Device for preferentially searching car-pooling friends in car-pooling process
By prioritizing the search for Pinyou Packaging, driver matching and route planning modules of Pinyou devices, the existing Pinyou logical conditions are solved, the carpooling order response rate and driver-passenger experience are improved, and traffic congestion and carbon emissions are reduced.
Patent Information
- Application Number
- CN202410922513.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-10
- Publication Date
- 2024-11-22
AI Technical Summary
The existing carpooling logic conditions are harsh, resulting in a large number of orders for only types of carpooling and a small number of drivers, resulting in a low response rate for carpooling and affecting the driver and passenger experience.
Design a carpooling device that prioritizes search for carpooling, including carpooling package module, driver matching module and route planning module. By calculating the order-sharing index, the carpooling order will be packaged and processed, and the play order will be performed after the delayed play time will be delayed. Make an appointment for a ride-sharing order, match the driver, and optimize the route.
It improves the response rate of ride-sharing orders, improves the driver and passenger experience, reduces the vehicle's empty time and traffic congestion, and reduces the travel cost and carbon emissions of each passenger.
Smart Images

Figure CN119006089A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of traffic services, and more particularly to a device for preferentially searching for carpooling partners. Background Art
[0002] Ride-sharing services are becoming increasingly popular in the ride-hailing industry. Ride-sharing services usually match passengers with the same destination through an application, allowing them to ride in the same car together and share the fare. By combining multiple passengers in one car, ride-sharing services help reduce traffic congestion on the road, improve traffic efficiency, and thus reduce emissions.
[0003] The existing carpooling logic conditions are demanding. If a passenger places an order for carpooling only, then only drivers who are on the carpooling trip can be found. Since there are many orders for carpooling only and the number of drivers on the carpooling trip is small, most carpooling orders cannot find drivers, which affects the response rate of carpooling orders. In order to improve the current situation and enhance the driver and passenger experience, the carpooling strategy is adjusted.
[0004] In view of the above problems, the present invention proposes a solution. Summary of the invention
[0005] In order to overcome the above defects of the prior art, the present invention provides a device for preferentially searching for carpooling partners to solve the problems existing in the above background technology.
[0006] To achieve the above object, the present invention provides the following technical solutions:
[0007] A device for finding carpooling partners in priority includes a carpooling partner packaging module, a driver matching module, and a route planning module. The modules are connected by signals. The steps of data processing between the modules are as follows:
[0008] The carpooling packing module is used to obtain the data of carpooling order a, count the number of carpooling orders in the area, and calculate the delayed order time according to the number of carpooling orders in the area;
[0009] The carpooling orders in the area are screened to obtain the pre-carpooling orders, and the carpooling index is calculated based on the data of the pre-carpooling orders and the carpooling order a;
[0010] The carpooling orders are packaged according to the order-matching index, and the orders are played after the delayed play time. In addition, a new carpooling reservation order is added while the carpooling orders are packaged;
[0011] The driver matching module is used to match drivers for the packaged ride-sharing orders and process the orders for drivers who have not turned on the ride-sharing switch;
[0012] The route planning module is used to plan and optimize routes based on the passenger's order information and the driver's location.
[0013] Preferably, calculating the delayed order time according to the number of carpooling orders in the area includes:
[0014] The operation area is divided into regular hexagonal grids of the same size. The number of carpooling orders in the set area is collected with the point where the carpooling order a is issued as the center. The delayed order time of carpooling order a is calculated based on the number of carpooling orders in the set area. The calculation formula is: DT is the delayed play time, NUM P It is expressed as the number of carpooling orders in the set area, and h and b are set constants.
[0015] Preferably, screening the carpooling orders in the area to obtain the pre-carpooling orders includes:
[0016] Filter out the carpooling orders in the area whose starting point is on the way of carpooling order a and whose travel radius does not exceed the set kilometers, and record them as pre-carpooling orders.
[0017] Preferably, calculating the order-sharing index based on the data of the pre-carpooling order and the carpooling order a includes:
[0018] Collect the starting and ending positions of the carpooling order a and the pre-carpooling order, and calculate the angle coefficient between the carpooling orders;
[0019] Collect the travel paths of carpooling order a and pre-carpooling order, and calculate the path similarity coefficient between the carpooling orders;
[0020] The carpooling index is calculated based on the angle coefficient and path similarity coefficient between carpooling orders. The calculation formula is OT=a1×AG+a2×PS, where OT represents the carpooling index, AG represents the angle coefficient, PS represents the path similarity coefficient, and a1 and a2 represent the weight coefficients of the angle coefficient and the path similarity coefficient.
[0021] Preferably, calculating the angle coefficient between the carpooling orders includes:
[0022] Using the longitude and latitude representation, convert the starting and ending positions of the carpooling order a and the pre-carpooling order into longitude and latitude coordinates to obtain the starting and ending coordinates of the carpooling order;
[0023] For each carpooling order, the direction vector from the starting coordinate to the end coordinate is calculated, which represents the direction of travel of the carpooling order;
[0024] Use the dot product method of vectors to calculate the angle between the carpooling order a and the direction vector of the pre-carpooling order;
[0025] The calculated angle is converted into degrees, and the angle coefficient is calculated based on the angle degree. The calculation formula is AG=30-θ, where AG represents the angle coefficient, 30 is the preset angle, and θ represents the angle between the direction vectors of the two carpooling orders.
[0026] Preferably, using the dot product method of vectors, calculating the angle between the carpooling order a and the pre-carpooling order direction vector includes:
[0027] The direction vector of the carpooling order is calculated by subtracting the starting coordinate from the ending coordinate of the carpooling order a and the pre-carpooling order;
[0028] The dot product of the vector is calculated based on the direction vector of the carpooling order a and the pre-carpooling order. The calculation formula is: where v ax With v bx represents the x-component of two vectors, v ay With v by represents the y components of two vectors;
[0029] Calculate the length of the vector of carpooling order a and pre-carpooling order respectively, and the calculation formula is: where v x With v y Represents the x component and y component of a vector, where the x component of the vector refers to the projection length of the vector on the x-axis;
[0030] According to the direction vector of the carpooling order a and the pre-carpooling order, the dot product of the vector and the length of the vector are calculated to obtain the angle between the direction vectors of the two carpooling orders. The calculation formula is:
[0031]
[0032] Preferably, calculating the path similarity coefficient between carpooling orders includes:
[0033] Divide the path of pre-carpooling orders and carpooling orders into a series of points with fixed intervals;
[0034] Use the shortest line segment connection method to connect the points on the two paths one by one to obtain a series of line segments;
[0035] In the line segments obtained by connecting the paths, the length of the line segments is calculated using the Euclidean distance method, and a distance matrix is constructed based on the length of the line segments. The length of the line segments is searched starting from the upper left corner of the distance matrix, and the maximum value is selected as the maximum continuous line segment length between the paths;
[0036] Among the line segments obtained by connecting the paths, calculate the path length that can traverse both paths at any time, and select the maximum value as the maximum possible value between the paths;
[0037] The path similarity coefficient is calculated based on the maximum continuous line segment length between the paths and the maximum possible value of the path length. The calculation formula is: Where PS represents the path similarity coefficient, MC represents the maximum continuous line segment length, and MP represents the maximum possible value of the path length.
[0038] Preferably, packaging the carpooling orders according to the carpooling index includes:
[0039] The number of passengers in each pre-carpooling order is collected, and the pre-carpooling order and carpooling order a are sorted from high to low in terms of their order-splitting index. The pre-carpooling order with the highest order-splitting index is selected and combined with carpooling order a, and the total number of passengers does not exceed the number of passengers that the online car-hailing vehicle can carry.
[0040] Technical effects and advantages of the present invention:
[0041] The data of carpooling order a is obtained, the number of carpooling orders in the area is counted, and the delayed order broadcasting time is calculated according to the number of carpooling orders in the area, and the order broadcasting index is calculated. The carpooling orders are packaged and processed according to the order broadcasting index, and the orders are broadcast after the delayed order broadcasting time. At the same time, new carpooling reservations are added, and drivers are matched for the packaged carpooling orders. For drivers who have not turned on the carpooling switch, the orders are processed, and route planning and optimization are performed, which effectively improves the response rate of carpooling orders and enhances the driver and passenger experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 It is the overall flow chart of the present invention. DETAILED DESCRIPTION
[0043] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are merely illustrative. The device for preferentially finding carpooling partners involved in the present invention is not limited to the various structures recorded in the following embodiments. All other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.
[0044] The present invention provides a device for finding carpooling partners in priority, including a carpooling partner packaging module, a driver matching module and a route planning module. The modules are connected by signals, and the data processing steps between the modules are as follows:
[0045] The carpooling packing module is used to obtain the data of the carpooling order a issued, obtain the delayed order time according to the number of carpooling orders in the area, and trigger the scheduling after the delayed order time, which can effectively improve the order packing rate. The data of the carpooling order a includes the starting position, the end position and the travel path of the order;
[0046] Find other real-time or scheduled carpooling orders during the delay time, record them as other carpooling orders, obtain data of other carpooling orders, calculate the carpooling order index according to the data of other carpooling orders and carpooling order a, package the carpooling orders according to the carpooling order index, play the packaged orders, and add a scheduled carpooling order while packaging the carpooling orders, wherein the data of other carpooling orders include the starting position, the end position and the travel path of the order;
[0047] If no orders that can be packaged and processed are matched within the delay time, the issued order a will be scheduled first, and real-time or scheduled carpooling matching will continue during the journey;
[0048] Packaged ridesharing orders allow vehicles to pick up multiple passengers in one trip, reducing idle time and increasing vehicle utilization. This can effectively reduce costs and improve efficiency; packaged ridesharing orders can spread the cost of the trip among multiple passengers, thereby reducing the travel cost of each passenger. This is attractive to price-sensitive passengers; reducing the number of times vehicles are on the road, thereby reducing traffic congestion. Multiple passengers sharing a car means fewer vehicles on the road, reducing traffic pressure; packaged ridesharing orders can reduce vehicle mileage, save energy, and reduce carbon emissions. This is in line with environmental awareness and helps reduce the negative impact of traffic on the environment.
[0049] In this embodiment, it should be specifically explained that the step of obtaining the delayed order time according to the number of carpooling orders in the area is to divide the operating area into regular hexagonal grids of the same size with a radius of 15 kilometers, take the issuing point of carpooling order a as the center, collect the number of carpooling orders in the set area, and calculate the delayed order time of carpooling order a according to the number of carpooling orders in the set area. The calculation formula is: DT is the delayed play time, NUM P It is represented by the number of carpooling orders in the set area. h and b are set constants, and their specific values are set by the operator. As the number of carpooling orders in the current area increases, the delayed broadcast time will also increase accordingly. This is to maximize the success rate and efficiency of carpooling.
[0050] When the number of carpooling orders increases, the vehicle resources in the area may become tight. In order to ensure that each carpooling order can find a suitable vehicle to take, the system may delay the broadcast of the order to wait for more idle vehicles to appear. When the number of carpooling orders increases, more route options and matching combinations may appear, which may require more computing time to determine the best carpooling route, thereby increasing the delay in broadcasting. A large number of carpooling orders will increase the load on the carpooling system and may cause the system to respond more slowly. In order to avoid system crashes or overloads, the system may adopt a strategy of delaying the broadcast of orders to relieve system pressure.
[0051] In this embodiment, it should be specifically explained that the steps of obtaining data of other carpooling orders are:
[0052] Online ride-hailing platforms need to classify and label existing order data to distinguish between ride-sharing orders and non-ride-sharing orders;
[0053] In the marked order data, the online ride-hailing platform needs to filter out only the orders that contain carpooling. This can be achieved by writing queries or filtering conditions to keep carpooling orders and exclude non-carpooling orders.
[0054] After obtaining the ride-hailing order data, the online ride-hailing platform may need to perform data cleaning and preprocessing to remove duplicate data, abnormal data, or missing data to ensure the accuracy and completeness of the data.
[0055] In this embodiment, it should be specifically explained that the step of calculating the order-sharing index based on the data of other carpooling orders and carpooling order a is as follows:
[0056] Filter other carpooling orders to obtain a pre-carpooling order, and the starting point of the pre-carpooling order must be on the way of carpooling order a, and the travel radius cannot exceed n kilometers, and the value of n can be controlled by the operator;
[0057] The angle coefficient between the carpooling order a and the starting position and the end position of the pre-carpooling order is calculated;
[0058] The path similarity coefficient between the carpooling orders is calculated based on the travel paths of the carpooling order a and the pre-carpooling order;
[0059] The carpooling index is calculated based on the angle coefficient and the path similarity coefficient between the carpooling orders, and the calculation formula is OT=a1×AG+a2×PS, wherein OT represents the carpooling index, AG represents the angle coefficient, PS represents the path similarity coefficient, a1 and a2 represent the weight coefficients of the angle coefficient and the path similarity coefficient, and this embodiment does not perform specific calculations on the specific values of a1 and a2.
[0060] In this embodiment, it should be specifically explained that the step of calculating the angle coefficient between the carpooling orders is:
[0061] Get the starting and ending positions of the carpooling order a and the pre-carpooling order, and use longitude and latitude to express them to get the starting and ending coordinates;
[0062] For each carpooling order, the direction vector from the starting position to the end position is calculated, that is, the end position coordinates minus the starting position coordinates. The resulting direction vector is a vector that represents the direction of travel of the carpooling order.
[0063] Use the dot product method of vectors to calculate the angle between the direction vectors of the two ridesharing orders. The dot product method of vectors is an operation used to calculate the product between two vectors. It multiplies the corresponding components of the two vectors and adds the results to obtain a scalar value (i.e., quantity). The result of the dot product is a scalar value that represents the degree of similarity between the two vectors. When the two vectors have the same direction, the result of the dot product is positive; when their directions are opposite, the result of the dot product is negative; when their directions are perpendicular, the result of the dot product is zero;
[0064] The calculated angle is converted into degrees, and the angle coefficient is calculated based on the angle degree. The calculation formula is AG=30-θ, where AG represents the angle coefficient, 30 is the preset angle, and θ represents the angle between the direction vectors of the two carpooling orders.
[0065] In this embodiment, it should be specifically explained that the steps of using the dot product method of vectors to calculate the angle between the direction vectors of two carpooling orders are:
[0066] Calculate the direction vectors of the two carpooling orders by subtracting the starting coordinates from the ending coordinates. For example, the starting coordinates of carpooling order a and pre-carpooling order are (x1, y1) and (x3, y3) respectively, and the ending coordinates are (x2, y2) and (x4, y4) respectively. Then the direction vector of carpooling order a is The direction vector of the pre-carpooling order is
[0067] The dot product of the vector is calculated based on the direction vector of the carpooling order a and the pre-carpooling order. The calculation formula is: where v ax With v bx represents the x-component of two vectors, v ay With v by represents the y components of two vectors;
[0068] Calculate the length of the vector of the carpooling order a and the pre-carpooling order, that is, the modulus of the vector, and the calculation formula is: where v x With vy Represents the x-component and y-component of a vector, where the x-component of the vector refers to the projection length of the vector on the x-axis. In a two-dimensional Cartesian coordinate system, a vector can be represented as an ordered pair (x, y), where x represents the component of the vector on the x-axis and y represents the component of the vector on the y-axis.
[0069] According to the direction vector of the carpooling order a and the pre-carpooling order, the dot product of the vector and the length of the vector are calculated to obtain the angle between the direction vectors of the two carpooling orders. The calculation formula is:
[0070]
[0071] In this embodiment, it should be specifically explained that the step of calculating the path similarity coefficient between the carpooling orders is to divide the path of the pre-carpooling order and the carpooling order a into a series of points with fixed intervals, represent the path of each order as a sequence of geographic coordinate points, and use the Frechet distance similarity measurement method to calculate the path similarity coefficient between the carpooling orders. The Frechet distance similarity measurement method is a distance measurement method for measuring the similarity between two paths. It takes into account the continuity and sequentiality of the paths. Its calculation method is to find a movement method in the two paths so that a point can be matched with a point on the other path at any time, and the distance between the paths does not exceed a given threshold.
[0072] In this embodiment, it should be specifically explained that the steps of calculating the path similarity coefficient between carpooling orders using the Frechet distance similarity measurement method are as follows:
[0073] Use the shortest segment connection method to connect the points on the two paths one by one to obtain a series of line segments, which describe the continuous movement process on the path. For example, the path of carpooling order a is (1,1), (2,2), (3,3), and the path of a pre-carpooling order is (1,1), (2,2), (3,4);
[0074] In the continuous connection between the paths, the maximum continuous line segment length of the two paths is calculated by constructing a distance matrix. For example, the distance matrix is constructed for the carpooling order path a and one of the pre-carpooling order paths as follows:
[0075] As shown in Table 1
[0076]
[0077] Table 1
[0078] Start from the upper left corner to find continuous line segments and keep a record of the maximum value. In this example, the maximum continuous line segment length is 1.
[0079] In the continuous connection between paths, the maximum possible value of the path length that can traverse two paths at any time is calculated. For example, for the carpooling order path a, the distance between any two points is calculated as
[0080] For one of the pre-carpooling order paths, the distance between any two points is calculated as
[0081] For a carpooling order, the maximum path distance is For one of the pre-rideshare order paths, the maximum distance is So the maximum possible value of the path length is
[0082] The maximum continuous line segment length between the paths is compared with the maximum possible value of the path length to determine the path similarity coefficient, which is calculated as Wherein PS represents the path similarity coefficient, MC represents the maximum continuous line segment length, and MP represents the maximum possible value of the path length. The path similarity coefficient represents the similarity between two paths, usually between 0 and 1, and is used to measure the similarity between two paths. The closer the score is to 1, the more similar the two paths are, and the closer the score is to 0, the less similar the two paths are.
[0083] In this embodiment, it should be specifically explained that the step of packaging and processing the carpooling orders according to the carpooling index is to collect the number of passengers in each pre-carpooling order, sort the carpooling indexes of the pre-carpooling orders and carpooling order a from high to low, and select the pre-carpooling order with the highest carpooling index and carpooling order a for combination, and the total number of passengers does not exceed the number of passengers that the online car-hailing vehicle can carry.
[0084] By bundling and assigning multiple carpooling orders to the same vehicle, on the one hand, you can directly find idle drivers; on the other hand, from the driver's perspective, carpooling for the entire trip is more attractive than a single carpooling order, reducing the vehicle's idle mileage and idle time, and improving vehicle utilization, thereby improving the efficiency of the entire carpooling system, minimizing the vehicle's driving distance and driving time, thereby reducing fuel costs and driver costs, and saving operating costs. Bundling and assigning orders can reduce passengers' waiting time and driving time, improve passengers' travel experience, and increase passenger satisfaction.
[0085] The newly added reserved carpooling order is a reserved carpooling order. If carpooling order A is a real-time order (departure immediately), the entire trip may be composed of multiple carpooling orders (A\B\C\D). The trip lasts for 5 hours. If carpooling order E is reserved for use in 4 hours and the itinerary route is the same, it can also be assigned to the driver in advance; if carpooling orders A\B\C\D are all reserved orders and the itinerary route is the same, they can also be assigned to the driver in advance.
[0086] The driver matching module is used to match drivers for the packaged ride-sharing orders and process the orders for drivers who have not turned on the ride-sharing switch;
[0087] In this embodiment, it should be specifically explained that the driver matching steps for the packaged ride-sharing orders are:
[0088] Collect information about all carpool-only orders that need to be packaged, including the order's starting location, end location, number of passengers, and departure time;
[0089] Collecting idle driver information within the set area of the order starting location, the driver information includes the driver's location, idle time, working status, etc., and comprehensively evaluating the order availability index;
[0090] The driver with the highest order acceptance index is selected to match the packaged carpooling orders.
[0091] The above-mentioned broadcast order processing for drivers who have not turned on the carpooling switch can reveal the entire trip information to drivers who have not turned on the carpooling switch, thereby increasing the order volume. These drivers can see the information of the entire trip and understand whether there are multiple orders that can be packaged, thereby increasing the order volume; after turning on the carpooling switch, drivers can accept more orders and increase their income by packaging orders. The packaging of multiple orders can reduce empty mileage and idle time, and increase income levels; by packaging orders, the utilization rate of vehicle resources can be optimized. Drivers who have not turned on the carpooling switch can improve the utilization rate of vehicles by accepting multiple orders, reduce idle time and empty mileage, and optimize resource allocation; packaging orders can reduce passengers' waiting time and driving time, and improve service quality. The packaging of multiple orders can more effectively meet passengers' needs and improve passenger satisfaction.
[0092] The route planning module is used to plan and optimize routes based on the passenger's order information and the driver's location to ensure the most efficient carpooling route;
[0093] In this embodiment, it should be specifically explained that the route planning steps according to the passenger's order information and the driver's location are:
[0094] Collect all packaged carpooling order information and the driver's current location, wherein all packaged carpooling order information includes the starting point, end point, and departure time of each order;
[0095] The driver’s current location information is used as the starting point for route planning, and the shortest path algorithm is used to determine the optimal travel sequence;
[0096] For the determined travel sequence, a route planning algorithm is used to plan the optimal driving route for the driver, and the optimal travel route is planned by taking into account factors such as traffic conditions and road restrictions. The route planning algorithm is a prior art for planning the optimal driving route for the driver, and this embodiment will not further describe this technology in detail.
[0097] During the trip, real-time traffic conditions are continuously monitored and route planning is updated in real time as needed to respond to emergencies or changes in road conditions, ensuring that the carpooling route remains efficient at all times.
[0098] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
[0099] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A device for finding carpooling partners first, characterized in that: It includes a ride-sharing module, a driver matching module, and a route planning module. The modules are connected through signals. The steps of data processing between the modules are as follows: The carpooling packing module is used to obtain the data of carpooling order a, count the number of carpooling orders in the area, and calculate the delayed order time according to the number of carpooling orders in the area; The carpooling orders in the area are screened to obtain the pre-carpooling orders, and the carpooling index is calculated based on the data of the pre-carpooling orders and the carpooling order a; The carpooling orders are packaged according to the order-matching index, and the orders are played after the delayed play time. In addition, a new carpooling reservation order is added while the carpooling orders are packaged; The driver matching module is used to match drivers for the packaged ride-sharing orders and process the orders for drivers who have not turned on the ride-sharing switch; The route planning module is used to plan and optimize routes based on the passenger's order information and the driver's location.
2. A carpooling priority searching device according to claim 1, characterized in that: The delayed order time calculated based on the number of carpooling orders in the area includes: The operation area is divided into regular hexagonal grids of the same size. With the point where the carpooling order a is issued as the center, the number of carpooling orders in the set area is collected, and the delayed order time of carpooling order a is calculated based on the number of carpooling orders in the set area. The calculation formula is DT = h × e b×NUMP , where DT is the delayed play time, NUM P It is expressed as the number of carpooling orders in the set area, and h and b are set constants.
3. The device for finding carpooling partners in priority according to claim 1, characterized in that: The pre-carpooling orders obtained by filtering the carpooling orders in the area include: Filter out the carpooling orders in the area whose starting point is on the way of carpooling order a and whose travel radius does not exceed the set kilometers, and record them as pre-carpooling orders.
4. The device for finding carpooling partners in priority according to claim 1, characterized in that: The calculation of the order-sharing index based on the data of pre-carpooling orders and carpooling orders a includes: Collect the starting and ending positions of the carpooling order a and the pre-carpooling order, and calculate the angle coefficient between the carpooling orders; Collect the travel paths of carpooling order a and pre-carpooling order, and calculate the path similarity coefficient between the carpooling orders; The carpooling index is calculated based on the angle coefficient and path similarity coefficient between carpooling orders. The calculation formula is OT=a1×AG+a2×PS, where OT represents the carpooling index, AG represents the angle coefficient, PS represents the path similarity coefficient, and a1 and a2 represent the weight coefficients of the angle coefficient and the path similarity coefficient.
5. The device for finding carpooling partners in priority according to claim 4, characterized in that: Calculating the angle coefficient between carpooling orders includes: Using the longitude and latitude representation, convert the starting and ending positions of the carpooling order a and the pre-carpooling order into longitude and latitude coordinates to obtain the starting and ending coordinates of the carpooling order; For each carpooling order, the direction vector from the starting coordinate to the end coordinate is calculated, which represents the direction of travel of the carpooling order; Use the dot product method of vectors to calculate the angle between the carpooling order a and the direction vector of the pre-carpooling order; The calculated angle is converted into degrees, and the angle coefficient is calculated based on the angle degree. The calculation formula is AG=30-θ, where AG represents the angle coefficient, 30 is the preset angle, and θ represents the angle between the direction vectors of the two carpooling orders.
6. A carpooling priority searching device according to claim 5, characterized in that: Using the dot product method of vectors, the angle between the carpooling order a and the pre-carpooling order direction vector is calculated as follows: The direction vector of the carpooling order is calculated by subtracting the starting coordinate from the ending coordinate of the carpooling order a and the pre-carpooling order; The dot product of the vector is calculated based on the direction vector of the carpooling order a and the pre-carpooling order. The calculation formula is: where v ax With v bx represents the x-component of two vectors, v ay With v by represents the y components of two vectors; Calculate the length of the vector of carpooling order a and pre-carpooling order respectively, and the calculation formula is: where v x With v y Represents the x component and y component of a vector, where the x component of the vector refers to the projection length of the vector on the x-axis; According to the direction vector of the carpooling order a and the pre-carpooling order, the dot product of the vector and the length of the vector are calculated to obtain the angle between the direction vectors of the two carpooling orders. The calculation formula is:
7. The device for finding carpooling partners in priority according to claim 4, characterized in that: Calculating the path similarity coefficient between carpooling orders includes: Divide the path of pre-carpooling orders and carpooling orders into a series of points with fixed intervals; Use the shortest line segment connection method to connect the points on the two paths one by one to obtain a series of line segments; In the line segments obtained by connecting the paths, the length of the line segments is calculated using the Euclidean distance method, and a distance matrix is constructed based on the length of the line segments. The length of the line segments is searched starting from the upper left corner of the distance matrix, and the maximum value is selected as the maximum continuous line segment length between the paths; Among the line segments obtained by connecting the paths, calculate the path length that can traverse both paths at any time, and select the maximum value as the maximum possible value between the paths; The path similarity coefficient is calculated based on the maximum continuous line segment length between the paths and the maximum possible value of the path length. The calculation formula is: Where PS represents the path similarity coefficient, MC represents the maximum continuous line segment length, and MP represents the maximum possible value of the path length.
8. The device for finding carpooling partners in priority according to claim 1, characterized in that: Packaging and processing of carpooling orders according to the carpooling index includes: The number of passengers in each pre-carpooling order is collected, and the pre-carpooling order and carpooling order a are sorted from high to low in terms of their order-splitting index. The pre-carpooling order with the highest order-splitting index is selected and combined with carpooling order a, and the total number of passengers does not exceed the number of passengers that the online car-hailing vehicle can carry.
