Bipartite graph-based shared network driver and passenger matching method

Abstract

The invention discloses a shared network driver and passenger matching method based on a bipartite graph. The method comprises the steps that a passenger submits a ride sharing request and uploads theride sharing request to a cloud server, the cloud server receives the request of each passenger and the positions and passenger carrying information of all taxies in the taxi set; the taxies meetingthe maximum number of passagners are screened according to a request of each passagner, wherein the taxis meet the constraint of passagners on the taxis and the time constraint to acquire candidate taxi sets corresponding to requests of the passengers; performing many-to-many matching on all the requests and all the candidate taxies to obtain optimal matching between the taxies and the requests, and orders are generated according to matching results; for unaccepted requests, the cloud server matches the unaccepted requests cyclically until all requests are accepted or no taxi can meet requestconstraints. The invention can be suitable for many-to-many ride-sharing matching of multiple passengers and multiple vehicles, the ride-sharing rate can be substantially increased, and traffic pollution is reduced.

Description

technical field [0001] The present application relates to the technical fields of computer and transportation, in particular to a bipartite graph-based sharing network driver-passenger matching method. Background technique [0002] The surge of urban population and the increasing demand for travel are one of the reasons for urban traffic congestion. In recent years, local governments have encouraged commuters to use public transport services. With the vigorous development of artificial intelligence and autonomous driving technology, the future mode of transportation will be transformed into a shared transportation mode dominated by self-driving taxi services. Different from traditional taxis, the self-driving taxi system must rely on an effective carpool matching algorithm to determine the matching between vehicles that still have vacancies and new passengers. [0003] Existing taxi dynamic share ride matching methods include: [0004] "Multi-level Screening Method for Ca...

AI Technical Summary

Problems solved by technology

The disadvantage is that the patent performs carpooling and matching according to user preferences, but the method of screening candidate sets using the principle of proximity can only provide users with partial options

The method of matching according to the user's choice is only suitable for solving the optimal matching situation of a single passenger or a single vehicle, which often leads to the failure of other passengers to obtain an ideal carpooling plan, unable to achieve cluster optimization of carpooling, and failing to achieve the overall optimization, failed to effectively improve the carpooling rate

Images