Potential car sharing group identification method based on position prediction

Abstract

The invention puts forward a method for selecting a matched potential car sharing group on the basis of the driving path of a car owner. A trie tree is used for processing the problem of high space complexity of a prediction model based on matrix processing, and an escaping mechanism is used for reasonably processing the problem of zero frequency. On the basis, an association rule is reasonably fused, connection among positions is enhanced, the amount of candidate position sets is reduced, algorithm operation efficiency is improved, and position prediction accuracy is further improved. By use of the method which is put forward by the invention, on the basis of the optimal driving path of the car owner, passenger selection is carried out, the intentions of the car owner are paid attention, and the requirements of the passenger are met.

Description

Technical field [0001] The invention belongs to the field of vehicle sharing, and specifically relates to a potential sharing group identification method based on position prediction. Background technique [0002] Carpooling, that is, carpooling and matching. The problem of car sharing and matching refers to the agreement between the owner and the passengers taking the ride. The owner is willing to let the passengers drive in the vehicle they drive and get the passengers off at the place deemed appropriate by both parties. The cost of sharing Travel mode shared by passengers (including car owners). In 1982, Daganzo first began to put forward the idea of ​​"shared rides" to solve the problem of balancing traffic flow distribution in the network. Although his hypothetical theory was not successful, through this attempt, it officially opened the scientific development of car sharing. In the prologue. [0003] The first thing to do in car sharing is to match passengers and select th...

AI Technical Summary

Problems solved by technology

[0008] The general Markov position prediction model based on matrix has the problems of high space complexity and zero frequency of data;

[0009] Traditional position prediction models are faced with the problems of low algorithm operation efficiency and long time consumption;

[0010] The solution to the car-sharing problem is generally to select the passengers first, and then choose the driving route of the car owner. This solution often pays too much attention to the choice of passengers, so that the original driving route of the car owner deviates greatly, ignoring the wishes of the car owner

Images