CN115049105A - Taxi track big data driven passenger carrying route recommendation method, device and medium - Google Patents

Abstract

The embodiment of the application discloses a taxi track big data driven passenger carrying route recommendation method, device and medium, which are used for rapidly planning the shortest route from a starting point to a terminal point for a vehicle. The method in the embodiment of the application comprises the following steps: acquiring a vehicle motion track data set of a target area; generating road network node data according to the vehicle motion track data set; generating a route recommendation model according to the road network node data, wherein the route recommendation model is fused with an heuristic algorithm and an optimization algorithm; acquiring a starting point and an end point of a user, wherein the starting point and the end point are positioned in the target area; and generating a recommended route from the starting point to the end point according to the route recommendation model.

Description

Method, device and medium for passenger route recommendation driven by taxi trajectory big data

technical field

The embodiments of the present application relate to the technical field of data processing, and in particular, to a method, device, and medium for recommending passenger routes driven by taxi trajectory big data.

Background technique

In the era of big data, with the rapid development of digital technology, the pace of life has become faster and faster, and with it, people's demand for travel has become higher and higher. In order to meet people's needs for efficient travel, a passenger route recommendation method driven by big data of taxi trajectories emerges as the times require.

In the prior art, the traditional method for recommending passenger routes driven by big data of taxi trajectories is usually based on greedy algorithm or genetic algorithm, which needs to visit a large number of nodes to calculate the shortest route. Due to the need to access a large number of nodes, it is easy to cause problems such as "high I/O overhead and high memory consumption". At the same time, in the case of a huge amount of calculation and limited computing power, it takes a lot of computing time to get the shortest route, which leads to problems such as "long running time and low computing efficiency".

In addition, blind cruising in urban traffic can easily lead to problems such as "high fuel consumption and serious traffic congestion". This application proposes a Gurobi optimization algorithm integrating angle and A ^* algorithm, which is applied to recommend optimal passenger routes for taxi trajectory big data in complex urban road networks. First, a method for extracting road network nodes based on taxi GPS directions is proposed to solve the problem of difficulty in extracting road network nodes from taxi GPS trajectory data. Secondly, an angle-based sharp point elimination method (ASPE) is constructed to optimize the search ability of the Gurobi algorithm to find the shortest route. Thirdly, a Gurobi optimization algorithm (A-Gurobi) based on the A ^* algorithm is designed, which uses the heuristic function of the A ^* algorithm to enhance the fast guidance ability from the starting point to the destination and improve the execution efficiency of the Gurobi algorithm.

SUMMARY OF THE INVENTION

The embodiments of the present application provide a method, device, and medium for recommending a passenger-carrying route driven by taxi trajectory big data, which can quickly plan the shortest route from the starting point to the ending point for the vehicle.

A first aspect of the embodiments of the present application provides a taxi trajectory big data-driven passenger route recommendation method, including:

Obtain the vehicle motion trajectory dataset of the target area;

generating road network node data according to the vehicle motion trajectory data set;

Generate a route recommendation model according to the road network node data, and the route recommendation model integrates a heuristic algorithm and an optimization algorithm;

Obtain the starting point and the ending point of the user, and the starting point and the ending point are located in the target area;

A recommended route from the start point to the end point is generated according to the route recommendation model.

Optionally, the generating a route recommendation model according to the road network node data includes:

The heuristic of the A ^* algorithm is integrated into the Gurobi algorithm to generate the A-Gurobi algorithm;

A route recommendation model is generated according to the road network node data and the A-Gurobi algorithm, wherein the road network node data is used as a parameter of the route recommendation model.

Optionally, the generating the recommended route from the start point to the end point according to the route recommendation model includes:

Connect the starting point with each road network node and the end point in the road network node data of the route recommendation model, respectively, to obtain a plurality of line segments, wherein each road network node corresponds to a line segment, and the starting point is connected to all the line segments. The line segment formed by the end point is the target line segment;

Calculate the degrees of the included angles formed by the target line segment and other line segments respectively;

Delete the road network node corresponding to the line segment whose angle formed by the target line segment is greater than or equal to the first preset angle to obtain the recommended road network node;

A recommended route is generated according to the recommended road network node, the starting point and the ending point.

Optionally, the generating the road network node data according to the vehicle motion trajectory data set includes:

Extracting a target movement trajectory data set from the vehicle movement trajectory data set, the operation status of the target movement trajectory data set is continuously empty vehicle-passenger-passenger, and the operation status includes empty vehicle and passenger;

Generate road network node data according to the target motion trajectory data set.

Optionally, the generating the road network node data according to the target motion trajectory data set includes:

Extract the time, coordinates and direction of each point from the target motion trajectory data set;

Determine the road network node according to the coordinates and direction of each point;

Generate road network node data according to road network nodes, and the road network node data includes a weighted undirected graph and an adjacency matrix.

Optionally, the determining the road network node according to the coordinates and direction of each point includes:

For any target motion trajectory data, determine the adjacency relationship between points according to the time of each point;

Calculate the direction change amount according to the adjacency relationship between the points and the coordinates of each point, and the direction change amount is the angle of two line segments formed by one point and two adjacent points respectively;

When the amount of change in the appearance direction is greater than or equal to the second preset angle, a point adjacent to the other two points among the corresponding three points is determined as a road network node.

Optionally, the generating the road network node data according to the road network node includes:

Determine the adjacency relationship between road network nodes;

Calculate the distance between adjacent road network nodes according to the coordinates of each road network node;

Generate an adjacency matrix and a weighted undirected graph according to the adjacency relationship between road network nodes and the distance between adjacent road network nodes;

Generate road network node data according to the weighted undirected graph and the adjacency matrix.

Optionally, before extracting the target motion trajectory data set from the vehicle motion trajectory data set, the method further includes:

Delete the isolated points in the vehicle motion trajectory data set.

A second aspect of the embodiments of the present application provides a taxi trajectory big data-driven passenger route recommendation device, including:

a first acquiring unit, used for acquiring the vehicle motion trajectory data set of the target area;

a first generating unit, configured to generate road network node data according to the vehicle motion trajectory data set;

a second generating unit, configured to generate a route recommendation model according to the road network node data, where the route recommendation model incorporates a heuristic algorithm and an optimization algorithm;

a second acquiring unit, configured to acquire the user's starting point and end point, where the starting point and the end point are located in the target area;

A third generating unit, configured to generate a recommended route from the start point to the end point according to the route recommendation model.

A third aspect of the embodiments of the present application provides a taxi trajectory big data-driven passenger route recommendation device, including:

processors, memories, input and output units, and buses;

the processor is connected to the memory, the input-output unit and the bus;

A program is stored in the memory, and the processor invokes the program to execute the taxi trajectory big data-driven passenger route recommendation method in the first aspect and any possible implementation manner of the first aspect.

A fourth aspect of the embodiments of the present application provides a computer-readable storage medium, where a program is stored in the computer-readable storage medium, and when the program is executed on a computer, the computer causes the computer to execute any one of the first aspect and the first aspect. A taxi trajectory big data-driven passenger route recommendation method in a possible implementation.

As can be seen from the above technical solutions, the embodiments of the present application have the following advantages:

The taxi trajectory big data-driven passenger route recommendation method provided by the embodiment of the present application first obtains the vehicle motion trajectory data set of the target area, then generates road network node data according to the vehicle trajectory data set, and then generates the road network node data according to the road network node data. A route recommendation model that integrates heuristic algorithms and optimization algorithms, and finally generates a recommended route from the starting point to the end point for the user according to the route recommendation model. Since the route recommendation model in the embodiment of the present application integrates heuristic algorithms and optimization algorithms, it is possible to visit fewer nodes in the process of generating the recommended route, compared with the big data-driven taxi trajectory based on the greedy algorithm for passenger-carrying. The route recommendation method reduces the amount of calculation, and can generate recommended routes faster under the same computing power, which is beneficial to improve computing efficiency. At the same time, the travel time of the user is reduced, which is beneficial to improve the user experience.

Description of drawings

1 is a schematic flowchart of an embodiment of a method for recommending passenger routes driven by taxi trajectory big data in an embodiment of the present application;

FIG. 2 is a schematic flowchart of another embodiment of a method for recommending a passenger-carrying route driven by taxi trajectory big data in an embodiment of the present application;

FIG. 3 is a schematic structural diagram of an embodiment of a taxi trajectory big data-driven passenger route recommendation device according to an embodiment of the application;

FIG. 4 is a schematic structural diagram of another embodiment of a taxi trajectory big data-driven passenger route recommendation device according to an embodiment of the application;

FIG. 5 is a schematic structural diagram of another embodiment of a taxi trajectory big data-driven passenger route recommendation device according to an embodiment of the present application.

Detailed ways

The embodiments of the present application provide a method, device, and medium for recommending a passenger-carrying route driven by taxi trajectory big data, which are used to quickly plan the shortest route from the start point to the end point for the vehicle.

The method of the present application can be applied to a server, a terminal, or other devices with logic processing capability, which is not limited in this application. For convenience of description, the following description takes the execution subject as the server as an example.

The embodiments of the present application will be described below with reference to the accompanying drawings.

Referring to FIG. 1, an embodiment of the taxi trajectory big data-driven passenger route recommendation method in the embodiment of the present application includes:

101. The server obtains a vehicle motion trajectory data set of the target area;

In practical applications, the generation of recommended routes is based on the big data of moving trajectories, and relevant information needs to be extracted from a large amount of data. Therefore, the server first needs to obtain the data set of vehicle motion trajectories in the target area. It should be noted that the vehicle motion trajectory data set includes large-scale vehicle motion trajectory data. Each vehicle motion trajectory data consists of multiple points, each point has corresponding attribute information, and each vehicle motion trajectory data is associated with a vehicle. The vehicle corresponds to the vehicle, and a vehicle may include one or more pieces of vehicle motion trajectory data, which is not specifically limited here.

102. The server generates road network node data according to the vehicle motion trajectory data set;

After obtaining the vehicle trajectory data set, the server can process the vehicle trajectory data set to generate road network node data, where the road network node data includes multiple road network nodes and distances between adjacent road network nodes.

103. The server generates a route recommendation model according to the road network node data;

After generating the road network node data, the server can generate a route recommendation model according to the road network node data, and provide a route recommendation service for the vehicle through the route recommendation model. It should be noted that the route recommendation model incorporates a heuristic algorithm and an optimization algorithm, and has the attributes of a heuristic algorithm and an optimization algorithm. That is, as long as the start and end points are given, the route recommendation model can generate the shortest route according to its own road network node data in a relatively short period of time.

104. The server obtains the starting point and the ending point of the user;

To generate a recommended route for a user, the user's starting point and end point need to be obtained first, and the starting point and the end point are located in the target area. Therefore, the server obtains the user's starting point and end point. It should be noted that both the starting point and the ending point are located in the target area, and the starting point and the ending point are expressed in the form of coordinates.

105. The server generates a recommended route from the start point to the end point according to the route recommendation model.

After the server obtains the user's starting point and ending point, it can input the starting point and the ending point into the route recommendation model, and generate a recommended route from the starting point to the ending point through the route recommendation model. After the recommended route is generated, the recommended route is fed back to the user, so that the user can start from the starting point and drive to the end point according to the recommended route, thereby completing the travel plan.

It should be noted that, the method for recommending a passenger carrying route driven by the taxi trajectory big data in this embodiment may be executed on a single server or a distributed server, which is not specifically limited here. When executed on a distributed server, since the task is divided into multiple servers, it can be executed in parallel on multiple servers, so the execution efficiency can be improved.

In this embodiment, the server first obtains the vehicle motion trajectory data set of the target area, then generates road network node data according to the vehicle trajectory data set, and then generates a route recommendation model that combines the heuristic algorithm and the optimization algorithm according to the road network node data, Finally, according to the route recommendation model, a recommended route from the starting point to the end point is generated for the user. Since the route recommendation model in the embodiment of the present application integrates heuristic algorithms and optimization algorithms, it is possible to visit fewer nodes in the process of generating the recommended route, compared with the big data-driven taxi trajectory based on the greedy algorithm for passenger-carrying. The route recommendation method reduces the amount of calculation, and can generate recommended routes faster under the same computing power, which is beneficial to improve computing efficiency. At the same time, the travel time of the user is reduced, which is beneficial to improve the user experience.

Referring to FIG. 2, another embodiment of the taxi trajectory big data-driven passenger route recommendation method in the embodiment of the present application includes:

201. The server obtains a vehicle motion trajectory data set of the target area;

In this implementation, step 201 is similar to step 101 in the foregoing embodiment, and details are not repeated here.

202. The server deletes the isolated points in the vehicle motion trajectory data set;

After the server obtains the vehicle motion trajectory data set, it can obtain any unprocessed vehicle motion trajectory data, and judge whether the points are continuous. If it is found that there are points that are not continuous with other points, it can be determined is an isolated point, that is, garbage data, and then delete processing is performed, and after the execution is completed, the piece of vehicle motion trajectory data is marked as an isolated point and deleted. The server repeatedly executes the acquisition of vehicle motion trajectory data, and determines whether there are isolated points, until all vehicle motion trajectory data are processed.

203. The server extracts the target motion trajectory data set from the vehicle motion trajectory data set;

After deleting the isolated points in the isolated points in the vehicle motion trajectory dataset, the server may perform extraction of the target motion trajectory dataset from the vehicle motion trajectory dataset. Specifically, any piece of vehicle motion trajectory data for data extraction is sequentially obtained, and then the data whose operating status is continuously empty-car-loaded-passenger is extracted and used as the target motion trajectory data, and the operating status includes empty and loaded. It should be noted that after extracting a segment of empty vehicle-passenger-passenger, data extraction is performed again from the next consecutive point, so each extracted segment of empty vehicle-passenger-passenger is to not repeat. After performing data extraction on a piece of vehicle motion track data, mark the vehicle motion track data as having undergone target motion track data extraction, and then execute the next piece of vehicle motion track data that is not marked as having undergone target motion track data extraction. The data extraction operation is performed until all vehicle motion trajectory data are marked as having been subjected to the target motion trajectory data extraction, so that the target motion trajectory data set can be obtained. It should be noted that the vehicle motion trajectory data is composed of multiple operating state segments, and each operating state segment is composed of multiple consecutive points.

204. The server extracts the time and coordinates of each point from the target motion trajectory data set;

After extracting the target motion trajectory data set, the server can extract the time, coordinate and direction information of each point, so as to use this information to generate road network node data.

205. For any target motion trajectory data, the server determines the adjacency relationship between points according to the time of each point;

For any piece of target motion trajectory data, the server may determine the adjacency relationship between the points according to the time of each point in the target motion trajectory data. Among them, for any point, among all the points before the time of the point, the point corresponding to the time closest to the time of the point is adjacent to the point; among all the points after the time of the point, it is adjacent to the point. The point whose time is closest to the point is adjacent to the point.

206. The server calculates the direction change amount according to the adjacency relationship between the points and the coordinates of each point;

After obtaining the adjacency relationship between the points of the target motion trajectory data, the server may calculate the direction change amount according to the adjacency relationship between the points and the coordinates of the points. Specifically, the server first obtains a target point, which is a point with two adjacent points, and then calculates the target point according to the coordinates of the target point and the coordinates of the two adjacent points of the target point. The straight line function with two adjacent points, and then calculate the angle of the two straight line functions, the angle is the direction change amount corresponding to the target point, the direction change amount represents the steering angle of the vehicle at the target point, the direction change amount Corresponding to 3 points, which are the target point and the two adjacent points of the target point. After calculating the direction change amount of a target point, the server marks the target point as the calculated direction change amount, and then calculates the direction change amount of the next target point that is not marked as the calculated direction change amount, until all the target points are The direction changes of , are all calculated.

207. When the amount of directional change is greater than or equal to the second preset angle, the server determines that a point adjacent to the other two points among the corresponding three points is a road network node;

In the process of calculating the direction change amount in the target motion trajectory data, each time a direction change amount is calculated, the server may determine the direction change amount, and if the direction change amount is greater than or equal to the second preset angle, determine Among the three points corresponding to the direction change, the point adjacent to the other two points is the road network node. It should be noted that the second preset angle is generally set to 90°, but may also be set to other degrees, which is not specifically limited here.

It should be noted that the server can not only perform the determination of the road network node in the process of calculating the directional variation in the target motion trajectory data, but also perform the road network node determination after calculating all the directional variations in the target motion trajectory data. The determination of the network node is not specifically limited here.

The server repeatedly executes steps 205 to 207, thereby determining the road network nodes of all target motion trajectory data.

208. The server determines an adjacency relationship between road network nodes;

Due to the huge data volume of the obtained vehicle motion trajectory data set of the target area, in the case of large-scale data, the road network nodes obtained by the server must be directly or indirectly connected. Since the adjacency relationship between road network nodes in each piece of target motion trajectory data is determined, the server can determine the adjacency relationship between all road network nodes after obtaining all road network nodes of the target motion trajectory data.

209. The server calculates the distance between adjacent road network nodes according to the coordinates of each road network node;

After determining the adjacency relationship between all road network nodes, the server can calculate the distance between two adjacent road network nodes according to the coordinates of the road network nodes. It should be noted that the distance calculated here is the Euclidean distance, that is, the distance of a straight line segment between two adjacent road network nodes.

210. The server generates an adjacency matrix and a weighted undirected graph according to the adjacency relationship between road network nodes and the distance between adjacent road network nodes;

The server can generate a weighted undirected graph about all road network nodes according to the adjacency relationship between all road network nodes and the distances between adjacent road network nodes. At the same time, the server may generate an adjacency matrix about all road network nodes according to the adjacency relationship between all road network nodes, and the adjacency matrix is used to represent the adjacency matrix of all road network nodes. A graph is a data structure composed of a vertex set V and an edge set E between vertices. It is represented by a two-tuple G(V, E). The vertex set of a weighted undirected graph is V={V ₁ , V ₂ , V ₃ , V ₄ , }, the edge set is E={e ₁₂ , e ₁₃ , e ₂₁ , e ₂₃ , e ₂₄ , e ₃₁ , e ₃₂ , e ₃₄ , e ₄₂ , e ₄₃ }, the weight can be reached by a vertex The distance to another vertex is represented. In this embodiment, all road network nodes are taken as the vertex set, the straight line segment between two adjacent road network nodes is taken as the edge set, and the distance of the straight line segment between the two adjacent road network nodes is taken as the weight .

211. The server generates road network node data according to the weighted undirected graph and the adjacency matrix;

After generating the weighted undirected graph and the adjacency matrix, the server also generates the road network node data, and the road network node data is the combination of the weighted undirected graph and the adjacency matrix.

212. The server integrates the heuristic of the A ^* algorithm into the Gurobi algorithm to generate the A-Gurobi algorithm;

The server can incorporate the heuristics of the A ^* algorithm into the Gurobi algorithm to generate the A-Gurobi algorithm. Specifically, the heuristic function of the A ^* algorithm is integrated into the Gurobi algorithm, and the heuristic function is shown in formula (1). Since the Gurobi algorithm is an optimization algorithm, the heuristic function of the A ^* algorithm is integrated into the A-Gurobi algorithm generated by the Gurobi algorithm, which not only has the characteristics of the optimization algorithm, but also has the characteristics of the heuristic algorithm.

f(n)=g(n)+h(n) (Equation 1)

Among them, f(n) represents the cost of getting to the target point, g(n) represents the cost from the initial node to any node n, and h(n) represents the heuristic evaluation cost from node n to the target point.

213. The server generates a route recommendation model according to the road network node data and the A-Gurobi algorithm;

After generating the road network node data and the A-Gurobi algorithm, the server can generate a route recommendation model according to the road network node data and the A-Gurobi algorithm, and provide users with route recommendation services through the route recommendation model. It should be noted that the route recommendation model incorporates a heuristic algorithm and an optimization algorithm, and has the attributes of a heuristic algorithm and an optimization algorithm. That is, as long as the start and end points are given, the route recommendation model can generate the shortest route according to its own road network node data in a relatively short period of time.

214. The server obtains the starting point and the ending point of the user;

In this embodiment, step 214 is similar to step 104 in the foregoing embodiment, and details are not repeated here.

215. The server connects the starting point with each road network node and the end point in the road network node data of the route recommendation model respectively, and obtains a plurality of line segments, wherein each road network node corresponds to a line segment, and the line segment formed by the start point and the end point is the target line segment;

After acquiring the user's starting point and ending point, the server can connect the starting point with the ending point and each road network node in the road network node data of the route recommendation model with a straight line segment to obtain multiple line segments. Each line segment corresponds to a road network node or end point, and each road network node or end point also corresponds to a line segment. Among all the line segments, the server may determine the line segment corresponding to the end point as the target line segment, and the target line segment represents the displacement from the start point to the end point.

216. The server calculates the degrees of the included angles formed by the target line segment and other line segments respectively;

Since the endpoints of each line segment include the starting point, all the line segments are directly connected, so an included angle can be formed between these line segments, and the server can separately calculate the included angle between the target line segment and every other line segment. In order to facilitate the calculation, the server can use the starting point as the coordinate origin, the east-west direction as the X-axis, and the north-south direction as the Y-axis, establish a rectangular coordinate system, and calculate the angle between the target line segment and each other line segment separately.

217. The server deletes the road network node corresponding to the line segment whose angle formed by the target line segment is greater than or equal to the first preset angle, and obtains the recommended road network node;

In practical applications, when the included angle between the line segment corresponding to a road network node and the target line segment exceeds a certain degree, it usually does not pass through the road network node. Therefore, the server can delete the road network node corresponding to the line segment whose angle formed by the target line segment is greater than or equal to the first preset angle, and retain the remaining road network nodes, which are the recommended road network nodes. network node.

In this embodiment, the server deletes the road network nodes other than the recommended road network nodes, which can reduce the calculation amount of the route recommendation model, which is beneficial to improve the operation efficiency of generating the recommended route, and at the same time, it can save resource overhead and reduce unnecessary resources. waste.

218. The server generates a recommended route according to the recommended road network node, the starting point and the ending point.

After obtaining the recommended road network node, the server can use the recommended road network node, start point and end point as the parameters of the route recommendation model, and use the A-Gurobi algorithm in the route recommendation model to generate the recommended route.

Referring to FIG. 3, an embodiment of the taxi trajectory big data-driven passenger route recommendation device in the embodiment of the present application includes:

The first obtaining unit 301 is used to obtain the vehicle motion trajectory data set of the target area;

a first generating unit 302, configured to generate road network node data according to the vehicle motion trajectory data set;

The second generating unit 303 is configured to generate a route recommendation model according to the road network node data, and the route recommendation model incorporates a heuristic algorithm and an optimization algorithm;

The second obtaining unit 304 is used to obtain the starting point and the ending point of the user, and the starting point and the ending point are located in the target area;

The third generating unit 305 is configured to generate a recommended route from the start point to the end point according to the route recommendation model.

In this embodiment, the first obtaining unit 301 first obtains the vehicle motion trajectory data set of the target area, then the first generating unit 302 generates road network node data according to the vehicle trajectory data set, and then the second generating unit 303 generates road network node data according to the road network node data A route recommendation model incorporating the heuristic algorithm and the optimization algorithm is generated, and finally the third generating unit 305 generates a recommended route from the start point to the end point for the user according to the route recommendation model. Since the route recommendation model in this embodiment of the present application integrates heuristic algorithms and optimization algorithms, the server can visit fewer nodes in the process of generating the recommended route, compared with the big data-driven taxi trajectory based on the greedy algorithm for passenger-carrying. The route recommendation method reduces the amount of calculation, and can generate recommended routes faster under the same computing power, which is beneficial to improve operational efficiency. At the same time, the travel time of the user is reduced, which is beneficial to improve the user experience.

Referring to FIG. 4, another embodiment of the taxi trajectory big data-driven passenger route recommendation device in the embodiment of the present application includes:

The first obtaining unit 401 is used to obtain the vehicle motion trajectory data set of the target area;

a first generating unit 402, configured to generate road network node data according to the vehicle motion trajectory data set;

The second generating unit 403 is configured to generate a route recommendation model according to the road network node data, and the route recommendation model incorporates a heuristic algorithm and an optimization algorithm;

The second obtaining unit 404 is used to obtain the starting point and the ending point of the user, and the starting point and the ending point are located in the target area;

The third generating unit 405 is configured to generate a recommended route from the start point to the end point according to the route recommendation model.

In this embodiment, the second generating unit 403 is specifically configured to:

The heuristic of the A ^* algorithm is integrated into the Gurobi algorithm to generate the A-Gurobi algorithm;

The route recommendation model is generated according to the road network node data and the A-Gurobi algorithm, wherein the road network node data is used as the parameter of the route recommendation model.

The third generating unit 405 is specifically used for:

Connect the starting point with each road network node and the end point in the road network node data of the route recommendation model respectively to obtain a plurality of line segments, wherein each road network node corresponds to a line segment, and the line segment formed by the starting point and the end point is used as the target line segment;

Calculate the degrees of the included angles formed by the target line segment and other line segments;

Delete the road network node corresponding to the line segment whose angle formed by the target line segment is greater than or equal to the first preset angle to obtain the recommended road network node;

Generate recommended routes based on recommended road network nodes, starting points and ending points.

In this embodiment, the first generating unit 402 may include an extracting module 4021 , a generating module 4022 and a deleting module 4023 .

The extraction module 4021 is used for extracting a target motion trajectory data set from the vehicle motion trajectory data set. The operation status of the target motion trajectory data set is continuously empty vehicle-passenger-passenger, and the operation status includes empty vehicle and passenger.

The generating module 4022 may include an extracting sub-module 40221, a determining sub-module 40222 and a generating sub-module 40223.

The extraction sub-module 40221 is used to extract the time, coordinates and direction of each point from the target motion trajectory data set.

Determine sub-module 40222 for:

For any target motion trajectory data, determine the adjacency relationship between points according to the time of each point;

Calculate the direction change amount according to the adjacency relationship between points and the coordinates of each point, and the direction change amount is the angle of two line segments formed by a point and two adjacent points respectively;

When the amount of change in the appearance direction is greater than or equal to the second preset angle, a point adjacent to the other two points among the corresponding three points is determined as a road network node.

Generate submodule 40223 for:

Determine the adjacency relationship between road network nodes;

Calculate the distance between adjacent road network nodes according to the coordinates of each road network node;

Generate an adjacency matrix and a weighted undirected graph according to the adjacency relationship between road network nodes and the distance between adjacent road network nodes;

Generate road network node data according to weighted undirected graph and adjacency matrix.

In this implementation, the functions of each unit and module correspond to the steps in the foregoing embodiment shown in FIG. 2 , which will not be repeated here.

Referring to FIG. 5, another embodiment of the taxi trajectory big data-driven passenger route recommendation device in the embodiment of the present application includes:

processor 501, memory 502, input and output unit 503 and bus 504;

The processor 501 is connected to the memory 502, the input and output unit 503 and the bus 504;

The memory 502 stores a program, and the processor 501 calls the program to execute the steps in the embodiments shown in FIGS. 1 to 2 .

In this embodiment, the functions of the processor 501 correspond to the steps in the foregoing embodiments shown in FIG. 1 to FIG. 2 , which will not be repeated here.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described systems, devices and units can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, for example, multiple units or components may be combined Either it can be integrated into another system, or some features can be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may also be electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware, or can be implemented in the form of software functional units.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or can be understood as a part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, removable hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), magnetic disk or optical disk and other media that can store program codes.

Claims (11)

1. A taxi track big data driven passenger carrying route recommendation method is characterized by comprising the following steps:

acquiring a vehicle motion track data set of a target area;

generating road network node data according to the vehicle motion track data set;

generating a route recommendation model according to the road network node data, wherein the route recommendation model is fused with an heuristic algorithm and an optimization algorithm;

acquiring a starting point and an end point of a user, wherein the starting point and the end point are positioned in the target area;

and generating a recommended route from the starting point to the end point according to the route recommendation model.

2. The taxi track big data driven passenger carrying route recommendation method according to claim 1, wherein the generating a route recommendation model according to the road network node data comprises:

a is to be ^* Fusing a heuristic algorithm into a Gurobi algorithm to generate an A-Gurobi algorithm;

and generating a route recommendation model according to the road network node data and the A-Gurobi algorithm, wherein the road network node data is used as a parameter of the route recommendation model.

3. The taxi track big data driven passenger carrying route recommendation method according to claim 1, wherein the generating of the recommended route from the starting point to the end point according to the route recommendation model comprises:

respectively connecting the starting point with each road network node in the road network node data of the route recommendation model and the end point to obtain a plurality of line segments, wherein each road network node corresponds to one line segment, and the line segment formed by the starting point and the end point is a target line segment;

calculating the degree of an included angle formed by the target line segment and other line segments;

deleting the road network nodes corresponding to the line segments of which the included angles formed with the target line segments are more than or equal to a first preset angle to obtain recommended road network nodes;

and generating a recommended route according to the recommended road network node, the starting point and the end point.

4. The taxi track big data driven passenger carrying route recommendation method according to any one of claims 1 to 3, wherein the generating road network node data according to the vehicle motion track data set comprises:

extracting a target motion trail data set from the vehicle motion trail data set, wherein the operation state of the target motion trail data set is empty vehicle-passenger continuously, and the operation state comprises empty vehicle and passenger;

and generating road network node data according to the target motion trail data set.

5. The taxi track big data driven passenger carrying route recommendation method according to claim 4, wherein the generating road network node data according to the target motion track data set comprises:

extracting the time and the coordinates of each point from the target motion trajectory data set;

determining road network nodes according to the time and the coordinates of each point;

and generating road network node data according to road network nodes, wherein the road network node data comprises a weighted undirected graph and an adjacency matrix.

6. The taxi track big data driven passenger carrying route recommendation method according to claim 5, wherein the determining road network nodes according to the time and the coordinates of each point comprises the following steps:

for any target motion trajectory data, determining the adjacency relation between points according to the time of each point;

calculating direction variation according to the adjacency relation between the points and the coordinates of each point, wherein the direction variation is the angle of two line segments formed by one point and the two adjacent points respectively;

and when the direction variation is larger than or equal to a second preset angle, determining the point adjacent to the other two points in the three corresponding points as the road network node.

7. The taxi track big data driven passenger carrying route recommendation method according to claim 5, wherein the generating of road network node data according to road network nodes comprises:

determining the adjacency relation between the road network nodes;

calculating the distance between adjacent road network nodes according to the coordinates of each road network node;

generating an adjacency matrix and a weighted undirected graph according to the adjacency relation between the road network nodes and the distance between the adjacent road network nodes;

and generating road network node data according to the weighted undirected graph and the adjacency matrix.

8. The taxi track big data driven passenger carrying route recommendation method according to claim 4, wherein before extracting the target motion track data set from the vehicle motion track data set, the method further comprises:

and deleting the isolated points in the vehicle motion trail data set.

9. A taxi track big data driven passenger carrying route recommendation device is characterized by comprising:

the first acquisition unit is used for acquiring a vehicle motion trail data set of a target area;

the first generation unit is used for generating road network node data according to the vehicle motion trail data set;

a second generation unit, configured to generate a route recommendation model according to the road network node data, where the route recommendation model is fused with an heuristic algorithm and an optimization algorithm;

a second obtaining unit, configured to obtain a start point and an end point of a user, where the start point and the end point are located in the target area;

a third generating unit, configured to generate a recommended route from the starting point to the ending point according to the route recommendation model.

10. A taxi track big data driven passenger carrying route recommendation device is characterized by comprising:

the device comprises a processor, a memory, an input and output unit and a bus;

the processor is connected with the memory, the input and output unit and the bus;

the memory stores a program, and the processor calls the program to execute the taxi track big data driven passenger carrying route recommendation method according to any one of claims 1 to 8.

11. A computer-readable storage medium having stored therein a program which, when executed on a computer, causes the computer to execute the taxi track big data driven passenger route recommendation method according to any one of claims 1 to 8.

Images