Customized bus route generation method based on trajectory clustering technology

Abstract

The invention belongs to the field of intelligent transportation, and relates to a customized bus route generation method based on a track clustering technology. The method comprises the following steps: acquiring taxi GPS track data; calculating the trajectory similarity from three dimensions of direction, geographic interval and trajectory sub-fragment, and fusing the features by considering domain knowledge and calculation complexity to obtain the trajectory similarity capable of reflecting the three features; based on the calculated trajectory similarity, using a DBSCAN model to cluster the trajectory set, and based on a multi-objective optimization method VIKOR, optimizing Eps and MinPts parameters in the DBSCAN; and based on the clustering results, carrying out potential line extraction on each clustering result. The method can be effectively applied to taxi GPS track data, and the central track of the clustering result is extracted, so that the final potential route is extracted.

Description

technical field [0001] The invention belongs to the field of intelligent transportation, and relates to a method for generating customized bus lines based on trajectory clustering technology. Background technique [0002] In recent years, a new type of surface bus service model called customized bus is emerging in many countries. Customized bus is a demand-responsive bus based on the concept of car sharing. Route planning is one of the keys to customized bus research. At present, most of the customized bus routes opened for travel needs in many regions are opened by a combination of questionnaire survey and appointment, while others are opened under the guidance of policies or cooperation with bus companies. However, the method of online questionnaire survey has the following problems: (1) low reliability: it is difficult to cover the overall commuting needs of urban residents (2) poor timeliness: it is difficult to make real-time adjustments according to changing travel ne...

AI Technical Summary

Problems solved by technology

However, the method of online questionnaire survey has the following problems: (1) low reliability: it is difficult to cover the overall commuting needs of urban residents (2) poor timeliness: it is difficult to make real-time adjustments according to changing travel needs

(3) cumbersome and expensive: manually designing lines not only consumes a lot of manpower and time costs

[0007] 2. The clustering results are difficult to apply. The clustering results of the general method have relatively discrete starting and ending points. It is difficult for researchers to extract OD lines with centralized travel from the clustering results as a reference for the potential opening of customized bus lines.

Images