information of the track with a preset road network, and calculating the space matching degree of the track and the central line of the road; analyzing the gradient change of the track based on the terrain data, and judging whether the track has the terrain rationality; Judging whether the track meets traffic rule compliance according to traffic rule constraint; and taking the space matching degree, the terrain rationality, the land utilization rationality and the traffic rule compliance as characteristic parameters to input a path rationality evaluation model, and outputting a corresponding path rationality score by the model.
5. The navigation trajectory data processing method according to claim 1, wherein the step of clustering the trajectory data includes: taking the physical violation degree score, the path rationality score and the user credibility score of each track as characteristic parameters, and carrying out standardized processing on the characteristic parameters to eliminate dimension differences; and calculating a distance matrix among tracks based on the feature similarity among tracks, and dividing track data into a plurality of track groups by adopting a clustering algorithm.
6. The method of claim 1, wherein the step of weighting and fusing the physical offensiveness score, the user credibility score, and the path rationality score to generate a composite rationality score for each track comprises: Taking the physical violation degree score, the user credibility score and the path rationality score corresponding to each track as input parameters; And carrying out fusion calculation on the scores according to a preset weighted fusion model, wherein the weight coefficient of each score is self-adaptively adjusted according to the clustering category to which the track belongs.
7. The method for processing navigation track data according to claim 6, wherein the step of adaptively adjusting the weight coefficient of each score according to the cluster category to which the track belongs comprises: counting the distribution characteristics of physical violation scores, user credibility scores and path rationality scores in each track cluster category; determining the confidence level of each scoring dimension according to the standard deviation of the scores in different clustering categories; And dynamically adjusting the weight coefficient in the weighted fusion model according to the confidence level.
8. The method of claim 1, wherein the step of setting a threshold according to the composite rationality score, and rejecting or marking trajectory data below the threshold as abnormal, thereby obtaining a quality-filtered effective trajectory dataset, comprises: Counting comprehensive rationality score distribution characteristics of all tracks; Calculating an abnormal threshold range according to the mean value and standard deviation of the scores; When the track score is lower than the standard deviation of the mean value minus a preset multiple, marking the track as abnormal; And eliminating all abnormal tracks to obtain an effective track data set.
9. A navigation track data processing system for use in the navigation track data processing method of any of claims 1-8, the system comprising: The data acquisition module is used for acquiring a track data set uploaded by a user, wherein each track data set comprises a time stamp sequence, a position coordinate sequence and corresponding user identification information; The physical verification module is used for carrying out superposition analysis on the track data and a preset physical barrier layer, identifying track segments which pass through physical barriers in the tracks, and calculating the physical violation score of each track according to the number of passes, the passing distance and the barrier severity; the user credibility evaluation module is used for constructing a user credibility model based on the historical track compliance and track data quality of each user, and generating a corresponding user credibility score for reflecting the reliability of the uploading track of the user; The track rationality evaluation module is used for calculating a track rationality index for each track based on a preset road network, terrain data, land utilization information and traffic rule constraint and generating a track rationality score of each track; the track cluster analysis module is used for carrying out cluster processing on track data according to the user credibility score, and identifying track groups with similar physical violation characteristics or similar path behavior characteristics so as to mine abnormal track modes or suspicious path segments; the weighted fusion module is used for carrying out weighted fusion on the physical violation degree score, the user credibility score and the path rationality score to generate a comprehensive rationality score of each track, wherein the weighted coefficient is adaptively adjusted according to the track clustering type; And the abnormal track screening module is used for setting a threshold value according to the comprehensive rationality score, and eliminating or marking track data lower than the threshold value as abnormal so as to obtain an effective track data set subjected to quality screening.

Description

Navigation track data processing method and system Technical Field The invention relates to the technical field of ground letter data processing, in particular to a navigation track data processing method and system. Background Along with the popularization of mobile internet, positioning technology (GPS, beidou) and intelligent terminals, massive users can continuously generate track data when using navigation application, and the data are important bases of map construction, path planning and travel analysis. However, the existing track data generally has the problem of group data pollution, part of tracks pass through physical barrier areas such as rivers, buildings, railway isolation belts and the like due to signal drift or uploading delay, and individual users can upload abnormal data due to illegal behaviors such as crossing lines, retrograde running and the like, so that map misjudgment and navigation misguidance are easily caused after a large number of abnormal tracks are acquired by the system. In the prior art, methods such as track smoothing, speed filtering or anomaly detection based on a single track are mostly adopted, but the methods lack comprehensive consideration of track physical rationality, user credibility and group consistency, and are difficult to effectively identify the anomaly track under complex terrain and large-scale data scenes, so that the accuracy and reliability of a navigation system are affected. Disclosure of Invention In order to make up for the defects, the invention provides a navigation track data processing method and system, which can realize multi-dimensional comprehensive evaluation and abnormal screening of track data aiming at the problem of group data pollution of the existing track data, thereby improving the reliability of a navigation system and the accuracy of path planning. In a first aspect, the present invention provides a method for processing navigation track data, including: acquiring a track data set uploaded by a user, wherein each track data set comprises a time stamp sequence, a position coordinate sequence and corresponding user identification information; carrying out superposition analysis on the track data and a preset physical obstacle layer, identifying track segments crossing physical obstacles in the track, and calculating the physical violation score of each track according to the crossing times, the crossing distance and the obstacle severity; based on the historical track compliance and track data quality of each user, constructing a user credibility model, and generating a corresponding user credibility score for reflecting the reliability of the uploading track of the user; Calculating a path rationality index for each track based on a preset road network, terrain data, land utilization information and traffic rule constraint, and generating a path rationality score of each track; clustering the track data according to the user credibility score, and identifying track groups with similar physical violation features or similar path behavior features so as to mine abnormal track modes or suspicious path segments; The physical violation degree score, the user credibility score and the path rationality score are subjected to weighted fusion to generate a comprehensive rationality score of each track, wherein the weighting coefficient is adaptively adjusted according to the track clustering category; and setting a threshold according to the comprehensive rationality score, and rejecting or marking the track data lower than the threshold as abnormal so as to obtain a quality-screened effective track data set. Preferably, the step of calculating a physical violation score for each track comprises: carrying out space superposition analysis on the track and the physical barrier layer, and identifying a track section crossing the barrier; Extracting characteristic parameters of each track segment, wherein the characteristic parameters at least comprise the number of times of crossing, the crossing distance, the type of obstacle and the crossing speed; and inputting the characteristic parameters into a pre-trained physical violation evaluation model, and outputting corresponding physical violation scores by the model. Preferably, the step of constructing the user credibility model comprises: Performing compliance analysis on the track data, and calculating the physical violation rate and the data integrity of the track; taking the physical violation rate, the data integrity and the track uploading frequency as user characteristic parameters; And inputting the user characteristic parameters into a user credibility evaluation model, and outputting a user credibility score by the model. Preferably, the step of generating a path rationality score for each track comprises: Matching the position information of the track with a preset road network, and calculating the space matching degree of the track and the central line of the road; analyzing the