Online car-hailing reservation method and system

Abstract

The invention relates to an online car-hailing reservation method and system. The method is applied to a user side and a server side which are in communication connection with each other. The server determines a plurality of different types of candidate online hailed cars corresponding to the travel information based on the travel information, and obtains the predicted cost and predicted arrival time of each candidate online hailed car through a cost and time prediction model when detecting that the candidate online hailed cars are hitchhiking cars and different types of real-time cars; wherein the cost and time prediction model is a multi-task learning model; and the user side displays the predicted cost and the predicted arrival time of the various candidate online hailed cars for the user to select. According to the method and the device, the problem that cross-interface data are inconsistent when different interfaces jump in the prior art is solved.

Description

Technical Field

[0001] This application relates to the technical field of order creation, and particularly to a method and system for online car-hailing reservation. Background Art

[0002] In the existing online car-hailing travel service, when a user places an order, they use different travel services on the home page according to the current actual situation, fill in the correct destination, and select a suitable category on the valuation page to place an order to ensure a smooth journey. The current method used in the market is to first select a travel mode on the home page and then estimate the price after entering the destination. However, when the user has doubts about the current price and wants to compare other travel modes, they need to return to the home page and enter the address again for price estimation. This process requires multiple clicks and jumps, increasing the complexity of the user's operation. Therefore, there is a need to provide a method and system for online car-hailing reservation. Summary of the Invention

[0003] In view of the above-mentioned disadvantages of the prior art, the purpose of this application is to provide a method and system for online car-hailing reservation, which improves the problem of inconsistent cross-interface data when jumping between different interfaces in the prior art.

[0004] To achieve the above object and other related objects, this application provides a method for online car-hailing reservation, which is applied to a user terminal and a service terminal that are communicatively connected to each other, and includes: the user terminal obtains the travel information of the user; the service terminal determines multiple different types of candidate online car-hailing corresponding thereto based on the travel information, and when detecting that the candidate online car-hailing is a shared ride and real-time car-hailing of different categories, obtains the estimated cost and estimated arrival time of each candidate online car-hailing through a cost and time prediction model; wherein, the cost and time prediction model is a multi-task learning model; the user terminal displays the estimated cost and estimated arrival time of various candidate online car-hailing for the user to select.

[0005] In an embodiment of this application, the service terminal determines multiple different types of candidate online car-hailing corresponding thereto based on the travel information, and when detecting that the candidate online car-hailing is a shared ride and real-time car-hailing of multiple different categories, obtains the estimated cost and estimated arrival time of each type of candidate online car-hailing through a cost and time prediction model, including: calculating the shared ride matching degree between the travel information and each currently running shared ride, and determining whether there is a shared ride that matches the travel information based on the shared ride matching degree; if there is no shared ride, determining at least one type of real-time car-hailing corresponding thereto based on the travel information, and determining the estimated cost and estimated arrival time of the real-time car-hailing; if there is a shared ride, performing the following process: determining different categories of real-time car-hailing corresponding thereto based on the travel information; using the shared ride that matches the travel information and various categories of real-time car-hailing as different types of candidate online car-hailing; obtaining the estimated cost and estimated arrival time of each type of candidate online car-hailing through a cost and time prediction model.

[0006] In one embodiment of the present application, calculating the matching degree between the trip information and each currently operating carpool, and determining whether there is a carpool matching the trip information based on the carpool matching degree includes: retrieving the trips of each currently operating carpool from the driving database, and screening out the carpool whose time difference between the arrival time at the starting point of the trip information and the reservation time is less than a preset time threshold; for each screened carpool: determining the matching degree between the carpool's trip and the driving route of the trip information; determining whether the matching degree is greater than a preset matching degree threshold, and when it is greater than the matching degree threshold, regarding this carpool as the carpool matching the trip information.

[0007] In one embodiment of the present application, determining different categories of real-time vehicle usage corresponding to the trip information includes: determining the trip distance based on the starting point and ending point of the trip information of the trip; determining whether the trip distance is less than a preset distance threshold: if so, determining the carpool matching degree between the trip information and each currently operating carpool, and determining whether there is a carpool matching the trip information based on the carpool matching degree; if there is, regarding the carpool, the dedicated car, and the express car as the real-time vehicle usage corresponding to the trip information; if not, regarding the dedicated car and the express car as the real-time vehicle usage corresponding to the trip information; otherwise, regarding the taxi, the dedicated car, and the express car as the real-time vehicle usage corresponding to the trip information.

[0008] In one embodiment of the present application, when the candidate online car-hailing vehicle is a carpool or a shared ride, obtaining the estimated cost and estimated arrival time of the candidate online car-hailing vehicle through the cost and time prediction model includes: determining its current location and the number of passengers to be carried based on the identifier of the online car-hailing vehicle; determining the driving route based on the current location of the candidate online car-hailing vehicle and the trip information; wherein, the trip route is the estimated driving route for the candidate online car-hailing vehicle to drive from the current location to the ending point of the trip information of the trip; inputting the driving route, the current location of the candidate online car-hailing vehicle, and the trip information into the time prediction module of the cost and time prediction model to obtain the estimated arrival time; inputting the driving route, the number of passengers to be carried, the estimated arrival time, and the trip information of the candidate online car-hailing vehicle into the cost prediction module of the cost and time prediction model to obtain the estimated cost.

[0009] In an embodiment of the present application, when the candidate online car-hailing vehicle is a luxury car, an economy car or a taxi, for each type of candidate online car-hailing vehicle, the estimated cost and the estimated arrival time of the candidate online car-hailing vehicle are obtained through a cost and time prediction model, including: determining a corresponding order dispatch area based on the starting point and the ending point of the trip in the trip information; determining the number of available drivers and the total number of current orders to be received corresponding to this type of candidate online car-hailing vehicle from the order dispatch area; calculating the current supply-demand ratio of this type of candidate online car-hailing vehicle based on the number of available drivers and the total number of orders to be received; inputting the supply-demand ratio and the trip information into the cost and time prediction model to predict the estimated cost and the estimated arrival time of this type of candidate online car-hailing vehicle.

[0010] In an embodiment of the present application, the multi-task learning model is a hard parameter sharing model.

[0011] In an embodiment of the present application, the hard parameter sharing model is an MT-DNN model.

[0012] In an embodiment of the present application, after the user terminal displays the estimated cost and the estimated arrival time of various candidate online car-hailing vehicles for the user to select, it further includes: the user terminal jumps to the online car-hailing vehicle interface of the corresponding type in response to the user's selection of the candidate online car-hailing vehicle.

[0013] In an embodiment of the present application, an online car-hailing reservation system is further provided. The system includes a user terminal and a server that are communicatively connected to each other. The user terminal includes: a first communication module for sending the user's trip information to the server and receiving the estimated cost and the estimated arrival time of various candidate online car-hailing vehicles sent by the server; a data acquisition module for acquiring the user's trip information; a display module for displaying the estimated cost and the estimated arrival time of various candidate online car-hailing vehicles for the user to select; the server includes: a second communication module for receiving the trip information sent by the user terminal and sending the estimated cost and the estimated arrival time of various candidate online car-hailing vehicles to the user terminal; a cost and time determination module for determining various different types of candidate online car-hailing vehicles corresponding thereto based on the trip information, and when it is detected that the candidate online car-hailing vehicle is a shared car and real-time car-hailing of different categories, obtaining the estimated cost and the estimated arrival time of each candidate online car-hailing vehicle through a cost and time prediction model; wherein, the cost and time prediction model is a multi-task learning model.

[0014] As described above, a car-hailing reservation method and system of the present application have the following beneficial effects: The server filters eligible shared cars and real-time cars of different categories based on the trip information provided by the user terminal, and uses a multi-task learning model to simultaneously predict the estimated cost and estimated arrival time of these candidate car-hailing services. By sharing underlying features, this model can improve calculation efficiency and reduce redundant calculations. The user terminal displays the estimated cost and estimated arrival time of all candidate car-hailing services, enabling the user to intuitively compare different travel modes, effectively improving the problem of inconsistent cross-interface data when jumping between different interfaces in the prior art. This allows the user to make the best choice among cost, waiting time, and convenience, thereby improving travel decision-making efficiency and simultaneously increasing the order matching rate and overall operation ability of the platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 FIG. is a schematic diagram of the interaction scenario between the user terminal and the server provided by an embodiment of the present application;

[0016] Figure 2 FIG. is a flowchart of a car-hailing reservation method provided by an embodiment of the present application;

[0017] Figure 3a FIG. is a schematic diagram of the home page interface of the user terminal provided by an embodiment of the present application;

[0018] Figure 3b FIG. is a schematic diagram of the shared car interface of the user terminal provided by an embodiment of the present application;

[0019] Figure 3c FIG. is a schematic diagram of the address selection interface provided by an embodiment of the present application;

[0020] Figure 3d FIG. is a schematic diagram of the price quotation interface of various vehicle types provided by an embodiment of the present application;

[0021] Figure 3e FIG. is a schematic diagram of the departure time selection interface of the shared car provided by an embodiment of the present application;

[0022] Figure 3f FIG. is a schematic diagram of the highway toll payment selection interface of the shared car provided by an embodiment of the present application;

[0023] Figure 3g FIG. is a schematic diagram of the shared car order confirmation interface provided by an embodiment of the present application;

[0024] Figure 4 FIG. shows a structural block diagram of a car-hailing reservation system provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] The following specific examples illustrate the implementation manners of the present application. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0026] It should be noted that the diagrams provided in the following embodiments only schematically illustrate the basic concept of the present application. Therefore, only the components related to the present application are shown in the diagrams, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and proportion of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0027] In the following description, a large number of details are explored to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0028] The existing process from placing an order to obtaining an estimate in online car-hailing applications usually has the following problems: First, the interface operation is cumbersome. Under normal circumstances, users need to select the service type on the home page. For example, after selecting the hitchhiking category and wanting to compare the real-time car rental price, they need to jump back to the home page, re-select the location, and then estimate the price on the real-time car rental interface. This process requires multiple clicks and jumps, increasing the operation complexity of users and making it difficult to intuitively view the prices of other categories. Second, the response delay is serious. Each page jump will cause a loading time, especially in a poor network environment, which will seriously affect the user experience. Third, the information is not coherent. When users switch between different pages, they may forget their previous selections, resulting in repeated operations and increasing the difficulty of use for users. Finally, there is also the problem of scattered functions. Since the selection of different categories and the price estimation functions are scattered on different pages, users need to switch back and forth between multiple pages, which is not conducive to quickly obtaining the required information.

[0029] To address the above problems, the present application provides a method for booking online car-hailing services. The server filters eligible shared rides and real-time car-hailing services of different categories based on the trip information provided by the user terminal, and uses a multi-task learning model to simultaneously predict the estimated fares and estimated arrival times of these candidate online car-hailing services. By sharing underlying features, this model can improve computational efficiency and reduce redundant calculations. The user terminal displays the estimated fares and estimated arrival times of all candidate online car-hailing services, enabling the user to intuitively compare different travel modes, effectively improving the problem of inconsistent cross-interface data during interface jumps in the prior art. This allows the user to make the best choice among cost, waiting time, and convenience, thereby enhancing the travel decision-making efficiency and simultaneously increasing the order matching rate and overall operation ability of the platform.

[0030] As Figure 1 shown, the present application is applied to online car-hailing scenarios such as shared rides, carpooling, express cars, and taxis to achieve real-time interaction between the user terminal and the server. Among them, the user terminal refers to the front-end interface where the user interacts with the server in the online car-hailing application, and can be run through mobile applications, web pages, mini-programs, etc. The server refers to the back-end system that processes trip data according to user requests to determine the estimated fares and estimated arrival times of various online car-hailing services and returns the calculation results. When the user needs to use a car, after entering the starting point and ending point on the user terminal, the user terminal obtains the geographical coordinates through GPS or map API and sends a trip request to the server. The server queries available drivers based on the fare and time prediction model, and when detecting the existence of shared rides and other types of real-time car-hailing services, calculates the estimated arrival times and estimated fares of various online car-hailing services and returns the optimal matching result to the user terminal. The user terminal will present the estimated fares and estimated arrival times of shared rides and other types of real-time car-hailing services on the same interface, enabling the user to understand the quotes of shared rides and other types of real-time online car-hailing services on the same interface for selection.

[0031] Please refer to Figure 2 , the method for booking online car-hailing services includes the following steps:

[0032] S1. The user terminal obtains the trip information of the user.

[0033] When a user needs to reserve an online car-hailing service, the user selects and enters trip information such as the starting point, ending point, and departure time of the trip independently on the user side. The user side will automatically locate the user's current location through methods such as GPS or map API and visually display it on the map interface to ensure that the user can clearly see their starting point. To facilitate the user to quickly select frequently used addresses and avoid repeated input, the user side also provides functions such as address search, favorite locations, and historical trips. Among them, address search means intelligently matching surrounding locations according to the keywords entered by the user, favorite locations means presenting the locations previously favorited by the user according to the user's trigger operation, and historical trips means that the user can directly select the most recently used destination to reduce the manual input steps.

[0034] S2. The server determines multiple different types of candidate online car-hailing services corresponding to the trip information, and when it detects that the candidate online car-hailing services are shared rides and real-time car-hailing services of different categories, it obtains the estimated cost and estimated arrival time of each candidate online car-hailing service through a cost and time prediction model; among them, the cost and time prediction model is a multi-task learning model.

[0035] The user side sends the user's trip information to the server. The server queries the current availability of shared rides and other types of real-time car-hailing services according to the trip information and screens out eligible candidate online car-hailing services. For shared rides and real-time car-hailing services of different categories such as express cars and taxis, the estimated cost and estimated arrival time of this type of candidate online car-hailing service are determined through a cost and time prediction model, and the results are returned to the user side. Among them, the estimated arrival time refers to the time when the candidate online car-hailing service is expected to arrive at the starting point of the trip. Optionally, to improve the calculation efficiency, the multi-task learning model is a hard parameter sharing model, which includes but is not limited to Cross-Stitch Network, Shared Bottom Model, and MT-DNN model, etc. Optionally, to achieve deep shared feature extraction, improve the model calculation efficiency, and reduce overfitting between tasks at the same time, the hard parameter sharing model is the MT-DNN model.

[0036] Specifically, step S2 includes S21 to S23 (not shown in the figure):

[0037] S21. Calculate the matching degree of the trip information with each currently running shared ride, and determine whether there is a shared ride that matches the trip information based on the matching degree of the shared ride.

[0038] After the server receives the user's trip information sent by the client, it compares all currently running shared ride trips, and determines the matching degree of the user's trip with each shared ride trip through factors such as route overlap degree, time proximity degree, and driving direction consistency to determine whether there is a shared ride that matches the user.

[0039] Specifically, step S21 includes the following process: first, the itineraries of each current hitchhiking car are retrieved from the driving database, and the hitchhiking cars whose arrival time at the starting point of the itinerary information and the reservation time differ by less than a preset time threshold are screened out. The server extracts the itinerary data of all current hitchhiking cars from the driving database, and calculates the estimated arrival time of each hitchhiking car at the starting point of the itinerary set by the user. The difference between the arrival time and the departure time entered by the user in the itinerary information is compared, and the hitchhiking cars whose time difference is less than the preset time threshold are screened out to ensure that the hitchhiking car can pick up and drop off users within a reasonable time range, thereby improving the matching success rate.

[0040] Secondly, for each selected ride-sharing car, the following process is performed: first, the matching degree between the ride-sharing car's itinerary and the driving path of the itinerary information is calculated. Then, it is determined whether the matching degree is greater than a preset matching degree threshold, and if it is greater than the matching degree threshold, the ride-sharing car is regarded as a ride-sharing car that matches the itinerary information.

[0041] The server performs the following process for each of the aforementioned selected ride-sharing cars: first, the established driving route of the ride-sharing driver is analyzed based on the map data, and the similarity between the route and the route from the user's starting point to the destination is calculated. If the overlapping mileage of the two is greater than or equal to the preset overlapping threshold and the driving directions are the same, it means that the two have a high degree of match. On the contrary, if the overlapping mileage of the two is less than the preset overlapping threshold, or the driving directions are opposite, it means that the matching degree is low. After calculating the matching degree of all the selected ride-sharing cars, the ride-sharing car with a matching degree greater than the matching degree threshold is selected as the ride-sharing car that matches the itinerary information, and finally a list of matching rides is formed for the user to choose.

[0042] S22. If there is no ride-sharing, determine at least one type of real-time car use corresponding to the ride based on the trip information, and determine the estimated cost and estimated arrival time of the real-time car use.

[0043] If a qualified rideshare is not matched, the server will automatically switch to the real-time car-use mode. At this time, the server will screen out at least one type of real-time online ride-hailing vehicles, such as available express trains and taxis, based on the user's itinerary information such as the starting point, end point, and departure time of the trip. The estimated cost and estimated arrival time of the online ride-hailing vehicle are determined through a preset pricing method, and the estimated cost and estimated arrival time are sent to the user for the user to choose. It can be understood that this application is for when there is a matching rideshare, the rideshare-related information and other real-time car-use information are displayed together on the same interface. Therefore, when a rideshare is not matched, the process of switching to real-time car-use belongs to the conventional online car-hailing scheduling strategy, which will not be described in detail here.

[0044] S23. If there is a rideshare, the following process is performed:

[0045] S231. Determine the corresponding real-time carpooling of different categories based on the travel information.

[0046] Based on the user's travel information, the server determines the currently available real-time carpooling categories. Among them, the real-time carpooling categories include but are not limited to online car-hailing services of different categories such as express cars, taxis, private cars, and shared rides. Combining the driving routes of these online car-hailing services, the server filters out the real-time carpooling of different categories that match the user's travel.

[0047] Specifically, S231 includes the following process:

[0048] First, determine the travel distance based on the starting point and ending point of the travel information. The server determines the optimal route from the starting point to the ending point of the travel by calling a preset path planning model according to the starting point and ending point of the travel provided by the user, and takes the length of the optimal route as the travel distance. Among them, the optimal route refers to the best driving route selected from multiple feasible routes according to a preset optimization goal (such as the shortest driving time, the shortest distance, etc.). It can be understood that the path planning model can be any model such as a deep neural network model or a reinforcement learning model, as long as it can achieve the best path planning, and it is not limited here.

[0049] Then the server determines whether the travel distance is less than a preset distance threshold. If the user's travel distance is less than the preset distance threshold, it means that the travel is short, and the carpooling matching process can be carried out. By determining the carpooling matching degree between the travel information and each currently running carpooling, and based on the carpooling matching degree, it is determined whether there is a carpooling that matches the travel information. Specifically, the server will query the currently running carpooling and filter out the carpooling orders whose estimated arrival time at the user-specified departure time is within the preset reservation time range. Calculate the carpooling matching degree between the user's travel and the carpooling driver's travel according to the best driving route. If the carpooling matching degree is greater than or equal to the preset matching degree threshold, it is determined that there is a carpooling plan, and the carpooling, private car, and express car can be used as the real-time carpooling corresponding to the travel information. On the contrary, if the carpooling matching degree is less than the preset matching degree threshold, it is determined that the carpooling is not feasible. At this time, the private car and express car will be used as the real-time carpooling corresponding to the travel information. On the other hand, if the travel distance is greater than or equal to the preset distance threshold, it means that the travel is far and carpooling is not available. At this time, the available taxis, private cars, and express cars will be used as the real-time carpooling corresponding to the travel information.

[0050] S232. Use the carpooling that matches the travel information and the real-time carpooling of various categories as different types of candidate online car-hailing services.

[0051] After the server calculates the matching plan for the user's itinerary, all eligible carpooling and real-time car-hailing of different categories are used as candidate online car-hailing services. Among them, carpooling matching is screened based on factors such as itinerary overlap, direction consistency, and time matching, while real-time car-hailing is screened based on factors such as the number of available vehicles and supply-demand situations. It can be understood that in the following, this application only elaborates on the price estimation and arrival time prediction for the selected candidate online car-hailing services. How to screen out the eligible candidate online car-hailing services from several available real-time car-hailing services is an existing method and will not be elaborated here.

[0052] S233. Obtain the estimated fees and estimated arrival times of various candidate online car-hailing services through the fee and time prediction model.

[0053] Specifically, when the candidate online car-hailing service is a carpool or shared ride, S233 includes the following processes:

[0054] First, determine the current location of the online car-hailing service and the number of passengers to be carried based on the identifier of the online car-hailing service. For candidate online car-hailing services such as carpooling or shared rides, the server determines the real-time location of the vehicle through the representation of the candidate online car-hailing service, and queries the ongoing orders of the vehicle by querying the order database, and then queries the total number of passengers that the vehicle needs to carry to complete the current order.

[0055] Secondly, determine the driving route based on the current location of the candidate online car-hailing service and the itinerary information; where the itinerary route is the estimated driving route for the candidate online car-hailing service to drive from the current location to the itinerary end point of the itinerary information. The server can generate the optimal driving route for the vehicle to start from the current location, pass through the user's starting point in sequence, and finally reach the itinerary end point based on the current location of the vehicle, itinerary information such as the user's itinerary start point and itinerary end point, and information such as real-time traffic conditions and road traffic conditions through the path planning model, and use it as the driving route.

[0056] Then, input the driving route, current location, and the itinerary information of the candidate online car-hailing service into the time prediction module of the fee and time prediction model to obtain the estimated arrival time. The estimated arrival time includes the estimated arrival time at the starting point and the estimated arrival time at the end point. The server inputs the driving route, current location, and the user's itinerary information into the time prediction module. This module uses an LSTM model or an XGBoost regression model, and combines a dynamic path optimization algorithm to analyze the driving route of the vehicle, calculates the estimated time for the vehicle to reach the user's starting point from the current location, further predicts the driving time from the starting point to the end point, and returns the estimated arrival time to the user terminal.

[0057] Finally, the driving route of the candidate online car-hailing vehicle, the number of passengers to be carried, the estimated arrival time, and the trip information are input into the fare prediction module of the fare and time prediction model to obtain the estimated fare. The server inputs the driving route, the number of passengers to be carried, the estimated arrival time, and the user's trip information into the fare prediction module of the fare and time prediction model, calculates the basic fare according to the driving route, including the starting price, mileage fee, time fee, etc. And in combination with the estimated arrival time and the real-time supply and demand situation, the pricing is dynamically adjusted. For example, the premium coefficient is increased during peak hours, or a discount is provided during low-demand hours. In addition, since the online car-hailing vehicle is a shared ride or a carpool, the fare for each passenger is also calculated according to the principle of sharing the passenger trip, and the final estimated fare of the online car-hailing vehicle is generated and returned to the user terminal. It can be understood that the fare prediction module can be a neural network model, a reinforcement learning model, an XGBoost regression model, etc., as long as it can achieve price prediction, and it is not limited here.

[0058] When the candidate online car-hailing vehicle is a luxury car, an economy car, or a taxi, for each type of candidate online car-hailing vehicle, S233 includes the following process:

[0059] First, determine the corresponding order assignment area based on the starting point and the ending point of the trip in the trip information. The server calls the map partitioning model, classifies the starting point of the user's trip into the corresponding order assignment grid, and combines the ending point of the trip to judge the cross-region situation of the trip. Among them, the division of the order assignment area is based on factors such as the urban road network structure, traffic flow, and historical order heat data, so as to ensure that the order can be matched with the most suitable driver resources.

[0060] Secondly, determine the number of idle drivers corresponding to this type of candidate online car-hailing vehicle and the total number of current pending orders in the order assignment area. The server queries the real-time vehicle scheduling database to obtain the vehicle information of the vehicles in the idle state in the order assignment area. The vehicle information includes, but is not limited to, information such as the current location of the vehicle and the vehicle type. The vehicle type includes, but is not limited to, taxis, economy cars, luxury cars, etc. At the same time, count the total number of pending orders in the order assignment area for subsequent analysis of the current supply and demand situation.

[0061] Then, based on the number of idle drivers and the total number of pending orders, calculate the current supply-demand ratio of this type of candidate online car-hailing vehicle. By calculating the ratio of the number of idle drivers to the total number of pending orders, the current supply-demand ratio of this type of candidate online car-hailing vehicle is obtained. The supply-demand ratio is used to represent the abundance degree of the current available driver resources relative to the user's car-hailing demand. The higher the supply-demand ratio, the more sufficient the idle drivers, and the greater the order acceptance success rate at this time.

[0062] Finally, input the supply-demand ratio and the trip information into the cost and time prediction model to predict the estimated cost and the estimated arrival time of this type of candidate online car-hailing service. The server inputs the supply-demand ratio and the user's trip information into the cost and time prediction model, calculates the initial cost based on the preset basic rates (such as the starting price, mileage fee, time fee, etc.), and dynamically adjusts the price in combination with the supply-demand ratio to obtain the estimated cost. When the supply-demand ratio > 1 (sufficient supply), the price is reduced based on the preset first premium coefficient on the basis of the initial cost; when the supply-demand ratio < 1 (strong demand), the price is increased based on the preset second premium coefficient on the basis of the initial cost to adjust the market supply and demand balance. It can be understood that the specific values of the first premium coefficient and the second premium coefficient can be adaptively set by those skilled in the art according to actual needs and are not limited herein. Further, the estimated arrival time of the driver can be predicted by using the cost and time prediction model through the trip information of this type of candidate online car-hailing service and the user's trip information, where the estimated arrival time includes the estimated arrival time at the trip starting point and the estimated arrival time at the trip end point.

[0063] S3. The user terminal displays the estimated costs and the estimated arrival times of various types of candidate online car-hailing services for the user to select. After receiving the estimated costs and the estimated arrival times of the hitchhiking cars and other types of real-time car-hailing services, the user terminal presents them to the user on the same interface for the user to select.

[0064] Further, after S3, it further includes: after the user terminal displays the estimated costs and the estimated arrival times of various types of candidate online car-hailing services for the user to select, it further includes: the user terminal jumps to the corresponding type of online car-hailing interface in response to the user's selection of a candidate online car-hailing service. Specifically, in response to the user's selection of a hitchhiking car, the user terminal jumps to the hitchhiking car reservation interface for the user to reserve a hitchhiking car, and in response to the user's selection of real-time car-hailing services such as express cars and taxis, it jumps to the corresponding online car-hailing payment interface for the user to pay the fee.

[0065] As Figure 3a shown, after the user opens the user terminal, the home page will present two service types, namely car-hailing and hitchhiking cars. On the one hand, the user can directly select the hitchhiking car service type and enter the Figure 3b shown interface. On the hitchhiking car interface, the user can click "Get on at ***" to locate the trip starting point, and click "Where are you going?" to select the specific trip end point through the Figure 3c shown option box. After the selection is completed, it will jump to the Figure 3e shown interface for the user to determine the specific departure time. After the user confirms the specific departure time, if it is detected that the trip will pass through the highway, it will jump to the Figure 3f shown interface for the user to select the payment method for the highway toll. After confirming the highway toll payment method, it jumps to the Figure 3gThe payment offer interface shown is provided for the user to confirm the payment of the final fee. During any of the above steps, if the user wants to compare the prices of hitchhiking cars and other real-time car-hailing services, they can directly click the return button on the corresponding interface and return to Figure 3b the hitchhiking car interface shown. By selecting the service type of car-hailing, the user can enter the offer interfaces of real-time car-hailing and hitchhiking cars. In addition, on the other hand, the user can also directly select the car-hailing service type on the Figure 3a home page shown. Clicking "Get in the car at ***" can locate the starting point of the user's trip, and clicking "Where are you going" can select the specific ending point of the trip through the option box shown in Figure 3c . After determining the trip information, it will jump to the Figure 3d offer interface shown. This interface will display the prices of hitchhiking cars and various different real-time car-hailing services (such as express cars, super express cars, etc.) so that users can more intuitively understand the prices of different car-hailing services. On this offer interface, the user can directly click on the hitchhiking car option to enter the Figure 3e interface shown and execute the aforementioned hitchhiking car payment process.

[0066] Please refer to Figure 4 . The hitchhiking car reservation system includes: a user terminal and a server that are communicatively connected to each other. The user terminal includes: a first communication module 11 for sending the user's trip information to the server and receiving the estimated fees and estimated arrival times of various candidate car-hailing services sent by the server. A data acquisition module 12 for acquiring the user's trip information. A display module 13 for displaying the estimated fees and estimated arrival times of various candidate car-hailing services for the user to select. The server includes: a second communication module 14 for receiving the trip information sent by the user terminal and sending the estimated fees and estimated arrival times of various candidate car-hailing services to the user terminal. A fee and time determination module 15 for determining multiple different types of candidate car-hailing services corresponding to the trip information based on the trip information, and when it is detected that the candidate car-hailing service is a hitchhiking car and real-time car-hailing services of different categories, obtaining the estimated fees and estimated arrival times of each candidate car-hailing service through a fee and time prediction model; wherein, the fee and time prediction model is a multi-task learning model.

[0067] For the specific limitations of the hitchhiking car reservation system, reference can be made to the limitations on the car-hailing reservation method in the above text, which will not be elaborated here. Each module in the above hitchhiking car reservation system can be implemented in whole or in part through software, hardware, and their combinations. The above modules can be embedded in the processor of the computer device in hardware format or independent of it, or stored in the memory of the computer device in software format so that the processor can call the corresponding operations of the above modules.

[0068] It should be noted that, in order to highlight the innovative part of this application, modules that are not closely related to solving the technical problems proposed in this application are not introduced in this embodiment. However, this does not mean that there are no other modules in this embodiment.

[0069] In summary, for a car-hailing reservation method and system disclosed in this application, by adding a switch for the hitchhiking entrance above the order placement area on the home page, users can quickly select to use the real-time car-hailing or hitchhiking service type; adding the hitchhiking category in the category on the valuation page, users can quickly compare different categories, effectively reducing the complexity of operations. Users can complete category selection and price estimation without multiple clicks and jumps. The simple interface design and convenient operation method improve the user experience and comfort. In addition, it can also alleviate the problem of excessive loading time caused by each page jump, thereby avoiding the problem that response latency affects the user experience. At the same time, it solves the situation of inconsistent information. When users switch between different pages, they will not forget their previous selections and avoid repeated operations. Finally, it reduces the dispersion of functions, enabling users to better quickly obtain information in category selection and price estimation, reducing the inconvenience caused by functions being scattered on different pages, and improving the efficiency of users to obtain the required information. Therefore, this application effectively overcomes various shortcomings in the prior art and has high industrial utilization value.

[0070] The above embodiments are only illustrative of the principles and effects of this application, and are not used to limit this application. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed in this application should still be covered by the claims of this application.

Claims

1. A method for booking online car-hailing, characterized in that, Applied to a client and a server that communicate with each other, the method includes: The client obtains the travel information of the user. The server determines multiple different types of candidate online car-hailing services corresponding to the travel information, and when it detects that the candidate online car-hailing services are shared rides and real-time car-hailing services of different categories, it obtains the estimated cost and estimated arrival time of each candidate online car-hailing service through a cost and time prediction model; wherein, the cost and time prediction model is a multi-task learning model. The client displays the estimated cost and estimated arrival time of various candidate online car-hailing services for the user to select.

2. The online car-hailing reservation method according to claim 1, wherein The server determines multiple different types of candidate online car-hailing services corresponding to the travel information, and when it detects that the candidate online car-hailing services are shared rides and real-time car-hailing services of multiple different categories, it obtains the estimated cost and estimated arrival time of various candidate online car-hailing services through a cost and time prediction model, including: Calculating the matching degree of the travel information with each currently running shared ride, and determining whether there is a shared ride that matches the travel information based on the matching degree of the shared ride. If there is no shared ride, then determining at least one category of real-time car-hailing service corresponding to the travel information, and determining the estimated cost and estimated arrival time of the real-time car-hailing service. If there is a shared ride, then perform the following process: Determining different categories of real-time car-hailing services corresponding to the travel information. Taking the shared ride that matches the travel information and various categories of real-time car-hailing services as different types of candidate online car-hailing services. Obtaining the estimated cost and estimated arrival time of various candidate online car-hailing services through a cost and time prediction model.

3. The online car-hailing reservation method according to claim 2, wherein The calculating the matching degree of the travel information with each currently running shared ride, and determining whether there is a shared ride that matches the travel information based on the matching degree of the shared ride, includes: Retrieving the trips of each currently running shared ride from the driving database, and screening out the shared rides whose time difference between the arrival time at the starting point of the travel information and the reservation time is less than a preset time threshold. For each screened shared ride: Determining the matching degree between the trip of the shared ride and the driving route of the travel information. Judging whether the matching degree is greater than a preset matching degree threshold, and taking the shared ride as the shared ride that matches the travel information when it is greater than the matching degree threshold.

4. The online car-hailing reservation method according to claim 2, wherein The determining different categories of real-time car-hailing services corresponding to the travel information, includes: Determining the travel distance based on the starting point and ending point of the travel information. Judging whether the travel distance is less than a preset distance threshold: If so, then determining the carpool matching degree of the travel information with each currently running carpool, and determining whether there is a carpool that matches the travel information based on the carpool matching degree. If there is, then taking carpool, private car, and express car as the real-time car-hailing services corresponding to the travel information. If there is no carpool, then taking private car and express car as the real-time car-hailing services corresponding to the travel information. Otherwise, then taking taxi, private car, and express car as the real-time car-hailing services corresponding to the travel information.

5. The online car-hailing reservation method according to any one of claims 1 or 2, characterized in that When the candidate online car-hailing service is a shared ride or a carpool, obtaining the estimated cost and estimated arrival time of the candidate online car-hailing service through a cost and time prediction model, includes: Determine the current location of the online car-hailing vehicle based on its identifier and the number of passengers to be carried; Determine the driving route based on the current location of the candidate online car-hailing vehicle and the trip information; wherein, the trip route is the expected driving route for the candidate online car-hailing vehicle to drive from the current location to the end point of the trip information; Input the driving route, current location and the trip information of the candidate online car-hailing vehicle into the time prediction module of the cost and time prediction model to obtain the expected arrival time; Input the driving route, the number of passengers to be carried, the expected arrival time and the trip information of the candidate online car-hailing vehicle into the cost prediction module of the cost and time prediction model to obtain the expected cost.

6. The online car-hailing reservation method according to any one of claims 1 or 2, characterized in that, When the candidate online car-hailing vehicle is a special car, an express car or a taxi, for each type of candidate online car-hailing vehicle, obtain the expected cost and expected arrival time of the candidate online car-hailing vehicle through the cost and time prediction model, including: Determine the corresponding order assignment area based on the starting point and end point of the trip information; Determine the number of idle drivers and the total number of current orders to be received corresponding to this type of candidate online car-hailing vehicle from the order assignment area; Calculate the current supply-demand ratio of this type of candidate online car-hailing vehicle based on the number of idle drivers and the total number of orders to be received; Input the supply-demand ratio and the trip information into the cost and time prediction model to predict the expected cost and expected arrival time of this type of candidate online car-hailing vehicle.

7. The online car-hailing reservation method according to claim 1, wherein, The multi-task learning model is a hard parameter sharing model.

8. The online car-hailing reservation method according to claim 7, wherein The hard parameter sharing model is an MT-DNN model.

9. The online car-hailing reservation method according to claim 1, wherein After the user terminal displays the expected cost and expected arrival time of various candidate online car-hailing vehicles for the user to select, it further includes: the user terminal jumps to the corresponding type of online car-hailing vehicle interface in response to the user's selection of a candidate online car-hailing vehicle.

10. A hitchhiking reservation system, characterized in that, The system includes a user terminal and a server that are communicatively connected to each other: The user terminal includes: A first communication module, configured to send the user's trip information to the server and receive the expected cost and expected arrival time of various candidate online car-hailing vehicles sent by the server; A data acquisition module, configured to acquire the user's trip information; A display module, configured to display the expected cost and expected arrival time of various candidate online car-hailing vehicles for the user to select; The server includes: A second communication module, configured to receive the trip information sent by the user terminal and send the expected cost and expected arrival time of various candidate online car-hailing vehicles to the user terminal; A cost and time determination module, configured to determine multiple different types of candidate online car-hailing vehicles corresponding to the trip information, and when it is detected that the candidate online car-hailing vehicle is a hitchhiking car and real-time car-hailing of different categories, obtain the expected cost and expected arrival time of each candidate online car-hailing vehicle through the cost and time prediction model; wherein, the cost and time prediction model is a multi-task learning model.

Images