Trip segment identification method based on GPS trajectory data

Abstract

The invention discloses a trip segment identification method based on GPS trajectory data. The method mainly includes the following steps of data cleaning and preprocessing, identification of trip endpoints, removal of false trip endpoints and merging of trip segments. On the basis of integrated advantages of the rule method and the clustering method, the method provides data cleaning and preprocessing processes, noise generated during the identification of trip endpoints and the merging of trip segments are effectively removed, the algorithm is simple and effective and the identification result is accurate and reliable. The distance between adjacent intervals of a trajectory point and the time difference are used for calculating the instantaneous velocity of the point, and the product of the velocity and the time is used for obtaining the trip distance. The method is suitable for trip segment identification based on the GPS trajectory data, GPS trajectory data mining algorithm systems are enriched, and a theoretical support is provided for large-scale travel investigation and promotion based on smart phones.

Description

technical field [0001] The invention relates to computer identification technology, in particular to a method for identifying travel segments based on GPS trajectory data. Background technique [0002] Traditional resident travel surveys generally use face-to-face interviews, paper questionnaires, emails, and telephone calls. These methods require respondents to fill in detailed travel characteristics, and they are filled in after the trip for a day or longer. The burden on the applicant is relatively large, and there are large errors in the data filled in. For example, 1) Travelers usually approximate the travel time as 5 minutes, 10 minutes or even 15 minutes, etc., and do not provide information such as the actual travel route; Between 20%-30%, these limit the accuracy of travel data. [0003] In recent years, with the rapid popularization of smart phones, the development of GPS positioning technology has made it possible to conduct travel surveys based on mobile phone ...

AI Technical Summary

Problems solved by technology

[0005] 1) Since there may be errors in the GPS signal, the existing method directly uses the speed value in the GPS recorded data for correlation calculation, and there will be a large deviation;

[0006] 2) Lack of perfect raw data cleaning and preprocessing methods, such as identification and removal of signal drift points;

[0007] 3) After the travel endpoint is identified, there is a lack of further methods for judging and removing false endpoints;

[0008] 4) The travel segment is not merged according to the traditional travel definition

[0009] Therefore, there is a large deviation between the identified travel segment and the real trip, and the method is not practical

Images