CN116415701A - Method and system for predicting arrival time of vehicle - Google Patents

Abstract

The embodiment of the specification provides a vehicle arrival time prediction method, which comprises the steps of obtaining position information of a target station and first positioning information reported by a target vehicle; determining a time interval from a first moment to a current moment corresponding to the first positioning information; and predicting the current position of the target vehicle at the current moment and the time of the target vehicle reaching the target station through a prediction model based on the position information of the target station, the first positioning information and the time interval.

Description

A vehicle arrival time prediction method and system

technical field

This specification relates to the field of information technology, in particular to a method and system for predicting the arrival time of a vehicle.

Background technique

Bus users hope to be able to view the real-time location of the bus and predict the arrival time on the user end, so as to decide when to arrive at the station and wait for the bus. However, the existing bus GPS tracking system has large intervals and delays, resulting in inaccurate bus location information, and based on this, the predicted arrival time is also inaccurate. Therefore, it is necessary to provide a vehicle arrival time prediction method and system to improve the accuracy of the predicted arrival time.

Contents of the invention

One of the embodiments of this specification provides a method for predicting the arrival time of a vehicle. The vehicle arrival time prediction method includes: obtaining the position information of the target site and the first positioning information reported by the target vehicle; determining the time interval from the first moment to the current moment corresponding to the first positioning information; The position information of the target vehicle, the first positioning information, and the time interval are used to predict the current position of the target vehicle at the current moment and the time when the target vehicle arrives at the target site through a prediction model.

One of the embodiments of this specification provides a vehicle arrival time prediction system, the vehicle arrival time prediction system includes an acquisition module, a determination module and a prediction module; the acquisition module is used to acquire the location information of the target site and the report of the target vehicle the first location information; the determination module is used to determine the time interval from the first moment corresponding to the first location information to the current moment; The positioning information and the time interval are used to predict the current position of the target vehicle at the current moment and the time when the target vehicle arrives at the target site through a prediction model.

One of the embodiments of this specification provides a vehicle arrival time prediction device, including a processor and a storage device, the storage device is used to store instructions, and when the processor executes the instructions, the vehicle arrival time prediction is realized method.

Description of drawings

This specification will be further illustrated by way of exemplary embodiments, which will be described in detail with the accompanying drawings. These examples are non-limiting, and in these examples, the same number indicates the same structure, wherein:

Fig. 1 is a schematic diagram of an application scenario of a vehicle arrival time prediction system according to some embodiments of this specification;

Fig. 2 is an exemplary flowchart of a vehicle arrival time prediction method according to some embodiments of the present specification;

Fig. 3 is an exemplary flow chart of determining a time interval according to some embodiments of the present specification;

Fig. 4 is a schematic diagram of two modeling modes of the prediction model according to some embodiments of the present specification;

Fig. 5 is an exemplary flowchart of training an initial model according to some embodiments of the present specification;

Fig. 6 is a block diagram of a vehicle arrival time prediction system according to some embodiments of the present specification.

Detailed ways

In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the following briefly introduces the drawings that need to be used in the description of the embodiments. Apparently, the accompanying drawings in the following description are only some examples or embodiments of this specification, and those skilled in the art can also apply this specification to other similar scenarios. Unless otherwise apparent from context or otherwise indicated, like reference numerals in the figures represent like structures or operations.

It should be understood that "system", "device", "unit" and/or "module" as used herein is a method for distinguishing different components, elements, parts, parts or assemblies of different levels. However, the words may be replaced by other expressions if other words can achieve the same purpose.

As indicated in the specification and claims, the terms "a", "an", "an" and/or "the" are not specific to the singular and may include the plural unless the context clearly indicates an exception. Generally speaking, the terms "comprising" and "comprising" only suggest the inclusion of clearly identified steps and elements, and these steps and elements do not constitute an exclusive list, and the method or device may also contain other steps or elements.

The flowchart is used in this specification to illustrate the operations performed by the system according to the embodiment of this specification. It should be understood that the preceding or following operations are not necessarily performed in the exact order. Instead, various steps may be processed in reverse order or simultaneously. At the same time, other operations can be added to these procedures, or a certain step or steps can be removed from these procedures.

Fig. 1 is a schematic diagram of an application scenario of a vehicle arrival time prediction system according to some embodiments of the present specification. In some embodiments, the vehicle arrival time prediction system 100 may include a server 110 , a network 120 , a user terminal 130 , a vehicle 140 and a storage device 150 .

In some embodiments, server 110 may be a single server or a group of servers. The server group can be centralized or distributed (eg, server 110 can be a distributed system). In some embodiments, server 110 may be local or remote. For example, server 110 may access information and/or data stored in vehicle 140 and/or storage device 150 via network 120 . As another example, server 110 may be directly connected to vehicle 140 and/or storage device 150 to access stored information and/or data. In some embodiments, the server 110 can be implemented on a cloud platform or on-vehicle computer. By way of example only, a cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-tier cloud, etc., or any combination thereof. In some embodiments, the server 110 may be deployed on the user terminal 130 (eg, the processor of the user terminal 130 may serve as the server 110 or a part thereof). In some embodiments, the server 110 may be deployed on a processing device of the online service platform (for example, the processing device of the online service platform may serve as the server 110 or a part thereof). In some embodiments, server 110 may be deployed on vehicle 140 (eg, a processor of vehicle 140 may be server 110 or a portion thereof).

In some embodiments, server 110 may include processing device 112 . Processing device 112 may process information and/or data related to a service request to perform one or more functions described herein. For example, the processing device 112 can predict the arrival time of the vehicle from the service request obtained by the user terminal 130 . For another example, the processing device 112 may obtain the latitude and longitude information of the target site. In some embodiments, processing device 112 may include one or more processing engines (eg, single-chip processing engines or multi-chip processing engines). By way of example only, processing device 112 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), an application-specific instruction set processor (ASIP), a graphics processing unit (GPU), a physical processing unit (PPU), a digital signal Processor (DSP), Field Programmable Gate Array (FPGA), Programmable Logic Device (PLD), Controller, Microcontroller Unit, Reduced Instruction Set Computer (RISC), Microprocessor, etc. or any combination thereof.

Network 120 may facilitate the exchange of information and/or data. In some embodiments, one or more components in the vehicle arrival time prediction system 100 (for example, the server 110, the user terminal 130, the vehicle 140 and/or the storage device 150) can send the vehicle arrival time prediction system 100 to the vehicle arrival time prediction system 100 through the network 120 to send information and/or data to other components. For example, the server 110 may acquire/obtain a service request from the user terminal 130 through the network 120 . In some embodiments, the network 120 may be any form of wired network or wireless network, or any combination thereof. By way of example only, network 120 may include a cable network, a cable network, a fiber optic network, a telecommunications network, an intranet, the Internet, a local area network (LAN), a wide area network (WAN), a wireless local area network (WLAN), a metropolitan area network (MAN) , Public Switched Telephone Network (PSTN), Bluetooth network, Zigbee network, Near Field Communication (NFC) network, etc. or any combination thereof. In some embodiments, network 120 may include one or more network switching points. For example, network 120 may include wired or wireless network exchange points, such as base stations and/or Internet exchange points 120-1, 120-2, ..., through which one or more components of vehicle arrival time prediction system 100 Can be connected to network 120 to exchange data and/or information.

In some embodiments, a user (eg, a passenger) may be the owner of the user terminal 130 . In some embodiments, the owner of the user terminal 130 may be someone other than the passenger. For example, owner A of user terminal 130 may use user terminal 130 to send a service request for passenger B, or to receive service confirmation and/or information or instructions from server 110 .

In some embodiments, the user terminal 130 may include a mobile device 130-1, a tablet 130-2, a laptop 130-3, etc. or any combination thereof. In some embodiments, the mobile device 130-1 may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, an augmented reality device, etc., or any combination thereof. In some embodiments, smart home devices may include smart lighting devices, smart electrical control devices, smart monitoring devices, smart TVs, smart cameras, walkie-talkies, etc., or any combination thereof. In some embodiments, wearable devices may include smart bracelets, smart footwear, smart glasses, smart helmets, smart watches, smart clothing, smart backpacks, smart accessories, etc., or any combination thereof. In some embodiments, a smart mobile device may include a smart phone, a personal digital assistant (PDA), a gaming device, a navigation device, a point of sale (POS), etc., or any combination thereof. In some embodiments, the virtual reality device and/or the augmented reality device may include a virtual reality helmet, virtual reality glasses, virtual reality goggles, augmented reality helmets, augmented reality glasses, augmented reality goggles, etc. or any combination thereof. For example, virtual reality devices and/or augmented reality devices may include Google™ Glass, OculusRift, HoloLens, GearVR, and the like. In some embodiments, the user terminal 130 may be a device with positioning technology for locating the location of the user and/or the user terminal 130 .

Vehicle 140 may be any type of vehicle. Vehicles 140 may include automated vehicles and non-autonomous vehicles. Autonomous vehicles are capable of sensing environmental information and navigating without human manipulation. Vehicle 140 may include the structure of a conventional vehicle. For example, vehicle 140 may include at least two control components configured to control operation of vehicle 140 . The at least two control components may include a steering device (eg, steering wheel), a braking device (eg, brake pedal), an accelerator, and the like. The steering device may be configured to adjust the heading and/or direction of the vehicle 140 . The braking device may be configured to perform a braking operation to stop the vehicle 140 . The accelerator may be configured to control the speed and/or acceleration of the vehicle 140 . Vehicle 140 may also include a Global Positioning System (GPS) module. In some embodiments, the vehicle 140 may report positioning information (eg, first positioning information) to the processing device 112 .

Storage device 150 may store data and/or instructions. In some embodiments, the storage device 150 may store data obtained from the user terminal 130 . In some embodiments, storage device 150 may store data and/or instructions used by server 110 to perform or use to perform the exemplary methods described herein. In some embodiments, the storage device 150 may include mass storage, removable storage, volatile read-write memory, read-only memory (ROM), etc., or any combination thereof. Exemplary mass storage may include magnetic disks, optical disks, solid state drives, and the like. Exemplary removable storage may include flash drives, floppy disks, optical disks, memory cards, compact disks, magnetic tape, and the like. Exemplary volatile read-write memory may include random access memory (RAM). Exemplary RAMs may include dynamic random access memory (DRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), static random access memory (SRAM), thyristor random access memory (T-RAM), and zero capacitance Random Access Memory (Z-RAM), etc. Exemplary read only memories may include mask read only memory (MROM), programmable read only memory (PROM), erasable programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), Compact Disk Read Only Memory (CD-ROM) and Digital Versatile Disk Read Only Memory, etc. In some embodiments, the storage device 150 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.

In some embodiments, the storage device 150 may be connected to the network 120 to communicate with at least one component of the vehicle arrival time prediction system 100 (eg, the server 110 , the user terminal 130 or the vehicle 140 ). At least one component of the vehicle arrival time prediction system 100 can access data or instructions stored in the storage device 150 via the network 120 . In some embodiments, the storage device 150 may be directly connected to or in communication with at least one component of the vehicle arrival time prediction system 100 (eg, the server 110 , the user terminal 130 or the vehicle 140 ). In some embodiments, storage device 150 may be part of server 110 . In some embodiments, the storage device 150 may be deployed on the server 110 , the user terminal 130 , and/or the vehicle 140 . In some embodiments, the storage device 150 can be deployed independently, and the server 110, the user terminal 130, and/or the vehicle 140 can access the storage device 150 directly or through the network 120 to obtain relevant data and/or instructions.

It should be noted that the above description of the vehicle arrival time prediction system is only for convenience of description, and does not limit this specification to the scope of the examples. It can be understood that for those skilled in the art, after understanding the principle of the system, it is possible to combine various components arbitrarily, or form a subsystem to connect with other components without departing from this principle.

Fig. 2 is an exemplary flowchart of a method for predicting the arrival time of a vehicle according to some embodiments of the present specification. As shown in FIG. 2 , the process 200 includes the following steps. In some embodiments, the process 200 may be performed by the processing device 112 .

Step 210, acquiring the location information of the target site and the first positioning information reported by the target vehicle. In some embodiments, step 210 may be performed by the obtaining module 610 .

The target site is the site where the target vehicle will arrive, and the predicted arrival time is the time when the target vehicle will arrive at the target site.

In some embodiments, the target site can be directly designated by the user. For example, the user can query the bus site (such as "XX Street Station") that the user is going to or the user is in through the user terminal 130. Station identified as the target site. In some embodiments, the target site may be determined by a machine. For example, when authorized by the user, the processing device 112 may obtain the location information of the user terminal 130, and determine a bus station near the user according to the location information of the user terminal 130, and determine it as the target station. It can be understood that when it is determined that there are multiple bus stops near the user, one bus stop can be selected as the target site according to a preset strategy, or each of the multiple bus stops can be determined as the target site, and each target Predicted arrival time at the station. In some embodiments, the method of determining the target site may also be a combination of manual and machine methods. For example, in the case of user authorization, the processing device 112 can recommend several sites near the user to the user terminal 130 according to the user's query history and/or the location information of the user terminal 130, and the user can select one of the sites through the user terminal 130, thereby Processing device 112 may determine the site as a target site.

Similarly, the target vehicle can be specified by the user, can also be determined by a machine, or can be determined by a combination of manual and machine. For example, the user may inquire about the bus to be boarded (such as "Route 128") through the user terminal 130, and the processing device 112 may determine the bus inquired by the user as the target stop. As another example, the processing device 112 can acquire the location information of the target site, and determine one or more vehicles (such as two, three, etc. vehicle) bus, and then the bus closest to the determined target site can be determined as the target vehicle. When it is determined that there are at least two buses closest to the target stop, the processing device 112 may use each of the determined at least two buses as a target vehicle, and predict an arrival time for each target vehicle. When it is determined that there are at least two buses closest to the target stop, the processing device 112 may also prompt the user to select one of them, so that the processing device 112 may determine this bus as the target vehicle.

The target vehicle may be equipped with a positioning system (such as GPS system, Beidou positioning system), and report the positioning information (such as GPS information) obtained through the positioning system to the processing device 112 on demand or continuously. The positioning information may indicate the position (such as latitude and longitude) of the positioning object (such as the target vehicle) at a certain moment (such as accurate to minutes and seconds). In some embodiments, the positioning information may also indicate more information, for example, the speed of the positioning object, the altitude of the positioning object, and the like. For example, the first positioning information may indicate the position of the target vehicle at the first moment. In some embodiments, the first moment may be the moment when the target vehicle reported the last location information before the current moment; correspondingly, the first location information may be the last location information reported by the target vehicle before the current moment.

Step 220, determine the time interval from the first moment corresponding to the first positioning information to the current moment. In some embodiments, step 220 may be performed by a determination module 620 .

Referring to FIG. 3 , the first moment corresponding to the historical position (the position indicated by the first positioning information) is t0, and the moment corresponding to the current position (current moment) is t1, based on the historical position of the target vehicle and the time from the first moment to the current moment interval (t1-t0), the current position of the target vehicle at the current moment can be predicted. Based on the current location of the target vehicle and the location of the target site, the time when the target vehicle arrives at the target site can be predicted. That is, based on the position information of the target site, the first positioning information and the time interval, the current position of the target vehicle and the time when the target vehicle arrives at the target site can be predicted. The first location information may be location information reported by the target vehicle at a certain moment (for example, the first moment) before the current moment. For specific prediction details, refer to the detailed description of step 230 later.

In some embodiments, the processing device 112 may determine the initiation time of the prediction request as the current time, for example, the prediction request may carry/include a timestamp to indicate the initiation time of the prediction request. In some embodiments, the processing device 112 may determine the time at which the prediction request is received as the current time. It can be understood that the current moment in step 220 is not necessarily the current moment that meets the user's expectations. The current moment that meets the user's expectations may refer to the real moment when the user terminal 130 displays the predicted current location. Refresh moment. Considering factors such as the time spent in prediction, network communication time, etc., the current time can also be estimated on the basis of the initiation/reception time of the prediction request, for example, the processing device 112 can preset the time period (such as 0.1s, 0.5s, 1s, 2s, 5s, etc.) as the current time.

Step 230, based on the location information of the target station, the first positioning information and the time interval, predict the current location of the target vehicle at the current moment and the time when the target vehicle arrives at the target station (or arrival time) through the prediction model. In some embodiments, step 230 may be performed by prediction module 630 .

The predictive model can be any model that predicts the vehicle's current location and arrival time. In some embodiments, the predictive model may include a machine learning model. Specifically, the predictive model may include one or more of a regression model, a decision tree, a neural network, and the like.

In some embodiments, the processing device 112 may input the location information of the target site, the first location information and the time interval into the prediction model. Furthermore, the processing device 112 may use the feature information to predict the current location of the target vehicle and the time when the target vehicle arrives at the target site through the prediction model. In some embodiments, the feature information can be extracted from the model input. Specifically, the model input can be directly used as feature information, or the model input (which can be regarded as input layer features) can be processed to obtain feature information (which can be regarded as intermediate layer features). In some embodiments, feature information may be obtained by the model from the storage device 150 and/or other application services (eg, map services).

In some embodiments, the feature information may include at least one of road information, road speed information, road congestion information, time period information, and traffic signal light information.

Wherein, the road information may include information of one or more roads (such as road signs) traveled by the target vehicle from the historical position (for example, the position indicated by the first positioning information) to the target site. For example, when the target vehicle is a bus, the road information may be a predetermined bus route, and the bus route may include representations of all roads from the starting point to the ending point. Specifically, the road information can be stored in the form of a sequence, and the order of the road signs in the sequence can be determined by the order in which the target vehicle passes through the roads when it travels to the target site.

The road speed information may indicate the vehicle speed on one or more roads of interest, for example, the target vehicle travels from the historical location to the target site via road a, road b and/or road c, correspondingly, the road speed information The vehicle traffic speed of road a, the vehicle traffic speed of road b and/or the vehicle traffic speed of road c may be included. The vehicle speed on the road may be a real-time speed, such as an average speed based on the statistics of the trajectory data of multiple vehicles currently on the road, or an estimated speed obtained through data mining.

Road congestion information can reflect the congestion status of one or more roads of interest, such as whether it is congested, the degree of congestion, and so on.

The period information may reflect which one of the preset periods the current period belongs to. Traffic conditions in different time periods are often different. For example, the whole day can be divided into morning peak (such as 8:00 am to 9:00 am), evening peak (such as 5:00 pm to 7:00 pm) and other time periods. Usually, the traffic volume during peak hours is large. , and the traffic speed is correspondingly slow. In some embodiments, it can be considered that the traffic volume in the morning peak is greater than that in the evening peak, that is, the traffic speed in the morning peak is considered to be slower than that in the evening peak. Of course, the whole day can also be divided into peak hours (such as including 8:00 am to 9:00 pm and 5:00 pm to 7:00 pm) and other time periods.

The traffic signal light information can reflect the signals indicated by one or more traffic lights on the road section where the target vehicle travels from the historical position (the position indicated by the first positioning information) to the target site within a period of time (such as red light indicating no traffic, yellow light Indicates readiness, green light indicates clearance) or signal change pattern (for example, red light duration, green light duration, amber light duration). In some embodiments, one or more of the above characteristic information can affect the traveling speed of the target vehicle, and further affect the current position of the target vehicle and the time when it arrives at the target site.

In some embodiments, the feature information may also include derivation route information and/or derivation time information. Wherein, the deduced route information may reflect the location corresponding to the first positioning information as the starting point, and the estimated location where the target vehicle will arrive after the above time interval. The derivation time information can reflect the estimated time required for the target vehicle to reach the target site starting from the current location. It is not difficult to see that the derivation route information can be used as the preliminary estimated current position, and the deduction time information can be used as the preliminary estimated arrival time. The prediction model can correct the derivation route information and/or derivation time information, so as to output the predicted current position and/or arrival time. In some embodiments, the model input can be used to make a preliminary estimate of the current position and/or arrival time, and then use other information in the feature information except the derivation route information and/or derivation time information to correct the preliminary estimate result, thereby The prediction result of the prediction model (predicted current position and/or arrival time) is obtained. As an example only, deduction can be carried out according to the predetermined driving route of the target vehicle and the passing speed of each road on the route to obtain the position corresponding to the first positioning information as the starting point and the target vehicle’s arrival position after the above-mentioned time interval, that is, to obtain a preliminary Estimated current location. Furthermore, deduction can be carried out according to the predetermined driving route of the target vehicle and the traffic speed of each road on the route, and the time required for the target vehicle to reach the target station is obtained starting from the preliminary estimated current position, that is, the preliminary estimated arrival time. station time.

Figure 4 shows two modeling ways of the prediction model. In some embodiments, the prediction model may include a current location prediction model and an arrival time prediction model. After the prediction request is initiated, the time interval from the first moment to the current moment corresponding to the first positioning information and the first positioning information are input to the current location Forecasting model, the output of the current position prediction model (predicted current position) is one of the inputs of the arrival time prediction model, and the input of the arrival time prediction model can also include the position information of the target site, and the output of the arrival time prediction model is Predicted arrival time. It can be seen that this modeling method includes two stages of current position prediction and arrival time prediction, and there may be certain errors in the prediction of each stage, that is, there may be double errors in this modeling method. In some other embodiments, the prediction model may be a fusion model of position prediction and arrival time prediction, that is, after the prediction request is initiated, the location information of the target station, the first positioning information and the time interval may be input into the fusion model, the The fusion model can simultaneously output the predicted current location and arrival time. In this way, unified modeling can be realized, and only one error may exist, thereby reducing the error of prediction and improving the accuracy of prediction.

When the prediction model is a machine learning model, the processing device 112 may obtain a training data set, and then use the training data set to train an initial model to obtain a trained prediction model. Wherein, the training data set may include features and/or labels of multiple training samples. In some embodiments, the features of each training sample may include the location information of the sample site, the first sample location information reported by the sample vehicle, and the first sample time corresponding to the first sample location information to the first sample location information reported by the sample vehicle. For the sample time interval at the second sample moment corresponding to the two-sample positioning information, the label of each training sample may include actual location information and actual arrival time information. The first sample location information and the second sample location information may be location information reported by the sample vehicles at two different times (for example, the first sample time and the second sample time). In some embodiments, the first sample instant may be earlier than the second sample instant. In some embodiments, the actual position information may include: the real position of the sample vehicle at the second sample time; or, the distance between the real positions of the sample vehicle between the first sample time and the second sample time. The actual arrival time information may include: the actual time when the sample vehicle arrives at the sample site; or, the time difference between the second sample moment and the actual time when the sample vehicle arrives at the sample site.

It should be noted that in the training phase, the second sample moment corresponding to the second sample location information reported by the sample vehicle can be used as the current moment corresponding to the predicted current position, and when the prediction model obtained by training is applied (that is, the prediction stage), generally based on the prediction request to determine the current moment corresponding to the predicted current location. This is because, in the process of providing prediction services to users, waiting for the target vehicle to report the second positioning information may require a time penalty (for example, there is a certain delay from uploading to receiving the positioning information), and further, it may cause a loss in the predicted position. The real-time performance is poor. For example, assuming that the user terminal 130 initiates a service request at 10:00, the second moment corresponding to the second positioning information is 10:01, and the processing device 112 receives the second positioning information at 10:02 (that is, there is 1 minute Delay). If the second moment (10:01) is used as the current moment corresponding to the predicted current location, the user terminal 130 may receive a service response (including the predicted location result) after 10:02, but the predicted location is the target The vehicle is at 10:01, and there is a time error of at least 1 minute. Obviously, when the initiation time of the service request (10:00) is used as the current time corresponding to the predicted current position, the processing device 112 does not need to wait for the arrival of the second positioning information, which saves a time error of 1 minute and can provide better real-time performance. location prediction service.

Fig. 5 is an exemplary flow chart of training an initial model according to some embodiments of the present specification. Process 500 may be performed by processing device 112 (eg, a training module) or other devices (eg, third-party devices). As shown in FIG. 5 , the process 500 may include the following steps.

Step 510, input the features of each training sample into the initial model to obtain prediction results.

Step 520, adjust the parameters of the initial model based on the difference between the prediction result of each training sample and the label.

In some embodiments, the value of the loss function may be calculated based on the prediction results and labels of each training sample. The loss function can reflect the difference between the prediction result and the label of each training sample. For the unified modeling method, the loss function can include two parts of loss: position loss and time loss. In some embodiments, the position loss may reflect the difference (error) between the predicted position of the sample vehicle in the prediction result and the real position of the sample vehicle at the second sample moment. In some embodiments, the time loss may reflect the difference (error) between the predicted time when the sample vehicle arrives at the sample site and the actual time when the sample vehicle arrives at the sample site in the prediction result. For example, for each training sample, the position loss can be expressed as the distance between the predicted position of the sample vehicle in the prediction result of the training sample and the real position of the sample vehicle at the second sample moment (such as taking the absolute value after calculating the position difference) , the time loss can be expressed as the time deviation (error) between the predicted time when the sample vehicle arrives at the sample site and the actual time when the sample vehicle arrives at the sample site in the prediction result of the training sample.

A loss function reflects the error over multiple training samples. In some embodiments, any predictive evaluation index in Mean Absolute Error (Mean Absolute Error, MAE), Mean Square Error (Mean Square Error, MSE), Root Mean Square Error (RootMean Square Error, RMSE) etc. can be used to construct the loss function.

In some embodiments, the loss function may include a weighted sum of a position loss and a time loss. Among them, the position loss/time loss can be weighted according to the importance of the position loss/time loss. For example, when the time loss is more important (for example, the user may pay more attention to the arrival time of the vehicle), the weight of the time loss can be greater than The weight of the position loss.

It can be understood that the training (model tuning process) is the optimization process of the loss function. When the value of the loss function is less than or equal to the set threshold, the training can be stopped, and the model parameters obtained at this time are the parameters of the prediction model. By way of example only, the training algorithm may employ a gradient descent algorithm. Specifically, a batch gradient descent algorithm, a stochastic gradient descent algorithm or other types of gradient descent algorithms may be used.

In some embodiments, after receiving the service response, the user terminal 130 may simultaneously display the current location of the target vehicle predicted by the prediction model and the time when the target vehicle arrives at the target site. For example, the graphical user interface of the user terminal 130 can include a map, which can mark the current position of the target vehicle predicted by the prediction model (such as a certain bus closest to the bus station queried by the user), and at the same time, the graphical interface can display the predicted The time when the target vehicle arrives at the target stop (such as the bus stop queried by the user) predicted by the model, such as 5 minutes later or 10:05.

It should be noted that the above descriptions about the process 200 and the process 500 are only for illustration and description, and do not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the process 200 and/or the process 500 under the guidance of this description. However, such modifications and changes are still within the scope of this specification. For example, in the training phase, the initiation time of the sample prediction request can also be used as the current time corresponding to the predicted current position, that is, the features of each training sample include the first sample location information corresponding to the first sample location information reported by the sample vehicle. The sample time interval from this moment to the time when the sample prediction request is initiated (received). Correspondingly, in the label of each training sample, the actual position can be the real position of the sample vehicle at the time when the sample prediction request is initiated (received) or the first sample The distance between the real position of the sample vehicle between this moment and the time when the sample prediction request is initiated (received). The time difference between the actual time the vehicle arrived at the sample site.

Fig. 6 is a block diagram of a vehicle arrival time prediction system according to some embodiments of the present specification.

In some embodiments, the vehicle arrival time prediction system 600 may include an acquisition module, a determination module and a prediction module.

In some embodiments, the obtaining module 610 may be used to obtain the location information of the target site and the first location information reported by the target vehicle.

In some embodiments, the confirming module 620 may be used to determine the time interval from the first moment corresponding to the first positioning information to the current moment.

In some embodiments, the prediction module 630 can be used to predict the current position of the target vehicle at the current moment and the time when the target vehicle arrives at the target site through a prediction model based on the position information of the target site, the first positioning information and the time interval.

In some embodiments, the prediction module 630 can be further used to input the position information of the target site, the first positioning information and the time interval into the prediction model; through the prediction model, the current position of the target vehicle and the arrival of the target vehicle can be predicted by using the feature information time at the target site. The feature information may include at least one of road information, road speed information, road congestion information, time period information, and traffic signal light information. In some embodiments, the feature information may also include derivation route information and/or derivation time information. The deduced route information may reflect the location corresponding to the first positioning information as the starting point, and the estimated location where the target vehicle will arrive after the aforementioned time interval. The derivation time information can reflect the estimated time required for the target vehicle to reach the target site starting from the current location. In some embodiments, the prediction model may be a fusion model of location prediction and arrival time prediction.

In some embodiments, the vehicle arrival time prediction system 600 may further include a display module (not shown). The display module can simultaneously display the current position of the target vehicle predicted by the prediction model and the time when the target vehicle arrives at the target site.

In some embodiments, the vehicle arrival time prediction system 600 may further include a training module (not shown). In some embodiments, the training module and the prediction module 630 (and/or the acquisition module 610, the confirmation module 620) may be deployed on different processing devices or processors (eg, servers). For example, the training module can train the prediction model offline on one or more processing devices; and the prediction module 630 can apply the prediction model on another processing device online to predict the current location of the vehicle and/or the arrival time of the vehicle.

In some embodiments, the training module can be used to obtain a training data set, and the training data set can include features and labels of a plurality of training samples; use the training data set to train the initial model; wherein, the features of each training sample can include sample sites The position information of the sample vehicle, the first sample location information reported by the sample vehicle, and the sample time interval from the first sample moment corresponding to the first sample location information to the second sample moment corresponding to the second sample location information reported by the sample vehicle, each The labels of each training sample can include actual location information and actual arrival time information. The actual position information may be: the real position of the sample vehicle at the second sample time; or the distance between the real positions of the sample vehicles between the first sample time and the second sample time. The actual arrival time information may be: the actual time when the sample vehicle arrives at the sample site; or the time difference between the second sample moment and the actual time when the sample vehicle arrives at the sample site.

In some embodiments, the training module can be further used to input the features of each training sample into the initial model to obtain a prediction result; based on the difference between the prediction result of each training sample and the label, adjust the parameters of the initial model. Wherein, the loss function may include a weighted sum of position loss and time loss. The position loss may reflect the difference between the predicted position of the sample vehicle in the prediction result and the real position of the sample vehicle at the second sample moment. The time loss can reflect the difference between the predicted time when the sample vehicle arrives at the sample site in the prediction results and the actual time when the sample vehicle arrives at the sample site. In some embodiments, the time loss may be weighted more than the position loss.

It should be noted that the above description of the modules is only for convenience of description, and does not limit this description to the scope of the examples. It can be understood that for those skilled in the art, after understanding the principle of the system, it is possible to combine various modules arbitrarily, or form a subsystem to connect with other modules without departing from this principle. In some embodiments, the acquisition module, confirmation module and prediction module disclosed in FIG. 6 may be different modules in one system, or one module may realize the functions of the above two or more modules. For example, each module may share one storage module, or each module may have its own storage module. Such deformations are within the protection scope of this specification.

The possible beneficial effects of the embodiments of this specification may include but are not limited to: (1) providing users with vehicle current location prediction services and arrival time prediction services at the same time, so that users can make further decisions; (2) improving location prediction and The real-time performance of arrival time prediction; (3) The fusion model of location prediction and arrival time prediction is designed, which can realize unified modeling, reduce prediction error and improve prediction accuracy.

The basic concept has been described above, obviously, for those skilled in the art, the above detailed disclosure is only an example, and does not constitute a limitation to this description. Although not expressly stated here, those skilled in the art may make various modifications, improvements and corrections to this description. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

Meanwhile, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and/or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that two or more references to "an embodiment" or "an embodiment" or "an alternative embodiment" in different places in this specification do not necessarily refer to the same embodiment . In addition, certain features, structures or characteristics in one or more embodiments of this specification may be properly combined.

In addition, unless explicitly stated in the claims, the order of processing elements and sequences described in this specification, the use of numbers and letters, or the use of other names are not used to limit the sequence of processes and methods in this specification. While the foregoing disclosure has discussed by way of various examples some embodiments of the invention that are presently believed to be useful, it should be understood that such detail is for illustrative purposes only and that the appended claims are not limited to the disclosed embodiments, but rather, the claims The claims are intended to cover all modifications and equivalent combinations that fall within the spirit and scope of the embodiments of this specification. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by a software-only solution, such as installing the described system on an existing server or mobile device.

In the same way, it should be noted that in order to simplify the expression disclosed in this specification and help the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of this specification, sometimes multiple features are combined into one embodiment, drawings or descriptions thereof. This method of disclosure does not, however, imply that the subject matter of the specification requires more features than are recited in the claims. Indeed, embodiment features are less than all features of a single foregoing disclosed embodiment.

In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of the embodiments use the modifiers "about", "approximately" or "substantially" in some examples. grooming. Unless otherwise stated, "about", "approximately" or "substantially" indicates that the stated figure allows for a variation of ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximations that can vary depending upon the desired characteristics of individual embodiments. In some embodiments, numerical parameters should take into account the specified significant digits and adopt the general digit reservation method. Although the numerical ranges and parameters used in some embodiments of this specification to confirm the breadth of the range are approximations, in specific embodiments, such numerical values are set as precisely as practicable.

Each patent, patent application, patent application publication, and other material, such as article, book, specification, publication, document, etc., cited in this specification is hereby incorporated by reference in its entirety. Application history documents that are inconsistent with or conflict with the content of this specification are excluded, and documents (currently or later appended to this specification) that limit the broadest scope of the claims of this specification are also excluded. It should be noted that if there is any inconsistency or conflict between the descriptions, definitions, and/or terms used in the accompanying materials of this manual and the contents of this manual, the descriptions, definitions and/or terms used in this manual shall prevail .

Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other modifications are also possible within the scope of this description. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims (10)

1. A vehicle arrival time prediction method, comprising: Obtain the location information of the target site and the first positioning information reported by the target vehicle; determining the time interval from the first moment corresponding to the first positioning information to the current moment; Based on the position information of the target site, the first positioning information, and the time interval, predict the current position of the target vehicle at the current moment and the time when the target vehicle arrives at the target site through a prediction model. 2. The method according to claim 1, wherein said predicting the current position of the target vehicle at the current moment and the time when the target vehicle arrives at the target site by a prediction model comprises: inputting the position information of the target site, the first positioning information and the time interval into the prediction model; Predict the current position of the target vehicle and the time when the target vehicle arrives at the target site through the predictive model, using feature information, the feature information includes road information, road speed information, road congestion information, time period information , at least one of traffic signal light information. 3. The method according to claim 2, wherein the feature information further includes derivation route information and/or derivation time information; the derivation route information reflects that starting from the position corresponding to the first positioning information, it is estimated that The location where the target vehicle arrives after the time interval; the derivation time information reflects the estimated time required for the target vehicle to arrive at the target site starting from the current location. 4. The method of claim 1, wherein the predictive model is obtained by: Obtain a training data set, the training data set includes features and labels of a plurality of training samples; training an initial model using said training data set; The features of each training sample include the location information of the sample site, the first sample location information reported by the sample vehicle, and the first sample time corresponding to the first sample location information to the second sample location information reported by the sample vehicle. The sample time interval of the second sample moment corresponding to the sample positioning information, the label of each training sample includes actual position information and actual arrival time information; The actual location information is: the real position of the sample vehicle at the second sample moment; or The distance between the real positions of the sample vehicles between the first sample moment and the second sample moment; The actual arrival time information is: the actual time the sample vehicle arrived at the sample site; or The time difference between the second sample moment and the actual time when the sample vehicle arrives at the sample site. 5. The method according to claim 4, wherein said training an initial model using said training data set comprises: Inputting the features of each training sample into the initial model to obtain prediction results; Adjusting the parameters of the initial model based on the difference between the prediction result of each training sample and the label; Wherein, the loss function includes a weighted sum of position loss and time loss, and the position loss reflects the difference between the predicted position of the sample vehicle in the prediction result and the real position of the sample vehicle at the second sample moment The time loss reflects the difference between the predicted time when the sample vehicle arrives at the sample site in the prediction result and the actual time when the sample vehicle arrives at the sample site. 6. The method of claim 5, wherein the time loss is weighted more than the position loss. 7. The method according to claim 1, wherein the prediction model is a fusion model of position prediction and arrival time prediction. 8. The method of claim 1, further comprising: At the same time, the current position of the target vehicle predicted by the prediction model and the time when the target vehicle arrives at the target site are displayed. 9. A vehicle arrival time prediction system, comprising an acquisition module, a determination module and a prediction module; The obtaining module is used to obtain the position information of the target site and the first positioning information reported by the target vehicle; The determination module is used to determine the time interval from the first moment corresponding to the first positioning information to the current moment; The prediction module is used to predict the current position of the target vehicle at the current moment and the arrival of the target vehicle through a prediction model based on the position information of the target site, the first positioning information and the time interval. the time of the target site. 10. A device for predicting the arrival time of a vehicle, comprising a processor and a storage device, the storage device is used to store instructions, and when the processor executes the instructions, it realizes the following: described method.

Images