TWI893616B - Road traffic condition prediction method, system, device, and storage medium - Google Patents

Abstract

A road traffic condition prediction method, system, device and storage medium is provided, the method includes: collecting the vehicle condition data of each vehicle and the road condition data of each road area at each prediction time point. Determining the current road area of each vehicle based on the vehicle condition data of each vehicle, and inputting the vehicle condition data of each vehicle and the road condition data of the current road area into a travel prediction model to obtain the predicted travel data of each vehicle predicted by the travel prediction model. Inputting the road condition data of each road area, the vehicle condition data of each vehicle, and predicted travel data into a condition prediction model to obtain the predicted traffic condition data of each road area at the next prediction time point by the condition prediction model outputted. Accurate prediction of road traffic conditions in the future is achieved.

Description

Road traffic state prediction method, system, device and storage medium

The present invention relates to the field of data processing technology, and in particular to a method, system, device, and storage medium for predicting road traffic conditions using a plurality of vehicles located in a plurality of road areas.

The growing number of motor vehicles in cities has dramatically increased traffic pressure, leading to frequent traffic jams in some areas. This is particularly true during rush hour, when many areas experience severe congestion. Traffic jams can easily lead to road rage, traffic violations, and overtaking, resulting in an increase of over 10% in traffic accidents and significant traffic congestion costs. The costs of traffic congestion primarily include time and fuel consumption. A three-minute stoppage uses as much fuel as one kilometer of driving.

Real-time traffic conditions cannot efficiently guide travel. For example, a transit point on the way to work may not be congested when leaving, but may be congested by the time you arrive. Existing methods for predicting traffic congestion, such as modeling and machine learning, have many shortcomings in practical application. Some can only predict traffic flow within a small area, within a short period of 15 minutes, and the prediction accuracy is low.

It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention and may include information that does not constitute prior art known to ordinary technicians in this field.

To address the problems in the existing technology, the present invention aims to provide a road traffic status prediction method, system, device, and storage medium to accurately predict road traffic conditions within a certain period of time in the future.

In the road traffic state prediction method of the present invention, relevant data for prediction is first collected to determine the current road area of each vehicle. The vehicle's predicted travel data is then predicted based on the vehicle's state data and the state data of the road area corresponding to the vehicle. The traffic state of each road area is then predicted based on the road area state data, the vehicle state data, and the predicted travel data obtained in the previous step, thereby achieving a more accurate prediction of road traffic conditions in the future. The road traffic status prediction method can be deployed on a cloud server and communicate with the vehicle's central control system, roadside units, smart traffic lights, and other systems. After predicting the road traffic status, the method promptly sends congestion warning information to vehicle users or other devices, enabling users to choose more appropriate travel routes and reducing road congestion.

To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention.

S100, S200, S300, S400, S110~S190, S190’, S510~S580: Steps

M100: Data Collection Module

M200: Road detection module

M300: Travel Prediction Module

M400: Status Prediction Module

600: Electronic equipment

610: Processing unit

620: Storage Unit

6201: RAM

6202: Cache memory

6203:ROM

6204: Programs/Utilities

6205: Programming Module

630: Bus

640: Display unit

650 I/O: Interface

660: Network interface card

700:External device

Figure 1 is a flow chart of the road traffic status prediction method according to the first embodiment of the present invention.

Figure 2 is a flowchart of determining the current driving route of a vehicle according to an embodiment of the present invention.

Figure 3 is a flow chart of selecting a recommended driving route according to an embodiment of the present invention.

Figure 4 is a schematic diagram of the structure of a road traffic status prediction system according to an embodiment of the present invention.

Figure 5 is a schematic diagram of the structure of a road traffic status prediction device according to an embodiment of the present invention.

The following is an explanation of the implementation of the "road traffic status prediction method, system, device and storage medium" disclosed in the present invention through specific concrete embodiments. Technical personnel in this field can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments. The details in this specification can also be modified and changed based on different viewpoints and applications without departing from the concept of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted in actual size. Please note in advance. The following implementation will further explain the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention. In addition, the term "or" used in this document may include any one or more combinations of the related listed items as the case may be.

[First embodiment]

As shown in FIG1 , an embodiment of the present invention provides a road traffic status prediction method for predicting road traffic status within a period of time in the future. The method includes the following steps:

S100: At each predicted time point, collect status data for each vehicle and each road area. In this embodiment, the vehicle status data includes historical route data and real-time driving data. The historical route data includes at least the departure time, starting point, end point, roads passed, intersections, and vehicle speed for each historical route of the vehicle. The historical route data is obtained from a historical database.

In this embodiment, the real-time vehicle driving data includes at least the vehicle's position, speed, and current driving route at the current prediction time. The real-time vehicle driving data can be obtained by obtaining vehicle-reported data and/or monitoring data from roadside units. A road area here refers to a road section or intersection. The state data collected for each road area includes the state data collected for each road section and intersection within a defined prediction range.

In this embodiment, the road area status data includes at least static data and real-time status data. The static data includes the road number corresponding to the road area, the road area number, the road area traffic direction number, the road length, the road area length, and the road area speed limit. The real-time status data includes information such as the traffic volume and speed at the current forecast time, the vehicle distribution data at the current forecast time, the road speed at the current forecast time, whether there is a traffic jam, the real-time green light duration in each direction at the intersection, and the vehicle queues in each direction at the intersection. The queues for vehicles in each direction at the intersection can be calculated using the location data reported by each vehicle. Changes in the queues are then calculated based on green lights and the speed limit in the road area.

S200: Determine the road area currently located by each vehicle based on the status data of each vehicle. Specifically, the road area currently located by each vehicle is determined based on the vehicle position at the current predicted time point and the coordinate range of each road area.

S300: Inputting the state data of each vehicle and the state data of the road area it is currently in into a travel prediction model to obtain predicted travel data for each vehicle as predicted by the travel prediction model. In this embodiment, the predicted travel data for each vehicle includes at least the predicted travel route, travel time, and uncongested vehicle speed. The uncongested vehicle speed is the vehicle's usual speed under normal driving conditions.

S400: Input the state data of each road area, the state data of each vehicle, and the predicted travel data into a state prediction model, and obtain the predicted traffic state data for each road area at the next predicted time point output by the state prediction model.

In this embodiment, the predicted traffic status data includes vehicle information, travel time, and predicted vehicle speed for each road zone at the next predicted time point. The vehicle information for each road zone at the next predicted time point is the predicted vehicle number for each road zone at the next predicted time point, the travel time is the predicted time required to pass through the road zone, and the predicted vehicle speed is the predicted speed of the vehicle when passing through the road zone.

In the road traffic status prediction method of the present invention, first, relevant data for prediction is collected through step S100, and the current road area of each vehicle is determined through step S200. The predicted travel data of the vehicle is predicted based on the vehicle status data and the status data of the road area corresponding to the vehicle through step S300. Finally, the traffic status of each road area is predicted through step S400 based on the road area status data, the vehicle status data, and the predicted travel data obtained through step S300, thereby achieving a more accurate prediction of the road traffic status in the future period. This road traffic status prediction method can be deployed on a cloud server and communicate with the vehicle's central control system, roadside units, smart traffic lights, and other systems. Once the traffic status is predicted, this method promptly sends congestion warning information to vehicle users or other devices, enabling users to choose more appropriate routes and reducing road congestion.

In this embodiment, the prediction time range can be divided into multiple prediction time periods, each of which includes multiple prediction time points set at intervals. At every preset prediction interval, the road traffic conditions for the next prediction time period are predicted, wherein the road traffic conditions at each prediction time point within the next prediction time period are predicted. For example, if the preset interval is set to 1 minute and the prediction time period is 1 hour, a prediction time point is set every 3 seconds within the 1 hour. In actual applications, the traffic status prediction process is initiated every minute, first predicting traffic conditions for the next 3 seconds, then 6 seconds, 9 seconds, and 12 seconds, until the traffic conditions for the next hour are predicted, at which point the current prediction period ends.

In this embodiment, after obtaining the predicted traffic state data for each road area at the next predicted time point output by the state prediction model in step S400, the following steps are also included:

Based on the predicted traffic status data for each road area, congestion areas and congestion durations are determined, and congestion warning information is sent to the corresponding vehicles. Here, the congestion area is, for example, a road area determined to meet severe congestion conditions based on the predicted traffic status data, and the congestion duration is the predicted duration of the congestion. The corresponding vehicles are those currently in the congested area or about to enter the congested area. By sending timely congestion warning information, vehicles can be prompted to change routes as quickly as possible, thereby reducing congestion.

By adopting the road traffic state prediction method of the present invention, two machine learning models, a travel prediction model and a state prediction model, such as a convolutional neural network model, are used for automatic learning and prediction. Among them, the travel prediction model can predict the current driving route of each vehicle, and the state prediction model can predict the vehicle distribution and current traffic capacity of each road and intersection, and predict the future congestion condition of each road. Specifically, it can accurately predict the distribution of all vehicles within a predicted time period, accurately predict the vehicle queues on all roads and intersections within a predicted time period, and accurately predict whether a specific road or intersection will be congested within a predicted time period. The predicted congestion times and locations are then released to users in advance, prompting them to choose less congested routes, thereby optimizing user travel routes and reducing congestion.

As shown in Figure 2, in step S100, when collecting the status data of each vehicle, the real-time driving data of each vehicle is collected, including the following steps:

S110: At the first prediction time point of a prediction time cycle, collect the vehicle's position, speed, and navigation information at the current prediction time point. The position, speed, and navigation information collected here can be obtained through communication with the vehicle's central control system.

S120: Determine whether the vehicle has a navigation route set. If so, S130: Set the set navigation route as the current driving route. Otherwise, S140: Determine whether the vehicle has a historical route that matches the vehicle's position at the current predicted time. If the vehicle has a historical route, S150: Query the vehicle's historical route data for the historical route corresponding to the vehicle's position at the current predicted time and use it as the current driving route. If the vehicle does not have a historical route, S160: Determine the vehicle's current road area based on the vehicle's current position and save it as the current driving route.

S170: From the second predicted time point onward, the vehicle's position at the current predicted time point is obtained. S180: Based on whether the vehicle's position at the current predicted time point matches the current driving route, a determination is made as to whether the current driving route needs to be revised. If revision is required, S190: Based on the vehicle's position at the current predicted time point, a historical route corresponding to the current predicted time point is retrieved and used as the revised current driving route. If revision is not required, S190': the route determined at the previous predicted time point is adopted as the current driving route.

Therefore, in this embodiment, when determining the vehicle's current driving route, the current route is prioritized based on the user-set navigation route. If the user has not set a navigation route, the vehicle's historical routes are used to match the vehicle's current location. If a match is found, the matching historical route is used as the current driving route. If a match is not found, the road area currently located by the vehicle is saved as the current driving route. At subsequent prediction time points, the vehicle's current driving route is continuously adjusted based on the vehicle's real-time location to ensure that the vehicle's current driving route is consistent with the actual driving route.

In this embodiment, in a prediction time period, at the first prediction time point, the status data of the vehicle and the status data of each road area are real-time data collected in real time. Except for the first prediction time point, at subsequent prediction time points, the status data of the vehicle and the status data of each road area are updated based on the predicted traffic status data obtained at the previous prediction time point. Therefore, not only the output data of the travel prediction model can be input and applied to the state prediction model, but the output data of the state prediction model at each prediction time point can also be used in the travel prediction model, making the travel prediction model and the state prediction model form a feedback closed loop, making the prediction more accurate.

In this embodiment, step S100 collects status data for each vehicle, including the following steps:

Collect status data for each vehicle currently traveling on the road. Obtain historical driving data for vehicles that have not yet traveled, and predict the vehicle's travel time based on this historical driving data. Predicting the vehicle's travel time based on this historical driving data can be achieved by using the aforementioned travel prediction model to implement travel pattern prediction, obtaining the predicted travel route, predicted travel time, predicted travel speed, etc. for the vehicles that have not yet traveled. Alternatively, a separate travel time prediction model can be used to predict the vehicle's travel time.

If the vehicle is predicted to depart within the current forecast time period based on the historical driving data of the vehicle that has not yet traveled, the status data of the vehicle that has not yet traveled is determined based on the predicted travel time and the historical driving data of the vehicle that has not yet traveled.

Therefore, in this embodiment, when predicting road traffic conditions, not only is every vehicle currently traveling on the road considered, but also vehicles that are not currently traveling but are predicted to travel within the current forecast period based on their historical driving data. Including these vehicles in the simulation calculations allows for a more comprehensive and accurate prediction of road congestion within the current forecast period.

In this embodiment, the predicted traffic state data includes the vehicle information, travel time and predicted speed distributed in each road area at the next predicted time point. When obtaining the predicted traffic state data through the state prediction model, any of the following implementation methods can be adopted: (1) When constructing and training the state prediction model, the output of the state prediction model is set to include the vehicle information, travel time and predicted speed distributed in each road area. The predicted traffic state data can be obtained directly from the output data of the state prediction model. When judging whether a road area is congested, when the travel time of a road area exceeds the preset travel time threshold and/or the predicted speed is less than the preset travel speed threshold, the road area is considered to meet the congestion criteria. Severe congestion conditions; (2) The output of the state prediction model is the vehicle information distributed in each road area. Based on the vehicle information distributed in each road area, it is judged whether each road area is congested. The predicted vehicle speed is divided into two situations: when the road area is congested and when it is not congested. When the road area is not congested, the predicted vehicle speed is the normal vehicle speed. When the road area is congested, the corresponding predicted vehicle speed can be obtained according to the degree of congestion. The queuing time and passing time of each road area are calculated based on the vehicle information distributed in each road area and the traffic light status.

In this embodiment, by accurately predicting the traffic conditions of each road, users can be better guided in route planning. As shown in Figure 3, in this embodiment, the road traffic condition prediction method further includes the following steps:

S510: Receive a route planning request from the user. For example, when the vehicle is started, the user sends a route planning request to the cloud server via the vehicle's central control platform. The route planning request includes the starting point, end point, current vehicle location, and normal vehicle speed.

S520: Obtain the optional driving routes for the vehicle bound by the user. The optional driving routes may include navigation routes set by the user, optional routes determined based on the starting and ending points, and the user's historical routes.

S530: Determine the congestion status of each optional driving route based on the predicted traffic status data for the road area of the optional driving routes. After determining the congestion status, the congestion status of each optional driving route can be sent to the user. For example, it is predicted that intersection 11 of Route 1 will be congested for 10 minutes in 30 minutes, while Route 2 will be completely unblocked within the next 30 minutes. The user can then choose the appropriate driving route. Alternatively, a recommended driving route can be automatically selected and provided to the user. This process further includes the following steps:

S540: Determine whether all available driving routes meet the severe congestion condition. If not, S550: Determine a recommended driving route based on the congestion status of each available driving route and send it to the user. If so, S560: Recommend parking locations within a preset distance based on the current location of the user's tied vehicle. A parking route is planned based on the parking locations and the current location of the user's tied vehicle, and sent to the user. These parking locations can be, for example, nearby suitable parking lots or leisure centers such as shopping malls and restaurants. The parking route guidance advises the user to avoid congested routes, thereby avoiding congestion.

S570: Continuously monitor the congestion status of the original optional driving route.

S580: If an alternative route no longer meets the severe congestion criteria exists, the alternative route is pushed to the user. After the user selects a specific route and begins driving, the system can continuously monitor the congestion status of the current route and promptly send warnings to the user when severe congestion occurs. Furthermore, if the current route becomes severely congested, the system can recommend alternative routes that are not severely congested, guiding the user to choose a more suitable and faster route.

As shown in FIG4 , an embodiment of the present invention further provides a road traffic status prediction system for implementing the road traffic status prediction method. The system includes:

The data collection module M100 is used to collect the status data of each vehicle and the status data of each road area at each predicted time point. The road judgment module M200 is used to determine the road area where each vehicle is currently located based on the status data of each vehicle. The travel prediction module M300 is used to input the status data of each vehicle and the status data of the road area in which it is currently located into the travel prediction model, and obtain the predicted travel data of each vehicle predicted by the travel prediction model. The state prediction module M400 is used to input the status data of each road area, the status data of each vehicle and the predicted travel data into the state prediction model, and obtain the predicted traffic state data of each road area at the next predicted time point output by the state prediction model.

In the road traffic status prediction system of the present invention, the functions of each module can be implemented using the specific implementation of the road traffic status prediction method described above, which will not be detailed here.

In the road traffic state prediction system of the present invention, the data collection module M100 first collects relevant data for prediction, the road judgment module M200 determines the current road area of each vehicle, and the travel prediction module M300 predicts the vehicle's predicted travel data based on the vehicle's state data and the state data of the vehicle's corresponding road area. The state prediction module M400 predicts the traffic state of each road area based on the road area state data, the vehicle's state data, and the predicted travel data obtained by the travel prediction module M300, thereby achieving a more accurate prediction of the road traffic state in the future period of time. This road traffic status prediction system can be deployed on a cloud server and communicate with the vehicle's central control system, roadside units, smart traffic lights, and other systems. After predicting traffic conditions using this road traffic status prediction method, it promptly sends congestion warning information to vehicle users or other devices, enabling users to choose more appropriate routes and reducing traffic congestion.

An embodiment of the present invention also provides a road traffic status prediction device, comprising a processor and a memory storing executable instructions for the processor. The processor is configured to execute the steps of the road traffic status prediction method by executing the executable instructions.

As shown in Figure 5, electronic device 600 is implemented as a general-purpose computing device. Components of electronic device 600 may include, but are not limited to, at least one processing unit 610, at least one storage unit 620, a bus 630 connecting various system components (including storage unit 620 and processing unit 610), and a display unit 640.

The storage unit stores program code that can be executed by the processing unit 610, causing the processing unit 610 to perform the steps of the various exemplary embodiments of the present invention described in the road traffic state prediction method section above. For example, the processing unit 610 may perform the steps shown in Figure 1.

The storage unit 620 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 6201 and/or a cache memory 6202, and may further include a read-only memory (ROM) 6203.

The storage unit 620 may also include a program/utility 6204 having a set (at least one) of program modules 6205. Such program modules 6205 include, but are not limited to, an operating system, one or more applications, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

Bus 630 can represent one or more of several types of bus structures, including a storage unit bus or storage unit controller, a peripheral bus, a graphics accelerator port, a processing unit, or a local bus using any of a variety of bus structures.

The electronic device 600 can also communicate with one or more external devices 700 (e.g., a keyboard, a pointing device, a Bluetooth device, etc.), one or more devices that enable a user to interact with the electronic device 600, and/or any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., a router, a modem, etc.). Such communication can be performed through an input/output (I/O) interface 650. Furthermore, the electronic device 600 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) via a network interface card 660. The network interface card 660 can communicate with other modules of the electronic device 600 via a bus 630. It should be understood that, although not shown, other hardware and/or software modules may be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

In the road traffic status prediction device, the program in the memory, when executed by the processor, implements the steps of the road traffic status prediction method. Therefore, the device can also achieve the technical effects of the above-mentioned road traffic status prediction method.

The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments. It should not be assumed that the specific implementation of the present invention is limited to this description. A person skilled in the art of the present invention may make simple deductions or substitutions without departing from the concept of the present invention, and these should be considered to fall within the scope of protection of the present invention.

S100, S200, S300, S400: Steps

Claims (9)

A method for predicting road traffic conditions comprises the following steps: At each prediction time point in a prediction time period, collecting first state data for each of a plurality of vehicles and second state data for each of a plurality of road areas; Determining a current road area for each of the vehicles based on each of the first state data; Inputting each of the first state data and the second state data for each of the currently located road areas into a travel prediction model to obtain predicted travel data for each of the vehicles as predicted by the travel prediction model; and Inputting the second state data for each of the road areas, the first state data for each of the vehicles, and the predicted travel data into a state prediction model to obtain predicted traffic state data for each of the road areas at the next prediction time point output by the state prediction model; Collecting the first state data for each vehicle further includes the following steps: At a first predicted time point in the predicted time period, collecting a first position, a first speed, and navigation information of the vehicle at the first predicted time point; Determining whether a navigation route is set for the vehicle; In response to determining that the navigation route is set for the vehicle, setting the set navigation route as a current driving route; and In response to determining that the navigation route is not set for the vehicle, querying a historical route corresponding to the first position at a current predicted time point based on historical route data of the vehicle, and setting the historical route as the current driving route. Collecting the first state data for each vehicle further includes the following steps: During the prediction time period, at a second prediction time point, a second position of the vehicle at the second prediction time point is obtained; and based on whether the second position of the vehicle at the second prediction time point matches the current driving route, a determination is made as to whether the current driving route needs to be corrected. The road traffic state prediction method of claim 1, wherein the first state data for each vehicle includes the historical route data and real-time driving data for each vehicle; the predicted travel data for each vehicle includes at least a predicted travel route, a travel time, and a non-congested vehicle speed; the second state data for each road area includes at least static data and real-time state data for each road area. A road traffic status prediction method as described in claim 1, wherein the predicted traffic status data of each road area includes distributed vehicle information, a travel time and a predicted vehicle speed for each road area within the next predicted time point. The road traffic state prediction method according to claim 2, wherein in one of the prediction time periods, except for the first prediction time point, the first state data of the vehicle and the second state data of the road area are updated according to the predicted traffic state data obtained at a previous prediction time point. The road traffic state prediction method of claim 1, wherein, after obtaining the predicted traffic state data for each road section at the next predicted time point output by the state prediction model, the method further comprises the following steps: Determining a congestion area and a congestion time based on the predicted traffic state data for each road section, and transmitting a congestion warning message to the corresponding vehicle. The road traffic state prediction method of claim 1, wherein collecting the first state data for each vehicle comprises the following steps: Collecting the first state data for each vehicle currently traveling on a road; Extracting historical driving data for a vehicle that has not yet traveled, and predicting a travel time for the vehicle that has not yet traveled based on the historical driving data; If the historical driving data of the vehicle that has not yet traveled predicts that the vehicle that has not yet traveled will depart within a current predicted time period, determining the first state data for the vehicle that has not yet traveled based on the predicted travel time and the historical driving data. The road traffic status prediction method of claim 1 further comprises: Receiving a route planning request from a user; Obtaining one or more optional driving routes associated with the vehicle by the user; Determining a congestion status for each optional driving route based on the predicted traffic status data for the road area associated with each optional driving route; Determining a recommended driving route based on the congestion status for each optional driving route and sending the recommended route to the user. The road traffic status prediction method as described in claim 7, wherein, after determining the congestion status of each of the optional driving routes, the method further comprises the following steps: Determining whether all of the optional driving routes meet a severe congestion condition; If so, recommending a parking location within a preset distance based on the current location of the vehicle bound to the user, planning a parking route based on the parking location and the current location of the vehicle bound to the user, and sending the parking route to the user; Continuously monitoring the congestion status of the original optional driving routes; If there is an alternative driving route that no longer meets the severe congestion condition, the alternative driving route that no longer meets the severe congestion condition is pushed to the user. A road traffic state prediction system, wherein the system is configured to implement the road traffic state prediction method described in any one of claims 1 to 8, the system comprising: a data acquisition module configured to collect the first state data of each vehicle and the second state data of each road area at each prediction time point; a road determination module configured to determine the current road area of each vehicle based on the first state data of each vehicle; a travel prediction module configured to input the first state data of each vehicle and the second state data of the current road area into a travel prediction model to obtain the predicted travel data of each vehicle as predicted by the travel prediction model; A state prediction module is configured to input the second state data for each road area, the first state data for each vehicle, and the predicted travel data into the state prediction model, and obtain the predicted traffic state data for each road area at the next predicted time point output by the state prediction model. Collecting the first state data for each vehicle further includes the following steps: At the first predicted time point in the prediction time period, collecting the first position, first speed, and navigation information of the vehicle at the first predicted time point; Determining whether the navigation route is set for the vehicle; In response to determining that the navigation route is set for the vehicle, setting the set navigation route as the current driving route; and In response to determining that the navigation route is not set for the vehicle, the historical route corresponding to the first position at the current predicted time point is retrieved from the historical route data of the vehicle and used as the current driving route. Collecting the first state data for each vehicle further includes the following steps: At the second predicted time point during the predicted time period, obtaining the second position of the vehicle at the second predicted time point; and Determining whether the current driving route needs to be corrected based on whether the second position of the vehicle at the second predicted time point matches the current driving route.