A method, apparatus, medium and device for hitchhike interline travel matching

Abstract

This application provides a technical solution for matching ridesharing trips, the method of which includes: determining the feature information corresponding to a ridesharing order, and determining the corresponding dynamic matching features based on the feature information; if the dynamic matching features meet preset triggering rules, determining the corresponding search space and the set of planned ridesharing trips of the matched ridesharing drivers based on the feature information; determining a set of candidate trip splitting points within the search space based on the set of planned ridesharing trips of the ridesharing drivers, the obtained key nodes of the road network, and real-time supply and demand hotspots; determining the score of each candidate trip splitting point for each candidate trip splitting point; selecting a preset number of candidate trip splitting points with the highest scores, and determining the simulated ridesharing trip and simulated ride-hailing trip corresponding to each candidate trip splitting point as the ridesharing trip matching result corresponding to that candidate trip splitting point. This can improve the accuracy of splitting ridesharing orders into trips and enhance the user experience.

Description

Technical Field

[0001] This application relates to the field of shared mobility data processing technology, and in particular to a technology for matching ride-sharing trips. Background Technology

[0002] With the rapid development of the internet, internet-based shared mobility has become a common mode of transportation in people's daily work and life. As one type of shared mobility model, ride-sharing, compared to the more flexible and personalized ride-hailing services, places greater emphasis on cost savings and mutual sharing. Its core advantage lies in achieving cost sharing through route matching.

[0003] Ride-sharing orders typically involve longer distances and have higher average order values. Industry data shows that although ride-sharing orders account for a small percentage of total orders, their revenue contribution is often significantly higher than their volume percentage, making them a major contributor to the revenue of ride-sharing platforms. However, current technologies usually employ a static "order-capacity" matching model. For a ride-sharing order to be accepted by a driver, a high degree of matching between the driver and passenger's origin, destination, and time window is often required. However, there is usually a significant discrepancy between the driver's planned route and the passenger's route. Generally, the longer the distance, the lower the degree of route compatibility, causing the probability of a successful ride-sharing order match to decrease exponentially with increasing distance. The completion rate for long-distance orders is typically much lower than that for short- or medium-distance orders. The inherent supply-demand mismatch in existing ride-sharing services has become a key bottleneck restricting the healthy operation and development of ride-sharing service platforms and hindering the improvement of the user experience for both passengers and drivers.

[0004] Therefore, how to dynamically reconstruct passengers' ride-sharing needs based on their ride-sharing requirements, actively couple them with real-time ride-sharing capacity, improve the accuracy of the reconstructed itinerary, and thus improve the overall completion rate of ride-sharing orders on the platform, is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] In order to at least partially solve the above-mentioned technical problems, the purpose of this application is to provide a method, apparatus, medium and equipment for matching ride-sharing trips.

[0006] According to one aspect of this application, a method for matching ridesharing trips is provided, wherein the method includes: Based on the ride-sharing orders posted by passengers, the corresponding feature information is determined, wherein the feature information includes at least: departure time, departure point, destination point, cities passed through, estimated trip distance and estimated trip time, and based on the feature information, the corresponding dynamic matching features are determined; Based on the dynamic matching features, it is determined in real time whether the preset triggering rules are met. If they are met, the corresponding search space is determined based on the starting point or the destination point. Based on the departure time, estimated travel time, cities passed through and the destination point, the set of matching ridesharing driver plans is determined. Based on the set of ridesharing driver plans, the obtained key nodes of the road network and real-time supply and demand hotspots, several candidate trip splitting points are determined in the search space to form a set of candidate trip splitting points. For each candidate trip splitting point in the candidate trip splitting point set, a simulated ridesharing trip and a simulated ride-hailing trip are constructed based on the starting point and the destination point. The route matching degree between the simulated ridesharing trip and each planned ridesharing trip in the set of planned ridesharing trips is calculated. The number of planned ridesharing trips with a route matching degree that meets a preset threshold and the highest route matching degree are determined. The estimated mileage, estimated cost, and estimated time corresponding to the simulated ridesharing trip and the estimated mileage, cost, and estimated time corresponding to the simulated ride-hailing trip are calculated respectively to determine the estimated total mileage, total cost, and total time corresponding to the candidate trip splitting point. Based on the number, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip, a preset scoring function is used to determine the score corresponding to the candidate trip splitting point. The set of candidate trip splitting points is traversed to obtain the score of each candidate trip splitting point in the set of candidate trip splitting points. A preset number of candidate trip splitting points with the highest scores are selected from the candidate trip splitting point set. The simulated ride-hailing trip and simulated carpooling trip of each candidate trip splitting point are taken as a carpooling connecting trip matching result, and a preset number of carpooling connecting trip matching results corresponding to the carpooling order are obtained.

[0007] Optionally, determining the corresponding dynamic matching features based on the feature information includes: Based on the aforementioned feature information, and according to the ride-sharing plan trip information published by all ride-sharing drivers, the number of matched ride-sharing plans, the degree of route matching for each matched ride-sharing plan trip, and the real-time waiting time are determined.

[0008] Optionally, the preset triggering rule includes any one of the following: If the number of matched ridesharing trips and the estimated trip mileage meet the first preset condition, then the event is triggered; If the highest matching degree of the matched ridesharing plan and the real-time waiting time meet the second preset condition, then it is triggered; If the real-time waiting time meets the third preset condition, then it will be triggered.

[0009] Optionally, the preset triggering rule further includes: If the historical connecting trip acceptance rate of the passenger who posted the ride-sharing order meets a preset threshold, the event is triggered, wherein the historical connecting trip acceptance rate is determined based on the passenger's historical behavior data.

[0010] Optionally, determining the corresponding search space based on the starting point or the destination point includes any one of the following: Centered on the starting point, and combining the estimated mileage and the first preset parameters, a first extended area and a second extended area are determined, and the relevant area between the first extended area and the second extended area is determined as the search space; Centered on the destination point, and combining the estimated mileage and the second preset parameters, a third extended region and a fourth extended region are determined, and the relevant area between the third extended region and the fourth extended region is determined as the search space.

[0011] Optionally, the step of determining the score corresponding to the candidate trip split point using a preset scoring function based on the quantity, the highest route matching degree, the estimated cost corresponding to the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip includes: A preset scoring function is used to calculate the score corresponding to the candidate trip split point by weighting the quantity, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip. The quantity, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip are each assigned a preset initial value with their respective weights and are continuously optimized.

[0012] Optionally, the determination of the respective corresponding preset initial values ​​of the weights includes: The historical behavior data of the passengers is clustered to determine the user profile tag type of the passengers, and based on the user profile tag type of the passengers, the preset weight initial values ​​of the quantity, the highest route matching degree, the estimated cost corresponding to the simulated ride-sharing trip, and the simulated ride-hailing trip are adjusted.

[0013] Optionally, the continuous optimization includes: Obtain historical full-path data collected based on a preset event model, and iteratively optimize the weights.

[0014] Optionally, the method for matching ridesharing trips further includes: The preset number of ridesharing trip matching results are structured and then sent to the client for display.

[0015] Optionally, the method for matching ridesharing trips further includes: Based on one or more ridesharing trip matching results confirmed by the passenger, a corresponding order group is created and published. The order group includes a master order and ride-hailing sub-orders and ridesharing orders corresponding to each ridesharing trip matching result confirmed by the passenger.

[0016] Optionally, the method for matching ridesharing trips further includes: Monitor the status of the order group and process the relevant orders in the order group when the status changes.

[0017] Optionally, the method for matching ridesharing trips further includes: Based on the ridesharing order posted by the passenger, the prepayment fee is determined. When the ridesharing trip matching result corresponding to the online car-hailing order is completed, the corresponding online car-hailing trip fee is deducted in real time. When the ridesharing trip matching result corresponding to the ridesharing order is completed, the settlement is performed.

[0018] According to another aspect of this application, an apparatus for matching ridesharing trips is provided, wherein the apparatus includes: The first module is used to determine the corresponding feature information based on the ride-sharing order posted by the passenger. The feature information includes at least: departure time, departure point, destination point, cities passed through, estimated trip distance and estimated trip time, and determines the corresponding dynamic matching features based on the feature information. The second module is used to determine in real time whether the preset triggering rules are met based on the dynamic matching features. If they are met, the corresponding search space is determined based on the starting point or the destination point. Based on the departure time, estimated travel time, cities passed through and the destination point, the matching set of ridesharing driver plans is determined. Based on the set of ridesharing driver plans, the obtained key nodes of the road network and real-time supply and demand hotspots, several candidate trip splitting points are determined in the search space to form a set of candidate trip splitting points. The third module is used to construct a simulated ridesharing trip and a simulated ride-hailing trip for each candidate trip splitting point in the candidate trip splitting point set, combining the starting point and the destination point. It calculates the route matching degree between the simulated ridesharing trip and each planned ridesharing trip in the set of planned ridesharing trips, determines the number of planned ridesharing trips with a route matching degree satisfying a preset threshold and the highest route matching degree, and calculates the estimated mileage, estimated cost, and estimated time corresponding to the simulated ridesharing trip and the simulated ride-hailing trip, respectively, to determine the estimated total mileage, total cost, and total time corresponding to the candidate trip splitting point. Based on the number, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip, a preset scoring function is used to determine the score corresponding to the candidate trip splitting point. The module iterates through the candidate trip splitting point set to obtain the score of each candidate trip splitting point in the set. The fourth module is used to select a preset number of candidate trip splitting points with the highest scores in the candidate trip splitting point set, and to take the simulated ride-hailing trip and simulated carpooling trip of each candidate trip splitting point as a carpooling connecting trip matching result, so as to obtain a preset number of carpooling connecting trip matching results corresponding to the carpooling order.

[0019] Optionally, the device for matching ridesharing trips further includes: The fifth module is used to perform structured processing on the preset number of ridesharing trip matching results and then send them to the client for display.

[0020] Optionally, the device for matching ridesharing trips further includes: The sixth module is used to create and publish corresponding order groups based on one or more ridesharing trip matching results confirmed by passengers. The order group includes a master order and ride-hailing sub-orders and ridesharing orders corresponding to each ridesharing trip matching result confirmed by the passenger.

[0021] Optionally, the device for matching ridesharing trips further includes: The seventh module is used to monitor the status of the order group and process the relevant orders in the order group when the status changes.

[0022] Optionally, the device for matching ridesharing trips further includes: The eighth module is used to determine the prepayment fee based on the ridesharing order published by the passenger. When the ridesharing sub-order corresponding to the ridesharing trip matching result is completed, the corresponding ridesharing trip fee is deducted in real time. When the ridesharing order corresponding to the ridesharing trip matching result is completed, the settlement is performed.

[0023] According to another aspect of this application, a computer-readable medium is provided, wherein computer-readable instructions are stored on the medium, which are executed by a processor to implement part or all of any of the above methods.

[0024] According to another aspect of this application, a device for matching ridesharing trips is provided, wherein the device includes: One or more processors; and a memory storing computer-readable instructions that, when executed, cause the processor to perform some or all of the operations described above.

[0025] Compared with the prior art, this application provides a method, apparatus, medium and equipment for matching ridesharing trips. The method includes: determining corresponding feature information based on the ride-sharing order posted by the passenger, wherein the feature information includes at least: departure time, departure point, destination point, cities passed through, estimated mileage, and estimated travel time; and determining corresponding dynamic matching features based on the feature information; determining in real time whether a preset triggering rule is met based on the dynamic matching feature, and if so, determining the corresponding search space based on the departure point or the destination point, and determining a set of matching ride-sharing driver planned trips based on the departure time, estimated travel time, cities passed through, and destination point; and determining several candidate trip splitting points within the search space based on the set of ride-sharing driver planned trips, obtained key road network nodes, and real-time supply and demand hotspots, forming a set of candidate trip splitting points; for each candidate trip splitting point in the set of candidate trip splitting points, constructing a simulated ride-sharing trip and a simulated ride-hailing trip in combination with the departure point and the destination point, and calculating the simulated ride-sharing trip and each ride-sharing driver's planned trip in the set of ride-sharing driver planned trips. The system calculates the route matching degree of a trip, determines the number of ridesharing driver plans with a route matching degree that meets a preset threshold, and the highest route matching degree. It also calculates the estimated mileage, estimated cost, and estimated time for the simulated ridesharing trip and the estimated mileage, cost, and time for the simulated ride-hailing trip, respectively, to determine the estimated total mileage, total cost, and total time for each candidate trip split point. Based on the number, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip, a preset scoring function is used to determine the score for each candidate trip split point. The system iterates through the candidate trip split point set to obtain the score for each candidate trip split point. A preset number of candidate trip split points with the highest scores are selected. The simulated ride-hailing trip and the simulated ridesharing trip for each candidate trip split point are combined into a single ridesharing trip matching result, resulting in a preset number of ridesharing trip matching results for the ridesharing order. This application dynamically couples the ride-sharing demand of passengers' ride-sharing orders with transportation capacity, providing passengers with several alternative ride-sharing connecting travel options corresponding to the ride-sharing order. This improves the accuracy of splitting ride-sharing orders into connecting travel options and enhances the user experience. At the same time, it can integrate different transportation resources of shared mobility service platforms that apply the above-mentioned technical solutions, thereby improving the overall order completion rate. Attached Figure Description

[0026] Other features, objectives, and potential technical effects of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1A schematic diagram is shown of a method for matching ridesharing trips according to one aspect of this application; Figure 2 This diagram illustrates a system for matching ridesharing trips according to one aspect of this application. Figure 3 This diagram illustrates the search space of an optional embodiment of one aspect of this application. Figure 4 A schematic diagram of an apparatus for matching ridesharing trips according to another aspect of this application is shown; The same or similar reference numerals in the accompanying drawings represent the same or similar parts. Detailed Implementation

[0027] The present application will now be described in further detail with reference to the accompanying drawings.

[0028] In a typical configuration of various embodiments of this application, the method execution entity, each trusted party of the system, and / or each module of the device may include one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0029] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0030] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include non-transitory computer-readable media, such as modulated data signals and carrier waves.

[0031] In existing technologies, the static "order-capacity" matching model used in ride-sharing often lacks suitable ride-sharing drivers for long-distance ride-sharing orders, resulting in a low completion rate for long-distance ride-sharing orders. This may passively increase the waiting time for passengers or lower the matching requirements, but it often reduces the timeliness of orders and the user experience.

[0032] This application provides a technical solution for matching ridesharing trips. First, based on the ridesharing order posted by the passenger, corresponding feature information and dynamic matching features are obtained. Then, based on the dynamic matching features, it is determined whether a preset triggering rule is met. If so, the trip of the ridesharing order is split into a preset number of matching trip results (or matching trip schemes) corresponding to the ridesharing order. These can serve as alternative long-distance trips for passengers to choose from, thereby maximizing the likelihood of the ridesharing order being executed. This can improve the user experience for both passengers and ridesharing drivers, and can also increase the overall ridesharing order completion rate for shared mobility service providers using this application's technical solution.

[0033] To further illustrate the technical means adopted and the effects achieved in this application, the technical solution of this application will be clearly and completely described below in conjunction with the accompanying drawings and embodiments and / or optional embodiments.

[0034] It should be noted that the collection, storage, use, processing, transmission, provision, disclosure, and application of relevant information and data involved in this application all comply with relevant laws, regulations, and standards, do not violate public order and good morals, do not harm the public interest, and necessary confidentiality measures have been taken.

[0035] Figure 1 The diagram illustrates a method for matching ridesharing trips according to one aspect of this application, wherein one embodiment of the method includes: S101 determines the corresponding feature information based on the ride-sharing order posted by the passenger. The feature information includes at least: departure time, departure point, destination point, cities passed through, estimated trip distance and estimated trip time, and determines the corresponding dynamic matching features based on the feature information. S102 Based on the dynamic matching features, it determines in real time whether the preset triggering rules are met. If they are met, it determines the corresponding search space based on the starting point or the destination point, and determines the set of matching ridesharing driver plans based on the departure time, estimated travel time, cities passed through and the destination point. Based on the set of ridesharing driver plans, the obtained key nodes of the road network and real-time supply and demand hotspots, it determines several candidate trip splitting points in the search space to form a set of candidate trip splitting points. S103 For each candidate trip splitting point in the candidate trip splitting point set, combined with the starting point and the destination point, a simulated ridesharing trip and a simulated ride-hailing trip are constructed respectively. The route matching degree between the simulated ridesharing trip and each planned ridesharing trip in the set of planned ridesharing trips is calculated. The number of planned ridesharing trips with a route matching degree that meets a preset threshold and the highest route matching degree are determined. The estimated mileage, estimated cost, and estimated time corresponding to the simulated ridesharing trip and the estimated mileage, cost, and estimated time corresponding to the simulated ride-hailing trip are calculated respectively to determine the estimated total mileage, total cost, and total time corresponding to the candidate trip splitting point. Based on the number, the highest route matching degree, the estimated cost corresponding to the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip, a preset scoring function is used to determine the score corresponding to the candidate trip splitting point. The candidate trip splitting point set is traversed to obtain the score of each candidate trip splitting point in the candidate trip splitting point set. S104 selects a preset number of candidate trip splitting points with the highest scores in the candidate trip splitting point set, and takes the simulated ride-hailing trip and simulated carpooling trip of each candidate trip splitting point as a carpooling connecting trip matching result, thereby obtaining a preset number of carpooling connecting trip matching results corresponding to the carpooling order.

[0036] This application provides a method for matching ridesharing trips, which is applied to a shared mobility service system 10. The system structure of the shared mobility service system 10 can be as follows: Figure 2 As shown, the system includes a shared mobility service platform 100 on the server side, and passenger clients 200, ride-sharing driver clients 300, and ride-hailing driver clients 400 that can interact with the shared mobility service platform 100. The shared mobility service platform 100 includes several computer devices capable of data transmission, processing, and storage. Passenger clients 200, ride-sharing driver clients 300, and ride-hailing driver clients 400 include smart terminal devices with relevant hardware and software environments and corresponding apps installed. These apps can be general-purpose, meeting the specific needs of passengers, ride-sharing drivers, and ride-hailing drivers respectively; or they can be dedicated apps installed on different clients, meeting only the specific needs of the corresponding passenger, ride-sharing driver, or ride-hailing driver. The computer devices include, but are not limited to, servers, network hosts, single network servers, or network server clusters. The smart terminal devices include, but are not limited to, smartphones, smart wearable devices, and tablets. The computer equipment and the smart terminal equipment mentioned herein are merely examples. Other existing or future equipment and / or resource platforms that are applicable to this invention should also be included within the scope of protection of this invention, and are hereby incorporated by reference.

[0037] In this application, passengers can publish ride-sharing requests through the passenger client 200. The shared mobility service platform 100 receives the ride-sharing request information published by the passenger, generates a ride-sharing order, and publishes it, which can be displayed on the ride-sharing driver client 300. Ride-sharing drivers can publish their planned trip information through the ride-sharing driver client 300. The shared mobility service platform 100 receives the planned trip information published by the ride-sharing driver, determines the planned trip of the ride-sharing driver corresponding to the driver (which can be an ordered sequence of trajectory points from the departure point to the destination point), and includes it in the ride-sharing driver planned trip database (or ride-sharing capacity pool).

[0038] In this application, the route matching degree can be determined based on the ride-sharing plan information posted by the ride-sharing driver and the ride-sharing order posted by the passenger, referring to information such as their respective departure points, destination points, departure times, cities passed through, estimated distance, and estimated travel time, and comprehensively considering factors such as time window overlap, route overlap, and directional angle. The closer the departure times, distances between their departure points, and distances between their destination points are, and the smaller the directional angle, the higher the route matching degree. The cities passed through can usually be obtained from the posted plan or order through an integrated electronic map.

[0039] For example, suppose a passenger posts a ridesharing order departing from point A1 in city A to point C1 in city C at 18:00, and a ridesharing driver posts a ridesharing order departing from point A2 in city A to point C2 in city C at 17:40. The expected trajectories / routes of both can be determined, and their path overlap can be calculated, assuming the path overlap meets a preset threshold. If the angle between the ridesharing driver's route direction (which can be the direction of the line connecting A2 to C2) and the ridesharing order's route direction (which can be the direction of the line connecting A1 to C1) is less than the preset threshold, then the two are considered to be in the same direction. If the time window is [17:30, 18:30], then the times also match. The path overlap, direction angle, and time matching degree can be combined, referring to a preset determination method, to determine the matching degree of the above ridesharing driver's planned route to the ridesharing order.

[0040] In this embodiment, in step S101, the shared mobility service platform 100 can determine the corresponding feature information based on the ride-sharing order posted by the passenger. The feature information corresponding to the ride-sharing order may include at least the following: the passenger's expected departure time, departure point (which may be the latitude and longitude of the departure location), destination point (which may be the latitude and longitude of the destination), cities included in the expected trip, expected trip mileage, and expected trip time. Furthermore, based on the feature information corresponding to the ride-sharing order, the dynamic matching features corresponding to the ride-sharing order can be determined.

[0041] Continuing in this embodiment, in step S102, the shared mobility service platform 100 can determine in real time whether the preset triggering rules are met based on the dynamic matching features obtained in step S101. If the dynamic matching features meet the preset triggering rules, the ride-sharing order splitting process is triggered. In the ride-sharing order splitting process, the corresponding search space can be determined based on the departure point or destination point in the feature information corresponding to the ride-sharing order. Based on the departure point, departure time, estimated travel time, cities passed through, and destination point in the feature information corresponding to the ride-sharing order, all matching ride-sharing driver planned trips are determined in the ride-sharing capacity pool to form a ride-sharing driver planned trip set. Then, based on the ride-sharing driver planned trip set, real-time acquired key road network nodes, and real-time supply and demand hotspots, several candidate trip splitting points are determined in the above search space to form a candidate trip splitting point set. For example, assuming the departure time for a passenger's ride-sharing order is T, and a fixed duration ΔT (e.g., 30 minutes) and a matching threshold P are set, a time window [T-ΔT, T+ΔT] can be determined. First, ride-sharing drivers whose departure times fall within this window from the real-time dynamic ride-sharing capacity pool can be filtered out to obtain an initial set of planned trips for these drivers (if the number is too small, the time window can be increased, for example, ΔT can be set to 60 minutes). Then, the matching degree between each driver's planned trip and the passenger's ride-sharing order's trip is calculated. Drivers satisfying P are then filtered out to obtain a set of all matched planned trips for these drivers. Through the integrated electronic map, key road network nodes (e.g., pick-up and drop-off points corresponding to highway entrances / exits, service areas, and main road intersections) can be obtained in real time. It can perform clustering on the planned trips of ride-sharing drivers and the departure and destination points of ride-sharing orders in the real-time dynamic capacity pool, and combine it with historical supply and demand hotspot data to obtain real-time supply and demand hotspots.

[0042] Continuing in this embodiment, in step S103, the shared mobility service platform 100 can evaluate each candidate trip splitting point in the candidate trip splitting point set obtained in step S102: First, combining the departure point and destination point in the feature information corresponding to the ride-sharing order, a simulated ride-sharing trip and a simulated ride-hailing trip are constructed respectively, that is, the ride-sharing trip is split into connecting trips, including a relatively short ride-hailing trip that can be flexibly connected and a relatively long ride-sharing trip with a high degree of matching along the route. Next, the matching degree between the simulated ridesharing trip and each planned ridesharing trip in the set of planned ridesharing trips obtained in step S102 is calculated. The number of planned ridesharing trips with a matching degree meeting a preset threshold and the highest matching degree are determined. The estimated mileage, estimated cost, and estimated time corresponding to the simulated ridesharing trip and the simulated ride-hailing trip are also calculated. Based on the above calculation results, the estimated total mileage, total cost, and total time corresponding to the candidate trip split point are determined. Then, based on the number of planned ridesharing trips with a matching degree meeting the preset threshold, the highest matching degree, the estimated cost of the simulated ridesharing trip corresponding to the candidate trip split point, and the estimated mileage of the corresponding simulated ride-hailing trip, a preset scoring function is used to determine the score corresponding to the candidate trip split point. By iterating through the candidate trip splitting point set obtained in step S102 and repeating the above steps, the score of each candidate trip splitting point in the candidate trip splitting point set can be obtained.

[0043] Continuing in this embodiment, in step S104, the shared mobility service platform 100 may select the top preset number of candidate trip splitting points with the highest scores in the candidate trip splitting point set, and take the simulated ride-hailing trip and simulated carpooling trip of each candidate trip splitting point in the preset number of candidate trip splitting points determined in step S103 as a carpooling connecting trip matching result, and obtain the preset number of carpooling connecting trip matching results corresponding to the carpooling order.

[0044] Through the above embodiments, the shared mobility service platform 100 can detect whether to trigger the splitting of a ride-sharing order based on its characteristic information. If a departure is confirmed, the splitting process is initiated, and the order is coupled with real-time ride-sharing capacity to filter candidate splitting points. Based on the scores of these candidate splitting points, a connecting travel matching solution is reconstructed. This allows for timely and autonomous identification of ride-sharing orders with low matching rates to the planned trips of drivers in the real-time ride-sharing capacity pool from a large number of real-time ride-sharing orders. By triggering splitting, the platform obtains the most likely connecting travel alternative for passengers, ensuring both the necessity and accuracy of platform intervention, increasing the likelihood of subsequent order completion, and improving the user experience.

[0045] Optionally, in step S101, determining the corresponding dynamic matching features based on the feature information includes: Based on the aforementioned feature information, and according to the ride-sharing plan trip information published by all ride-sharing drivers, the number of matched ride-sharing plans, the degree of route matching for each matched ride-sharing plan trip, and the real-time waiting time are determined.

[0046] In this optional embodiment, the shared mobility service platform 100 can, based on the characteristic information corresponding to the determined passenger's ride-sharing order, attempt to match each ride-sharing plan trip information published by all ride-sharing drivers in the ride-sharing capacity pool in real time, and determine the dynamic matching characteristics corresponding to the ride-sharing order, namely: the number of successfully matched ride-sharing plans, the degree of matching the route corresponding to each successfully matched ride-sharing plan trip, and the real-time waiting time after the ride-sharing order is published (i.e., from the time of publication to the current time before the ride-sharing driver has accepted the order).

[0047] Optionally, in step S102, the preset triggering rule includes any one of the following: If the number of matched ridesharing trips and the estimated trip mileage meet the first preset condition, then the event is triggered; If the highest matching degree of the matched ridesharing plan and the real-time waiting time meet the second preset condition, then it is triggered; If the real-time waiting time meets the third preset condition, then it will be triggered.

[0048] The preset triggering rules for ride-sharing orders can be set based on the operational goals and business logic of the actual application scenario. In this optional embodiment, the preset triggering rules in step S102 may include any of the following: If the number of successfully matched ride-sharing trips in the dynamic matching features of the ride-sharing order meets a preset threshold (e.g., less than 3), and the estimated mileage in the feature information of the ride-sharing order also meets a preset threshold (e.g., more than 100 kilometers), that is, the number of successfully matched ride-sharing trips and the corresponding estimated mileage in the dynamic matching features of the ride-sharing order meet the first preset condition, then the splitting of the ride-sharing order can be triggered.

[0049] If the highest on-route matching degree of the successfully matched ride-sharing trip in the dynamic matching features of the ride-sharing order meets a preset threshold (e.g., not exceeding 60%), and the real-time waiting time also meets a preset threshold (e.g., exceeding 30 minutes), that is, the highest on-route matching degree and real-time waiting time of the successfully matched ride-sharing trip in the dynamic matching features of the ride-sharing order meet the second preset condition, then the splitting of the ride-sharing order can be triggered.

[0050] If the real-time waiting time in the dynamic matching feature of the ride-sharing order meets the third preset condition (e.g., more than 2 hours), the splitting of the ride-sharing order can be triggered.

[0051] Optionally, the preset triggering rule further includes: If the historical connecting trip acceptance rate of the passenger who posted the ride-sharing order meets a preset threshold, the event is triggered, wherein the historical connecting trip acceptance rate is determined based on the passenger's historical behavior data.

[0052] Triggering rules can also be set by combining passengers' historical behavior data. In this optional embodiment, the preset triggering rule in step S102 may include: if the historical connecting trip adoption rate of the passenger who posted the ride-sharing order meets a preset threshold (for example, 25%, that is, more than 25% of the ride-sharing orders he has completed are completed in the form of connecting trips), then the splitting of the ride-sharing order can also be triggered.

[0053] In this application, a corresponding context can also be generated based on the specific rules for the ride-sharing order, which can be sent to the passenger's client 200 and displayed on the relevant page. This context can also be used as the passenger's historical behavior data for continuous optimization.

[0054] Optionally, in step S102, determining the corresponding search space based on the starting point or the destination point includes any one of the following: Centered on the starting point, and combining the estimated mileage and the first preset parameters, a first extended area and a second extended area are determined, and the relevant area between the first extended area and the second extended area is determined as the search space; Centered on the destination point, and combining the estimated mileage and the second preset parameters, a third extended region and a fourth extended region are determined, and the relevant area between the third extended region and the fourth extended region is determined as the search space.

[0055] In this optional embodiment, in step S102, the shared mobility service platform 100 can use the latitude and longitude of the departure point in the feature information corresponding to the ride-sharing order as the center, and combine it with the estimated mileage and the first preset parameters in the feature information to determine a first extended area and a second extended area, and determine the relevant area between the first extended area and the second extended area as the search space. Alternatively, the shared mobility service platform 100 can also use the latitude and longitude of the destination point in the feature information corresponding to the ride-sharing order as the center, and combine it with the estimated mileage and the second preset parameters in the feature information to determine a third extended area and a fourth extended area, and determine the relevant area between the third extended area and the fourth extended area as the search space. The search space determined by any of the above methods can ensure the basic connection distance for subsequent connecting trips determined based on the determined split points, while limiting the maximum connection range, so as to balance the time cost, cost and flexibility of splitting into connecting trips, and provide passengers with a better experience. Among them, the extended area can be a circle or block shape determined according to the estimated mileage and preset parameters, and the relevant area as the search space can be determined according to the corresponding extended area. An exemplary search space is as follows: Figure 3 As shown in a and b in Figure 3, assuming a starting point A, a destination point C, and an expected travel distance D, the first preset parameters may include the minimum radius Dmin, the inner diameter percentage PAin, the maximum radius Dmax, and the outer diameter percentage PAout. The second preset parameters may include the minimum radius Dmin, the inner diameter percentage PCin, the maximum radius Dmax, and the outer diameter percentage PCout. The radius RAin of the first extended region and the radius RAout of the second extended region can be calculated according to the following formulas (1) and (2). The radius RCin of the third extended region and the radius RCout of the fourth extended region can be determined according to the following formulas (3) and (4). The relevant regions (the shaded areas in a or b of Figure 3) can be determined as the search space.

[0056] (1) (2) (3) (4) In this application, the shared mobility service platform 100 can automatically collect passenger behavior data in real time and statistically analyze historical passenger behavior data to obtain relevant statistical features, such as the exposure rate, adoption rate, completion rate, cancellation rate, and adoption rate curves and completion rate curves for connected travel plans with different connection distances. The shared mobility service platform 100 can dynamically adjust the preset first and / or second preset parameters based on the relevant statistical features. For example, if the relevant statistical features show that the passenger adoption rate drops sharply when the connection distance exceeds 25 kilometers, D_max can be automatically reduced from 50 kilometers to 30 kilometers, and this can be verified based on subsequently collected passenger behavior data. Therefore, in actual implementation, the initial values ​​of the first and / or second preset parameters can be preset, and the shared mobility service platform 100 can automatically collect passenger behavior data in real time, statistically analyze relevant features, and dynamically adjust the parameters based on the statistical features. Of course, manual adjustment is also possible depending on the actual application scenario.

[0057] Optionally, in step S103, for a candidate trip splitting point, a simulated ridesharing trip and a simulated ride-hailing trip are constructed by combining the departure point and the destination point, including any one of the following: Construct simulated ride-hailing trips from the starting point to the candidate trip splitting point and simulated carpooling trips from the candidate trip splitting point to the destination point, respectively. Simulated ridesharing trips from the starting point to the candidate trip splitting point and simulated ride-hailing trips from the candidate trip splitting point to the destination point are constructed respectively.

[0058] In this optional embodiment, in step S103, for each candidate trip splitting point in the candidate trip splitting point set obtained in step S102, a simulated ride-sharing trip and a simulated ride-hailing trip can be constructed by combining the departure point and destination point in the feature information corresponding to the passenger's ride-sharing order. One optional implementation is: simulated ride-hailing trip + simulated ride-sharing trip, that is, a simulated ride-hailing trip from the departure point to the candidate trip splitting point and a simulated ride-sharing trip from the candidate trip splitting point to the destination are constructed respectively. Another optional implementation is: simulated ride-sharing trip + simulated ride-hailing trip, that is, a simulated ride-sharing trip from the departure point to the candidate trip splitting point and a simulated ride-hailing trip from the candidate trip splitting point to the destination are constructed respectively.

[0059] Optionally, in step S103, the step of determining the score corresponding to the candidate trip split point using a preset scoring function based on the quantity, the highest route matching degree, the estimated cost corresponding to the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip includes: A preset scoring function is used to calculate the score corresponding to the candidate trip split point by weighting the quantity, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip. The quantity, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip are each assigned a preset initial value with their respective weights and are continuously optimized.

[0060] In this optional embodiment, in step S103, the shared mobility service platform 100 can use the number of ridesharing driver planned trips with a matching degree that meets a preset threshold, the highest matching degree among them, the estimated cost of a simulated ridesharing trip from the candidate trip split point to the destination, and the estimated mileage of a simulated ride-hailing trip from the departure point to the candidate trip split point as weighting factors. A preset scoring function is used, combined with the preset initial values ​​of the weights of each weighting factor, to perform a weighted calculation, thereby obtaining the score corresponding to the candidate trip split point. An exemplary preset scoring function can be expressed as follows: (5) (5) Where Bi is the candidate trip split point; Fi is the estimated cost of the simulated ridesharing trip from the candidate trip split point to the destination. The highest rate of matching the route within a rideshare driver's planned trip, where the matching rate meets a preset threshold. The number of ridesharing drivers whose planned trips meet a preset threshold for route matching. The estimated mileage of the simulated ride-hailing trip from the starting point to the candidate trip split point; , , , Fi, respectively , , Each has its corresponding weight, and the sum of the values ​​of all weights is 1.

[0061] The weights of the following preset values ​​can be adjusted based on the passenger profile tag type to make the score calculated by the weight combination more accurate. These values ​​include the number of ridesharing driver plans that meet the preset threshold for the degree of matching along the route, the highest degree of matching along the route, the estimated cost of the simulated ridesharing trip from the candidate trip split point to the destination, and the estimated mileage of the simulated ride-hailing trip from the departure point to the candidate trip split point.

[0062] Optionally, the determination of the respective corresponding preset initial values ​​of the weights includes: The historical behavior data of the passengers is clustered to determine the user profile tag type of the passengers, and based on the user profile tag type of the passengers, the preset weight initial values ​​of the quantity, the highest route matching degree, the estimated cost corresponding to the simulated ride-sharing trip, and the simulated ride-hailing trip are adjusted.

[0063] In this optional embodiment, the shared mobility service platform 100 can perform clustering processing on the historical behavior data of the collected passenger, and determine the user profile tag type of the passenger based on the clustering processing results. Based on the user profile tag type of the passenger, and combined with the preset user profile tag type and preset weight initial value adjustment rules or mapping relationship, the platform can adjust the preset weight initial values ​​corresponding to the number of ridesharing driver planned trips with a matching degree that meets the preset threshold, the highest matching degree among them, the estimated cost of the simulated ridesharing trip from the candidate trip split point to the destination, and the estimated mileage of the simulated ride-hailing trip from the departure point to the candidate trip split point. This allows for the use of differentiated weight combinations to calculate scores for passengers with different user tag types.For example, clustering can be performed based on passengers' historical behavior data: First, collect passengers' historical behavior data, such as behavior data within a preset time period (e.g., the most recent 3 months). This may include: posted ridesharing order information, waiting time for each order's acceptance or cancellation, negotiation information (e.g., whether negotiation was initiated for each order and negotiation data), reasons for order cancellation (e.g., price, waiting time, or others), connecting trip related information (e.g., whether it was split into connecting trips, whether the connecting trip's split points were changed), and discount usage (e.g., whether coupons were used or not, what type of coupons, etc.). Second, process the collected passenger history... The behavioral data is processed and quantified to obtain passenger-related behavioral feature vectors. For example, price features can be obtained based on data such as negotiation information, order cancellation reasons, and discount usage; time features can be obtained based on data such as waiting time and order cancellation reasons; and distance and location features can be obtained based on data such as connecting trip information. The third step is to standardize / normalize the passenger-related behavioral feature vectors (because the feature values ​​in different feature vectors are of different types and / or ranges, direct clustering will cause anomalies, so standardization / normalization is required to ensure that the feature values ​​in the processed vectors are all in the range [0,1]); the fourth step is to perform clustering, which can be done using K-means clustering. Common clustering algorithms such as Means predefine the mapping relationship between several passenger user profile label types and behavioral feature vectors. For example, these might be categorized as price-sensitive (sensitive to cost / price), time-sensitive (sensitive to waiting time), distance-sensitive (sensitive to connection distance), indifferent (insensitive to cost / price, waiting time, and connection distance), and mixed (sensitive but difficult to attribute). The standardized / normalized passenger behavioral feature vectors are input into the clustering algorithm for clustering, resulting in several clusters. Each cluster corresponds to a user profile label type, and the cluster can be determined based on the mean value of the feature vectors of all passengers within that cluster. The corresponding user profile label types can be used as well. For example, if the average feature value of the price vector for all passengers in one cluster is the highest, then the user profile label type for all passengers in that cluster can be identified as price-sensitive; similarly, if the average feature value of the time vector for all passengers in another cluster is the highest, then the user profile label type for all passengers in that cluster can be identified as time-sensitive. Furthermore, the clustering results can be verified based on passengers' historical behavior data. For instance, for passengers whose user profile label type is price-sensitive, is their acceptance rate for lower-priced connecting travel options significantly higher than for other connecting travel options? If so, the relevant clustering results are reliable.Furthermore, since passenger behavior can change, the passenger user profile tag type may also change. In order to match in real time, the passenger user profile tag type can be dynamically updated periodically or irregularly by combining the updated historical behavior data of passengers, so as to dynamically adjust the weight combination of the corresponding preset scoring function and ensure the accuracy of each score as much as possible.

[0064] To continuously improve the adoption and completion rates of the split connecting travel plans among subsequent passengers, thereby enhancing the user experience, the weight combination can be continuously optimized. Optionally, the continuous optimization includes: Obtain historical full-path data collected based on a preset event model, and iteratively optimize the weights.

[0065] In this optional embodiment, historical full-path data of several passengers can be collected based on a preset event model as online A / B test data. Through online A / B testing, the weight combinations of the aforementioned dimensions are continuously iterated and tested, and the results are continuously optimized. The online A / B test data refers to passenger behavior data obtained through comparative experiments. Specifically, participating passengers (operators can select passengers for participation based on actual application scenarios) can be randomly divided into an experimental group and a control group. The experimental group uses weight combination A, and the control group uses weight combination B. During the experiment, the interaction behavior data of the two groups of passengers is recorded, and relevant characteristics are statistically analyzed: test ridesharing order volume, acceptance rate of split connecting trips, completion rate, cancellation rate, etc. The statistical characteristics of the two groups are compared to determine which weight combination is better. A / B testing can be conducted periodically after collecting passenger behavior data, and the weight values ​​can be updated based on the test results for continuous optimization. Machine learning models such as Bayesian optimization and reinforcement learning can also be used to continuously optimize the weight values. One approach combines continuous tuning with a Bayesian optimization model. The weight parameters are treated as hyperparameters, with the completion rate as the objective function. Gaussian process modeling is used to explore the parameter space. Its advantages include high sample efficiency, making it suitable for A / B testing with limited samples. The input to the Bayesian optimization algorithm is sample data consisting of historical weight parameter combinations and their corresponding completion rates; the output is the optimized weight parameter combination to be tested and validated. Another approach combines continuous tuning with a reinforcement learning model. Each weight parameter combination is treated as a decision, and the strategy is adjusted in real-time based on statistical characteristics (e.g., adoption rate). These states can include passenger characteristics, ride-sharing order characteristics, and candidate connecting travel scheme characteristics. Based on the weight combination selected by the model, a positive reward is given if the passenger adopts and completes the order, and a negative reward is given if the order is canceled or not completed. The model updates based on each reward result, thus achieving continuous tuning.

[0066] In this application, data points are pre-implemented across the entire data processing chain of the shared mobility service platform 100, defining a unified event model and specifying the format and content of each collected data record. For example, the following types of events can be defined in this application: 1. Trigger Event: Recorded when intervention is required for a ride-sharing order. The recorded data includes at least the ride-sharing order ID, which preset trigger rule was triggered, and the number of matches. 2. Exposure Event: Recorded when the connecting travel matching results are sent to the client and displayed to passengers. The recorded data includes at least one or more recommended connecting travel matching results and the display page. 3. Click Event: Recorded when a passenger clicks on a recommended connecting travel matching result on a relevant page in the client. The recorded data includes at least the click location / card / button and the passenger's real-time waiting time. 4. Browsing Event: Recorded when a user enters the details page of a recommended connecting travel matching result after clicking on it in the client. The recorded data includes at least the browsing time on the details page and which one or more recommended connecting travel matching results were viewed. 5. Selection Event: Recorded when a user selects a recommended connecting trip match on the client. The recorded data should at least include the user's selected connecting trip match; 6. Order Placement Event: Recorded when the ride-sharing service platform 100 creates an order group corresponding to the passenger's selection. The recorded data should at least include the order group ID, main order ID, and a list of sub-order IDs; 7. Order Acceptance Event: Recorded when a ride-sharing order or sub-order is confirmed by a ride-sharing driver. The recorded data should at least include the accepted order ID or sub-order ID, the accepting ride-sharing driver, and the order acceptance timestamp; 8. Fulfillment Event: Recorded when the entire connecting trip is completed. The recorded data should at least include the actual completed connecting trip mileage and actual cost; 9. Settlement Event: Recorded when payment is completed. The recorded data should at least include the final payment amount. By connecting all these events along the entire path, the entire process of each ride-sharing order from its inception to its completion can be reconstructed, forming a data closed loop. Based on the collected full-path data, historical passenger behavior data and core indicators can be obtained, and a core indicator dashboard can be established. Attribution analysis of the full-path data can also be performed to iteratively evaluate and adjust the weight combination of preset trigger rules and preset scoring functions. For example, core indicators may include trigger rate (which reflects the coverage of ride-sharing order splitting triggers; too high an indicator may indicate that normal completion of long-distance ride-sharing is not ideal, while too low an indicator indicates that the conditions for triggering splitting are too strict), exposure click-through rate (this indicator can be used as a core indicator; if too low an indicator indicates that the connecting travel matching results recommended by the shared mobility service platform are not attractive enough to passengers and need optimization), order connecting conversion rate (which reflects the conversion efficiency of ride-sharing orders into connecting sub-orders), completion rate, and the ratio of completed connecting sub-orders to completed normal ride-sharing orders (the comparison of connecting sub-order completion rate and ride-sharing order completion rate reflects the effectiveness of ride-sharing order splitting), etc.Attribution analysis can be used to evaluate the effect of different triggering rules and weight combinations of different scoring functions. An exemplary attribution analysis process is as follows: First, label each split ride-sharing order and record which preset triggering rule it was triggered by and the weight combination used by the preset scoring function. Second, divide the passengers corresponding to the split ride-sharing orders into different groups according to the labels (it should be ensured that these passengers are not significantly different in other aspects). Third, calculate the ROI (Return on Investment) for each group. ROI = (Revenue generated - Cost consumed) ÷ Cost consumed × 100%, where revenue generated = additional completed orders × average order value. (Assuming that for the same 1000 exposures of the connecting travel plan, group 1 can complete 10 more orders, while group 2 can only complete 2 more orders. If the average order value of each split ride-sharing order is 200 yuan, then the revenue generated by group 1 is 2000 yuan, and the revenue generated by group 2 is 400 yuan.) The costs incurred are mainly technical costs and user experience costs. Technical costs refer to development and computing power investment, while user experience costs can be quantified by the negative feedback rate, such as the percentage of users who clicked "not interested." The fourth step is to compare the ROI of different groups to determine the weight combination of the preset trigger rules and / or preset scoring functions for the groups that should be optimized. (For example, if the ROI of group 1 is 50% and the ROI of group 2 is 10%, then the weight combination of the preset trigger rules and / or preset scoring functions for group 1 should be adopted and continuously optimized.)

[0067] Optionally, the method for matching ridesharing trips further includes: S105 performs structured processing on the preset number of ridesharing trip matching results and sends them to the client for display.

[0068] In this optional embodiment, in step S105, the shared mobility service platform 100 can perform structured processing on a preset number of ridesharing trip matching results corresponding to the passenger's ridesharing order obtained in step S104, and send personalized recommendations to the passenger client 200 after structured processing, so as to expose and display them in the relevant pages of the passenger client 200 in a contextualized manner. For example, it can be displayed directly as a non-intrusive card, or on the ridesharing order-related page, after the passenger triggers an operation, it can display the context (e.g., "There are fewer ridesharing drivers currently, intelligent recommendation of faster departure options"). When the passenger performs further operations, they can enter a detailed page displaying each recommended ridesharing trip matching solution, so that the passenger can browse and understand feasible alternative solutions that can meet their ridesharing travel requirements. The structured data obtained after structured processing may include: trip splitting point information corresponding to each ridesharing trip matching result, information on the two segments of the trip, estimated time and estimated cost, estimated order acceptance rate improvement tag information, etc. The estimated order acceptance rate increase can be determined based on the initial order acceptance rate corresponding to the ridesharing order and the joint trip order acceptance rate corresponding to the proposed connecting trip. The initial order acceptance rate can be determined using an order acceptance rate prediction model trained on historical passenger behavior data. The joint trip order acceptance rate corresponding to the proposed connecting trip can be determined based on two dimensions: the number of ridesharing driver planned trips whose route matching degree at the trip split point meets a preset threshold, and the highest route matching degree among them, combined with a preset lookup table. Continuing with the above example, the preset lookup table used to determine the joint trip order acceptance rate corresponding to the trip split point Bi can be as shown in Table 1 below. Table 1

[0069] If a passenger selects one or more ride-sharing trip matching results after browsing, the results will be sent to the shared mobility service platform 100 through the passenger client 200 (for example, through "one-click ride-sharing order", passengers can select one or more ride-sharing trip matching options recommended by the platform, and after submission, the platform will automatically create the main order and multiple corresponding sub-orders).

[0070] Optionally, the method for matching ridesharing trips further includes: S106 creates and publishes a corresponding order group based on one or more ridesharing trip matching results confirmed by the passenger. The order group includes a master order and ride-hailing sub-orders and ridesharing orders corresponding to each ridesharing trip matching result confirmed by the passenger.

[0071] In this optional embodiment, in step S106, the shared mobility service platform 100 can create a corresponding collaborative order group based on one or more ridesharing matching results selected and confirmed by the passenger in step S104, and publish it to the passenger client 200 to expose and display the order group to the passenger for browsing. The order group includes a main order (which can be the original ridesharing order, serving as an anchor point) and ridesharing sub-orders corresponding to each ridesharing matching result confirmed by the passenger (i.e., one ridesharing matching result confirmed by a passenger corresponds to one ridesharing order and one ridesharing order). Passengers can access relevant information about the order group on the relevant page of the passenger client 200, including but not limited to: order group ID, main order ID, sub-order ID list, and / or the estimated trip, estimated cost, and estimated travel time corresponding to each order. The ride-sharing service platform 100 can manage the lifecycle of an order group using an order collaboration service. Through order collaboration, a ridesharing order is broken down and packaged into one or more complete connecting travel service orders, including ridesharing sub-orders and ride-hailing sub-orders. When any sub-order in the order group is accepted, the order collaboration service automatically cancels the main order and other connecting travel service orders within it, ensuring that only the connecting travel service corresponding to that sub-order is executed. If a passenger accepts a set of ride-hailing and ridesharing orders from the order group, the passenger's client 200 will click to confirm, and the confirmation result will be sent to the ride-sharing service platform 100.

[0072] Optionally, the method for matching ridesharing trips further includes: S107 monitors the status of the order group and processes the relevant orders in the order group when its status changes.

[0073] In this optional embodiment, in step S107, the shared mobility service platform 100 can monitor the status of published order groups in real time. When the status changes, it processes the relevant orders in the order group. When an order group is created, the status of the order group, the main order, and all sub-orders is "pending acceptance." The status of the order group and its corresponding main and sub-orders may also include "accepted," "in progress," "cancelled," and "completed." The status of the order group can be monitored through message subscription. When the status changes, notifications are sent to relevant subscribers (such as passenger client 200, ride-sharing driver client 300, ride-hailing driver client 400, and relevant components / services of the shared mobility service platform 100). For example, when the ride-sharing service platform 100 receives a group of ride-hailing and carpooling orders from an order group selected by a passenger based on interactive operations on the passenger client 200, it can change the status of the ride-hailing and carpooling orders in the order group to "accepted" and publish the ride-hailing order to the ride-hailing driver's client and the carpooling order to the carpooling driver's client, respectively. When a ride-hailing order is accepted, its status is changed to "in execution"; when a carpooling order is accepted, its status is changed to "in execution".

[0074] Optionally, the method for matching ridesharing trips further includes: S108 determines the prepayment fee based on the ridesharing order published by the passenger. When the ridesharing trip matching result corresponding to the online car-hailing order is completed, the corresponding online car-hailing trip fee is deducted in real time. When the ridesharing trip matching result corresponding to the ridesharing order is completed, the settlement is performed.

[0075] In this optional embodiment, in step S108, the shared mobility service platform 100 can further determine the prepayment fee based on the ride-sharing order posted by the passenger. After the passenger pays the prepayment fee, the platform posts a ride-hailing order to the ride-hailing driver client 400 and a ride-sharing order to the ride-sharing driver client 300. When the ride-hailing order is accepted and completed by the ride-hailing driver, the corresponding ride-hailing fare is deducted from the prepayment paid by the passenger in real time. When the ride-sharing order is accepted and completed by the ride-sharing driver, the platform settles the accounts based on the actual ride-sharing fare and the remaining prepayment fee after deduction, with refunds for overpayments and additional payments for underpayments.

[0076] The above embodiments and / or optional embodiments provide a method for matching ridesharing trips, which can perceive the characteristics of ridesharing orders in real time, accurately identify ridesharing orders that meet preset triggering rules, and trigger trip splitting at the optimal time. It can be actively coupled with real-time ridesharing capacity supply to dynamically determine the optimal preset number of split-trip matching results (or ridesharing matching schemes), improving the accuracy of splitting eligible ridesharing orders into ridesharing trips and enhancing user experience. Furthermore, based on the interactive operations of passengers and drivers on the client, the order groups corresponding to the ridesharing matching results selected by passengers are dynamically tracked and managed throughout the entire process through process collaboration. This enables dynamic monitoring, coordinated processing, and settlement of the status of the main order and sub-orders within the order group, ensuring seamless connection between cross-order and cross-trip trips, providing a guarantee for the smooth completion of ridesharing orders, and improving user experience. It can also improve the overall completion rate of ridesharing orders on the shared mobility service platform 100.

[0077] Figure 4 This application illustrates an apparatus for matching ridesharing trips according to another aspect of the present application, wherein, in one embodiment, the apparatus includes: The first module 410 is used to determine the corresponding feature information based on the ride-sharing order posted by the passenger. The feature information includes at least: departure time, departure point, destination point, cities passed through, estimated trip distance and estimated trip time, and determines the corresponding dynamic matching features based on the feature information. The second module 420 is used to determine in real time whether the preset triggering rules are met based on the dynamic matching features. If they are met, the corresponding search space is determined based on the starting point or the destination point. Based on the departure time, estimated travel time, cities passed through and the destination point, the matching set of ridesharing driver plans is determined. Based on the set of ridesharing driver plans, the obtained key nodes of the road network and real-time supply and demand hotspots, several candidate trip splitting points are determined in the search space to form a set of candidate trip splitting points. The third module 430 is used to construct a simulated ridesharing trip from each candidate trip splitting point in the candidate trip splitting point set to the destination point, calculate the route matching degree between the simulated ridesharing trip and each planned ridesharing trip in the planned ridesharing trip set, determine the number of planned ridesharing trips with a route matching degree that meets a preset threshold and the highest route matching degree, and construct a simulated ride-hailing trip from the starting point to the candidate trip splitting point, and calculate the pre-set route matching degree corresponding to the simulated ridesharing trip. The estimated mileage, estimated cost, and estimated time are used to determine the estimated total mileage, total cost, and total time corresponding to the candidate trip splitting point. Based on the quantity, the highest route matching degree, the estimated cost corresponding to the simulated ride-sharing trip, and the estimated mileage of the simulated ride-sharing trip, a preset scoring function is used to determine the score corresponding to the candidate trip splitting point. The candidate trip splitting point set is traversed to obtain the score of each candidate trip splitting point in the candidate trip splitting point set. The fourth module 440 is used to select a preset number of candidate trip splitting points with the highest scores in the candidate trip splitting point set, and to take the simulated ride-hailing trip and simulated carpooling trip of each candidate trip splitting point as a carpooling connecting trip matching result, so as to obtain a preset number of carpooling connecting trip matching results corresponding to the carpooling order.

[0078] In this embodiment, the device is deployed in a shared mobility service platform, and the software and hardware environment of the shared mobility service platform is the same as that of the aforementioned shared mobility service platform 100.

[0079] In this embodiment of the device, the first module 410 can determine the corresponding feature information based on the ride-sharing order posted by the passenger. The feature information corresponding to the ride-sharing order may include at least the following: the passenger's expected departure time, departure point (which may be the latitude and longitude of the departure location), destination point (which may be the latitude and longitude of the destination), cities included in the expected trip, expected trip mileage, and expected trip time. Furthermore, based on the feature information corresponding to the ride-sharing order, the dynamic matching features corresponding to the ride-sharing order can be determined.

[0080] Continuing with this device embodiment, the second module 420 of the device can determine in real time whether the preset triggering rules are met based on the dynamic matching features obtained through the first module 410. If the dynamic matching features meet the preset triggering rules, the ride-sharing order splitting process is triggered. In the ride-sharing order splitting process, the corresponding search space can be determined based on the departure point and destination point in the feature information corresponding to the ride-sharing order. Based on the departure time, estimated trip time, cities passed through, and destination point in the feature information corresponding to the ride-sharing order, all matching ride-sharing driver planned trips are determined in the ride-sharing capacity pool to form a ride-sharing driver planned trip set. Then, based on the ride-sharing driver planned trip set, real-time acquired key road network nodes, and real-time supply and demand hotspots, several candidate trip splitting points are determined within the above search space to form a candidate trip splitting point set.

[0081] Continuing with this device embodiment, the third module 430 of the device can evaluate each candidate trip splitting point in the candidate trip splitting point set obtained by the second module 420: First, by combining the departure point and destination point in the feature information corresponding to the ride-sharing order, a simulated ride-sharing trip and a simulated ride-hailing trip are constructed respectively, that is, the ride-sharing trip is split into connecting trips, including a relatively short ride-hailing trip that can be flexibly connected and a relatively long ride-sharing trip with a high degree of matching along the route. Secondly, when using the simulated ridesharing trip, the degree of route matching between it and each planned ridesharing trip in the set of planned ridesharing trips obtained through the second module 420 is calculated. The number of planned ridesharing trips with a route matching degree meeting a preset threshold and the highest route matching degree are determined. The estimated mileage, estimated cost, and estimated time corresponding to the simulated ridesharing trip and the simulated ride-hailing trip are also calculated respectively. Based on the above calculation results, the estimated total mileage, total cost, and total time corresponding to the candidate trip split point are determined. Then, based on the number of planned ridesharing trips with a route matching degree meeting the preset threshold, the highest route matching degree, the estimated cost of the simulated ridesharing trip corresponding to the candidate trip split point, and the estimated mileage of the corresponding simulated ride-hailing trip, a preset scoring function is used to determine the score corresponding to the candidate trip split point. By iterating through the candidate trip splitting point set obtained through the second module 420 and repeating the above steps, the score of each candidate trip splitting point in the candidate trip splitting point set can be obtained.

[0082] Continuing with this device embodiment, the fourth module 440 of the device can select the top preset number of candidate trip splitting points with the highest scores in the candidate trip splitting point set. The simulated ride-hailing trip and simulated carpooling trip of each candidate trip splitting point in the preset number of candidate trip splitting points determined by the third module 430 are taken as a carpooling connecting trip matching result, thus obtaining a preset number of carpooling connecting trip matching results corresponding to the carpooling order.

[0083] Through the device described in the above embodiment, the shared mobility service platform 100 can detect whether to trigger the splitting of a ride-sharing order based on its characteristic information. If it is determined that the order needs to be dispatched, the splitting process is triggered, and the order is coupled with real-time ride-sharing capacity to filter candidate splitting points. Based on the scores of the candidate splitting points, the connecting travel matching scheme is reconstructed. This ensures the necessity and accuracy of platform intervention and also increases the likelihood of subsequent order completion.

[0084] Optionally, the device for matching ridesharing trips further includes: The fifth module 450 is used to perform structured processing on the preset number of ridesharing trip matching results and then send them to the client for display.

[0085] In this optional device embodiment, the fifth module 450 of the device can perform structured processing on a preset number of ridesharing matching results corresponding to the passenger's ridesharing order obtained by the fourth module 440, and send the structured results to the passenger client 200 for exposure on the relevant page of the passenger client 200, allowing the passenger to browse and understand feasible alternatives that can meet their ridesharing requirements. If the passenger selects one or more of the ridesharing matching results after browsing, the results will be sent to the shared mobility service platform 100 through the passenger client 200.

[0086] Optionally, the device for matching ridesharing trips further includes: The sixth module 460 is used to create and publish corresponding order groups based on one or more ridesharing trip matching results confirmed by passengers. The order group includes a master order and ride-hailing sub-orders and ridesharing orders corresponding to each ridesharing trip matching result confirmed by the passenger.

[0087] In this optional device embodiment, the sixth module 460 of the device can create and publish a corresponding collaborative order group based on one or more ridesharing trip matching results selected and confirmed by the passenger through the passenger client 200, received through the fifth module 450. This order group is then sent to the passenger client 200 to be displayed to the passenger for browsing. The order group includes a master order and corresponding ride-hailing sub-orders and ridesharing orders for each ridesharing trip matching result confirmed by the passenger (i.e., one confirmed ridesharing trip matching result corresponds to one ride-hailing order and one ridesharing order). Passengers can browse relevant information about the order group on the relevant page of the passenger client 200, including but not limited to: order group ID, master order ID, list of sub-order IDs, and / or the estimated trip, estimated cost, and estimated travel time for each order. Module 460 can use order collaboration services to manage the lifecycle of the order group. Through order collaboration, the ridesharing order is broken down and packaged into one or more complete connecting travel service orders, including ridesharing sub-orders and ride-hailing sub-orders. When any sub-order in the order group is accepted, the order collaboration service automatically cancels the main order and other connecting travel service orders within it, ensuring that only the connecting travel service corresponding to that sub-order is executed. If a passenger accepts a set of ride-hailing and ridesharing orders from the order group, the passenger's client 200 will click to confirm, and the confirmation result will be sent to the shared mobility service platform 100.

[0088] Optionally, the device for matching ridesharing trips further includes: The seventh module 470 is used to monitor the status of the order group and process the relevant orders in the order group when its status changes.

[0089] In this optional device embodiment, the seventh module 470 of the device can monitor the status of published order groups in real time. When the status changes, the relevant orders in the order group are processed. For example, when the shared mobility service platform 100 receives a group of ride-hailing orders and carpooling orders from an order group selected by a passenger based on interactive operations on the passenger client 200, it can change the status of the ride-hailing order and carpooling order in the order group to "accepted" and publish the ride-hailing order to the ride-hailing driver's client and the carpooling order to the carpooling driver's client, respectively. When a ride-hailing order is accepted, its status is changed to "in execution"; when a carpooling order is accepted, its status is changed to "in execution".

[0090] Optionally, the device for matching ridesharing trips further includes: The eighth module 480 is used to determine the prepayment fee based on the ride-sharing order published by the passenger. When the ride-sharing sub-order corresponding to the ride-sharing trip matching result is completed, the corresponding ride-hailing trip fee is deducted in real time. When the ride-sharing order corresponding to the ride-sharing trip matching result is completed, the settlement is performed.

[0091] In this optional device embodiment, the eighth module 480 of the device can determine the prepayment based on the ride-sharing order posted by the passenger. After the passenger pays the prepayment, a ride-hailing order is posted to the ride-hailing driver's client 400, and a ride-sharing order is posted to the ride-sharing driver's client 300. When the ride-hailing order is accepted and completed by the ride-hailing driver, the corresponding ride-hailing fare is deducted from the prepayment paid by the passenger in real time. When the ride-sharing order is accepted and completed by the ride-sharing driver, settlement is made based on the actual ride-sharing fare and the remaining prepayment after deduction, with refunds for overpayments and additional payments for underpayments.

[0092] Through the above-described device embodiments and / or optional embodiments, the characteristics of ride-sharing orders can be perceived in real time, accurately identifying ride-sharing orders that meet preset triggering rules and triggering trip splitting at the optimal time. This can be actively coupled with real-time ride-sharing capacity supply to dynamically determine the optimal preset number of split-trip matching results (or connecting trip matching schemes), improving the accuracy of splitting eligible ride-sharing orders into connecting trips and enhancing user experience. Furthermore, based on the interactive operations of passengers and drivers on the client, the order groups corresponding to the connecting trip matching results selected by passengers are dynamically tracked and managed throughout the entire process through process collaboration. This enables dynamic monitoring, coordinated processing, and settlement of the status of the main order and sub-orders within the order group, ensuring seamless connection between cross-order and cross-trip connections, providing a guarantee for the smooth completion of ride-sharing orders and improving user experience. It can also improve the overall completion rate of ride-sharing orders on the shared mobility service platform 100.

[0093] In the above-described device embodiments and / or optional embodiments, any method steps that are not mentioned in the executable steps of the various components of the device are the same as those in the aforementioned related method embodiments and / or optional embodiments, and will not be repeated here.

[0094] According to another aspect of this application, a computer-readable medium is also provided, the computer-readable medium storing computer-readable instructions that can be executed by a processor to implement some or all of the foregoing method embodiments and / or optional embodiments.

[0095] It should be noted that the method embodiments and / or optional embodiments in this application do not strictly limit the order of execution of each step, as long as the method embodiments and / or optional embodiments can solve the defects existing in the prior art, achieve the inventive purpose of this application, and obtain beneficial effects. The method embodiments and / or optional embodiments in this application can be implemented in software and / or combinations of software and hardware. The software program involved in this application can be executed by a processor to implement the steps or functions of the above embodiments. Similarly, the software program of this application (including related data structures) can be stored in a computer-readable recording medium.

[0096] Furthermore, part or all of this application can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to this application through the operation of the computer. The program instructions invoking the methods of this application may be stored in a fixed or removable recording medium, and / or transmitted via data streams in broadcast or other signal carrying media, and / or stored in the working memory of a computer device operating according to the program instructions.

[0097] According to another aspect of this application, a device for matching ridesharing trips is also provided. The device includes: a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to run part or all of the methods and / or technical solutions of the foregoing embodiments.

[0098] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0099] In this application, when terms such as "upper," "lower," "left," "right," "front," "rear," "top," "bottom," "inner," "outer," "middle," "vertical," "horizontal," "lateral," and "longitudinal" are used, the indicated orientation and / or positional relationship is based on the orientation and / or positional relationship shown in the accompanying drawings. These terms are primarily for the purpose of better describing this application and its embodiments, and are not intended to limit the indicated device, element, or component to having a specific orientation, or to be constructed and operated in a specific orientation. Furthermore, some of the above terms, in addition to indicating orientation or positional relationship, can also be used to indicate other meanings; for example, the term "upper" can also be used in some cases to indicate a certain dependency or connection relationship. Those skilled in the art can understand the specific meaning of these terms in this application according to the specific circumstances. Furthermore, the terms "installation," "setup," "equipped with," "connection," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral structure; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection via an intermediate medium; and they can refer to an internal connection between two devices, components, or constituent parts. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0100] Furthermore, the terms "first," "second," etc., are primarily used to distinguish different devices, units, modules, elements, circuits, or components (which may be the same or different in specific type and construction), and are not intended to indicate or imply the relative importance, order, and / or quantity of the indicated devices, units, modules, elements, circuits, or components. Unless otherwise stated, "a plurality of" means two or more.

[0101] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in the apparatus claims may also be implemented by a single unit or device through software and / or hardware.

Claims

1. A method for matching ridesharing trips, characterized in that, The method includes: Based on the ride-sharing orders posted by passengers, the corresponding feature information is determined, wherein the feature information includes at least: departure time, departure point, destination point, cities passed through, estimated trip distance and estimated trip time, and based on the feature information, the corresponding dynamic matching features are determined; Based on the dynamic matching features, it is determined in real time whether the preset triggering rules are met. If they are met, the corresponding search space is determined based on the starting point or the destination point. Based on the departure time, estimated travel time, cities passed through and the destination point, the set of matching ridesharing driver plans is determined. Based on the set of ridesharing driver plans, the obtained key nodes of the road network and real-time supply and demand hotspots, several candidate trip splitting points are determined in the search space to form a set of candidate trip splitting points. For each candidate trip splitting point in the candidate trip splitting point set, a simulated ridesharing trip and a simulated ride-hailing trip are constructed based on the starting point and the destination point. The route matching degree between the simulated ridesharing trip and each planned ridesharing trip in the set of planned ridesharing trips is calculated. The number of planned ridesharing trips with a route matching degree that meets a preset threshold and the highest route matching degree are determined. The estimated mileage, estimated cost, and estimated time corresponding to the simulated ridesharing trip and the estimated mileage, cost, and estimated time corresponding to the simulated ride-hailing trip are calculated respectively to determine the estimated total mileage, total cost, and total time corresponding to the candidate trip splitting point. Based on the number, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip, a preset scoring function is used to determine the score corresponding to the candidate trip splitting point. The set of candidate trip splitting points is traversed to obtain the score of each candidate trip splitting point in the set of candidate trip splitting points. A preset number of candidate trip splitting points with the highest scores are selected from the candidate trip splitting point set. The simulated ride-hailing trip and simulated carpooling trip of each candidate trip splitting point are taken as a carpooling connecting trip matching result, and a preset number of carpooling connecting trip matching results corresponding to the carpooling order are obtained.

2. The method according to claim 1, characterized in that, The step of determining the corresponding dynamic matching features based on the feature information includes: Based on the aforementioned feature information, and according to the ride-sharing plans published by all ride-sharing drivers, the number of matching ride-sharing plans, the degree of route matching for each matching ride-sharing plan, and the real-time waiting time are determined.

3. The method according to claim 2, characterized in that, The preset triggering rule includes any one of the following: If the number of matched ridesharing trips and the estimated trip mileage meet the first preset condition, then the event is triggered; If the highest matching degree of the matched ridesharing plan and the real-time waiting time meet the second preset condition, then it is triggered; If the real-time waiting time meets the third preset condition, then it will be triggered.

4. The method according to claim 3, characterized in that, The preset triggering rules also include: If the historical connecting trip acceptance rate of the passenger who posted the ride-sharing order meets a preset threshold, the event is triggered, wherein the historical connecting trip acceptance rate is determined based on the passenger's historical behavior data.

5. The method according to claim 1, characterized in that, The determination of the corresponding search space based on the starting point or the destination point includes any one of the following: Centered on the starting point, and combining the estimated mileage and the first preset parameters, a first extended area and a second extended area are determined, and the relevant area between the first extended area and the second extended area is determined as the search space; Centered on the destination point, and combining the estimated mileage and the second preset parameters, a third extended region and a fourth extended region are determined, and the relevant area between the third extended region and the fourth extended region is determined as the search space.

6. The method according to claim 1, characterized in that, For a candidate trip split point, a simulated ridesharing trip and a simulated ride-hailing trip are constructed by combining the departure point and the destination point, including any one of the following: Construct simulated ride-hailing trips from the starting point to the candidate trip splitting point and simulated carpooling trips from the candidate trip splitting point to the destination point, respectively. Simulated ridesharing trips from the starting point to the candidate trip splitting point and simulated ride-hailing trips from the candidate trip splitting point to the destination point are constructed respectively.

7. The method according to claim 1, characterized in that, Based on the number of trips, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip, a preset scoring function is used to determine the score corresponding to the candidate trip split point, including: A preset scoring function is used to calculate the score corresponding to the candidate trip split point by weighting the quantity, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip. The quantity, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip are each assigned a preset initial value with their respective weights and are continuously optimized.

8. The method according to claim 7, characterized in that, The determination of the respective corresponding preset initial values ​​of the weights includes: The historical behavior data of the passengers is clustered to determine the user profile tag type of the passengers, and based on the user profile tag type of the passengers, the preset weight initial values ​​of the quantity, the highest route matching degree, the estimated cost corresponding to the simulated ride-sharing trip, and the simulated ride-hailing trip are adjusted.

9. The method according to claim 7, characterized in that, The continuous optimization includes: Obtain historical full-path data collected based on a preset event model, and iteratively optimize the weights.

10. The method according to claim 1, characterized in that, The method further includes: The preset number of ridesharing trip matching results are structured and then sent to the client for display.

11. The method according to claim 10, characterized in that, The method further includes: Based on one or more ridesharing trip matching results confirmed by the passenger, a corresponding order group is created and published. The order group includes a master order and ride-hailing sub-orders and ridesharing orders corresponding to each ridesharing trip matching result confirmed by the passenger.

12. The method according to claim 11, characterized in that, The method further includes: Monitor the status of the order group and process the relevant orders in the order group when the status changes.

13. The method according to claim 11, characterized in that, The method further includes: Based on the ridesharing order posted by the passenger, the prepayment fee is determined. When the ridesharing trip matching result corresponding to the online car-hailing order is completed, the corresponding online car-hailing trip fee is deducted in real time. When the ridesharing trip matching result corresponding to the ridesharing order is completed, the settlement is performed.

14. A device for matching ridesharing trips, characterized in that, The device includes: The first module is used to determine the corresponding feature information based on the ride-sharing order posted by the passenger. The feature information includes at least: departure time, departure point, destination point, cities passed through, estimated trip distance and estimated trip time, and determines the corresponding dynamic matching features based on the feature information. The second module is used to determine in real time whether the preset triggering rules are met based on the dynamic matching features. If they are met, the corresponding search space is determined based on the starting point or the destination point. Based on the departure time, estimated travel time, cities passed through and the destination point, the matching set of ridesharing driver plans is determined. Based on the set of ridesharing driver plans, the obtained key nodes of the road network and real-time supply and demand hotspots, several candidate trip splitting points are determined in the search space to form a set of candidate trip splitting points. The third module is used to construct a simulated ridesharing trip and a simulated ride-hailing trip for each candidate trip splitting point in the candidate trip splitting point set, combining the starting point and the destination point. It calculates the route matching degree between the simulated ridesharing trip and each planned ridesharing trip in the set of planned ridesharing trips, determines the number of planned ridesharing trips with a route matching degree satisfying a preset threshold and the highest route matching degree, and calculates the estimated mileage, estimated cost, and estimated time corresponding to the simulated ridesharing trip and the simulated ride-hailing trip, respectively, to determine the estimated total mileage, total cost, and total time corresponding to the candidate trip splitting point. Based on the number, the highest route matching degree, the estimated cost of the simulated ridesharing trip, and the estimated mileage of the simulated ride-hailing trip, a preset scoring function is used to determine the score corresponding to the candidate trip splitting point. The module iterates through the candidate trip splitting point set to obtain the score of each candidate trip splitting point in the set. The fourth module is used to select a preset number of candidate trip splitting points with the highest scores in the candidate trip splitting point set, and to take the simulated ride-hailing trip and simulated carpooling trip of each candidate trip splitting point as a carpooling connecting trip matching result, so as to obtain a preset number of carpooling connecting trip matching results corresponding to the carpooling order.

15. The apparatus according to claim 14, characterized in that, The device further includes: The fifth module is used to perform structured processing on the preset number of ridesharing trip matching results and then send them to the client for display.

16. The apparatus according to claim 15, characterized in that, The device further includes: The sixth module is used to create and publish corresponding order groups based on one or more ridesharing trip matching results confirmed by passengers. The order group includes a master order and ride-hailing sub-orders and ridesharing orders corresponding to each ridesharing trip matching result confirmed by the passenger.

17. The apparatus according to claim 16, characterized in that, The device further includes: The seventh module is used to monitor the status of the order group and process the relevant orders in the order group when the status changes.

18. The apparatus according to claim 16, characterized in that, The device further includes: The eighth module is used to determine the prepayment fee based on the ridesharing order published by the passenger. When the ridesharing sub-order corresponding to the ridesharing trip matching result is completed, the corresponding ridesharing trip fee is deducted in real time. When the ridesharing order corresponding to the ridesharing trip matching result is completed, the settlement is performed.

19. A computer-readable medium, characterized in that, It stores computer-readable instructions that are executed by a processor to implement the method as described in any one of claims 1 to 13.

20. A device for matching ridesharing trips, characterized in that, The device includes: One or more processors; and A memory storing computer-readable instructions, which, when executed, cause the processor to perform the method as described in any one of claims 1 to 13.

Images