Method for predicting destination and recommending driving route

Abstract

The invention discloses a method for predicting a destination and recommending a driving route. The method comprises the following steps: S1, acquiring historical itinerary data of a user, wherein thehistorical itinerary data of the user include the departure time, departure sites, a driving tracks, the arrival time and arrival sites; S2, analyzing the acquired historical itinerary data; S3, predicting the destination, namely predicting the position where the user might go according to the historical itinerary data, the current time and the current position of the user; S4, recommending the route, namely recommending the route according to the destination and in combination with combined data learned at a server side; and S5, processing a route result, and recommending the optimal drivingroute to the user. By the method, the destinations and the time periods for driver outgoing are clustered, the periodic law of the driver outgoing can be completely analyzed, the law is divided intoworking days and holidays, and the holidays consider national statutory holidays except weekends, and when multiple possible destinations exist in the same time period, the outgoing probability is calculated.

Description

technical field [0001] The invention relates to the field of traffic travel, and more specifically, to a method for predicting a destination and recommending a driving route. Background technique [0002] With the development of cities, driving conditions are becoming more and more complex. People are increasingly relying on driving route recommendation software. Now commonly used are navigation applications such as Baidu Map App and Gaode Map App, which provide travel destination predictions. And driving route recommendation function. The method of realization is that the car owner needs to input the address of home and work unit in advance, predict that the car owner will go to the work unit when the car owner is at home during the working hours in the morning, and predict that the car owner will go home when the car owner is at the company during the off-duty time period in the afternoon, and then according to the real-time traffic conditions and the user input Routing p...

AI Technical Summary

Problems solved by technology

[0003] 1. The predicted destination is limited, and it is only applicable to office workers with very fixed travel patterns. It can only predict the two locations of home and work. For example, the car owner who visits customer A on Wednesday morning is unpredictable

[0004] 2. The forecast time period is limited, and non-working days such as weekends and holidays are unpredictable

[0005] 3. The recommended driving route does not follow the owner's travel habits, but only considers factors such as the shortest travel time

Images