CN-121998370-A - Network contract vehicle order matching method based on driving direction fitting degree and dynamic weight

CN121998370ACN 121998370 ACN121998370 ACN 121998370ACN-121998370-A

Abstract

The invention relates to the technical field of intelligent transportation, in particular to a network contract vehicle order matching method based on traveling direction fitness and dynamic weight. The method comprises the steps of obtaining a direction fit score by calculating an included angle cosine value between a driving direction and an order destination direction obtained by a driver historical GPS track, dynamically adjusting weights of the direction fit score and a space distance in a comprehensive score according to order information such as a time period, a category and the like of the about-vehicle time, and finally sequencing candidate drivers based on the comprehensive score to achieve optimal matching. According to the intelligent scheduling method, through the combination of the direction fitting degree and the dynamic weight, the optimization of the receiving and driving path, the reduction of the free driving rate and the improvement of the system matching efficiency are realized, and the intelligent scheduling method is suitable for intelligent scheduling scenes of different time periods and order types.

Inventors

JIANG FENG
YUAN ZIHAN
TANG HAOZHE
DONG ZEJIAO
CAO LIPING
Abaho G. Geshen

Assignees

哈尔滨工业大学

Dates

Publication Date: 20260508
Application Date: 20260205

Claims (4)

1. The network contract vehicle order matching method based on the traveling direction fitness and the dynamic weight is characterized by comprising the following steps of: Acquiring the coordinates of the get-on points and the destination coordinates of an order to be matched, the real-time position of a candidate vehicle and a historical GPS track, and predicting the running direction angle of the candidate vehicle based on the historical GPS track; Constructing an order destination direction vector based on the coordinates of the get-on point and the destination coordinates, calculating a destination direction angle, calculating an included angle between the predicted candidate vehicle running direction angle and the order destination direction angle, and calculating a direction fit score according to the included angle; Calculating the space distance from the current position of the candidate vehicle to the point of the order on the vehicle and the distance score thereof, and obtaining a real-time traffic congestion index in a preset range of the current position of the candidate vehicle; And step four, matching decision and order distribution, namely sequencing the candidate vehicles according to the comprehensive priority grades, and distributing the to-be-matched order to the candidate vehicle driver with the highest grade.
2. The method for matching a network contract vehicle order based on traveling direction fitness and dynamic weight according to claim 1, wherein in the first step, longitude and latitude data of a vehicle point to be matched are obtained to obtain a vehicle point coordinate, longitude and latitude data of a destination are obtained to obtain a destination coordinate, and a traveling direction angle of a candidate vehicle is predicted by a linear fitting method based on a candidate vehicle history GPS track point sequence 。
3. The method for matching a network contract vehicle order based on traveling direction fitness and dynamic weight according to claim 2, wherein in the second step, a destination direction vector of the order to be matched is constructed by taking the coordinates of the upper vehicle point of the order to be matched as an origin and the destination coordinates as a destination point, and a destination direction angle is calculated Calculating the travel direction angle of the predicted candidate vehicle Direction angle with the destination of an order to be matched Included angle of (2) According to the included angle Calculating a directional fitness score The calculation formula is: 。
4. The method for matching network contract vehicle orders based on driving direction fitness and dynamic weight according to claim 3, wherein in step three, the spatial distance from the current position of the candidate vehicle to the point on the vehicle on the order to be matched is calculated Sum distance score , The method comprises the steps of obtaining real-time traffic congestion indexes in a preset range of the current position of a candidate vehicle, dynamically determining direction fitness weights according to the time period of the time of the order, the order category attribute and the real-time traffic congestion indexes And distance weight When the real-time traffic congestion index exceeds a preset threshold value, the direction fitness weight is improved To reduce the distance weight When the real-time traffic congestion index is lower than a preset threshold value, reducing the weight of the direction fitness To increase the distance weight Is based on the score of the direction fit degree Distance score Direction fitness weight Distance weight Calculating a composite priority of candidate vehicles , 。

Description

Network contract vehicle order matching method based on driving direction fitting degree and dynamic weight Technical Field The invention relates to the technical field of intelligent transportation, in particular to a network contract vehicle order matching method based on traveling direction fitness and dynamic weight. Background The core task of the network vehicle-restraining platform is to realize efficient and accurate matching of massive orders and transport capacity. Existing matching strategies are mostly based on static or instantaneous spatial proximity, e.g., giving priority to orders to vehicles that are "about car longitude" and "about car latitude" spatially closest to the passenger's "departure location". The matching strategy taking the distance as a single dimension is direct and simple in calculation, but has obvious limitation that the system can send an order to a driver who is approaching but is traveling in the opposite direction, and the driver needs to turn around or detour to take the drive, so that the driving mileage and waiting time are increased, and the resource waste and experience are reduced. Although some improved methods introduce multidimensional features such as price and running time, and adopt weighted summation to make decisions, the weights of the improved methods are usually fixed values, and cannot be dynamically adjusted according to real-time traffic conditions or order characteristics, so that further improvement of matching effect is limited. Disclosure of Invention The invention aims to overcome the defects of the prior art, and provides a network contract vehicle order matching method based on the travelling direction fitting degree and dynamic weight, which obtains a direction fitting degree score by calculating an included angle cosine value of the travelling direction and the order destination direction obtained by a driver historical GPS track; meanwhile, according to order information, such as a time period of time of vehicle-holding, category and the like, the weights of the direction fit degree and the space distance in the comprehensive score are dynamically adjusted, and finally, candidate drivers are ordered based on the comprehensive score to achieve optimal matching. In order to achieve the above purpose, the present invention adopts the following specific technical scheme: The invention provides a network contract vehicle order matching method based on traveling direction fitting degree and dynamic weight, which comprises the following steps: Acquiring the coordinates of the get-on points and the destination coordinates of an order to be matched, the real-time position of a candidate vehicle and a historical GPS track, and predicting the running direction angle of the candidate vehicle based on the historical GPS track; Constructing an order destination direction vector based on the coordinates of the get-on point and the destination coordinates, calculating a destination direction angle, calculating an included angle between the predicted candidate vehicle running direction angle and the order destination direction angle, and calculating a direction fit score according to the included angle; Calculating the space distance from the current position of the candidate vehicle to the point of the order on the vehicle and the distance score thereof, simultaneously obtaining a real-time traffic congestion index in a preset range of the current position of the candidate vehicle, dynamically determining a direction fitness weight and a distance weight by combining a time period of the order, an order category attribute and the real-time traffic congestion index, and calculating the comprehensive priority score of the candidate vehicle based on the direction fitness score, the distance score and the weights; And step four, matching decision and order distribution, namely sequencing the candidate vehicles according to the comprehensive priority grades, and distributing the to-be-matched order to the candidate vehicle driver with the highest grade. In the first step, longitude and latitude data of a boarding point of an order to be matched are obtained to obtain a boarding point coordinate, longitude and latitude data of a destination are obtained to obtain a destination coordinate, and a running direction angle of a candidate vehicle is predicted by a linear fitting method based on a candidate vehicle history GPS track point sequence。 Further, in the second step, the coordinates of the boarding point of the order to be matched are taken as an origin point, the coordinates of the destination are taken as an end point, a destination direction vector of the order to be matched is constructed, and a destination direction angle is calculatedCalculating the travel direction angle of the predicted candidate vehicleDirection angle with the destination of an order to be matchedIncluded angle of (2)According to the included angleCalculating a directional fitness sc