Apparatus for predicting traffic flow of road and method thereof

Abstract

The present disclosure relates to an apparatus for predicting traffic flow of a road and a method thereof. The device comprises a memory which stores traffic information in a preset time period according to dates; and a controller that divides the dates into a specified number of categories, determines a target category corresponding to a target date among the categories, and predicts a traffic flow of the target date based on traffic information of the dates included in the target category.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of priority from Korean Patent Application No. 10-2023-0112254 filed in the Korean Intellectual Property Office on August 25, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to a technology for predicting road traffic flow based on an artificial neural network model. Background Art

[0004] In general, artificial neural network (ANN) is the field of artificial intelligence, which is an algorithm that allows machines to learn by simulating human neural structures. Recently, artificial neural networks have been applied to image recognition, language recognition, natural language processing, etc., and have shown excellent results. Artificial neural networks include an input layer that receives input, a hidden layer for actual learning, and an output layer that returns the results of the operation. An artificial neural network including multiple hidden layers is called a deep neural network (DNN), which is also a kind of artificial neural network.

[0005] Artificial neural networks allow computers to learn by themselves based on data. When trying to use artificial neural networks to solve a problem, it may be necessary to prepare a suitable artificial neural network model and the data to be analyzed. The artificial neural network model used to solve the problem is trained based on the data. Before training the model, it may be necessary to first divide the data into two types. That is, the data should be divided into a training data set and a validation data set. The training data set is used to train the model, and the validation data set is used to verify the performance of the model.

[0006] There are various reasons for validating an artificial neural network model. An artificial neural network developer tunes the model by modifying the model's hyperparameters based on the model's validation results. Furthermore, model validation is performed to select a suitable model from among various models. The reasons why model validation may be necessary are explained in more detail below.

[0007] The first is prediction accuracy. As a result, the purpose of artificial neural networks is to achieve good performance on out-of-sample data that was not used for training. Therefore, after creating a model, it is necessary to check how the model will perform on out-of-sample data. However, since the model should not be validated using the training dataset, the accuracy of the model should be measured using a validation dataset that is separate from the training dataset.

[0008] The second is to improve the performance of the model by adjusting the model. For example, overfitting can be prevented. Overfitting refers to the model being overtrained on the training data set. For example, when the training accuracy is high but the validation accuracy is low, overfitting may be suspected. In addition, overfitting can be understood in more detail through the training loss and validation loss. When overfitting occurs, it may be necessary to prevent overfitting to increase validation accuracy. Overfitting can be prevented by using schemes such as regularization or dropout.

[0009] Meanwhile, conventional techniques for predicting traffic flow on roads determine a day of the week for a date to be predicted, collect traffic speeds corresponding to the same past date of the day of the week, and predict traffic flow on the date to be predicted based on the traffic speeds.

[0010] Since the conventional technology predicts the traffic flow on a date to be predicted based only on the traffic speed on the same day of the past week, it is difficult to predict the traffic flow with high accuracy and also difficult to predict various traffic conditions. Summary of the invention

[0011] One aspect of the present disclosure provides an apparatus and method for predicting the traffic flow of a road, which can predict the traffic flow of a target date with high accuracy by: storing traffic information within a preset time period by date; dividing the dates into a specified number of categories; determining a target category corresponding to the target date among the categories; and predicting the traffic flow of the target date based on the traffic information of the dates included in the target category.

[0012] Another aspect of the present disclosure provides an apparatus and method for predicting traffic flow on a road, which can predict the traffic flow on a target date with high accuracy by: storing traffic information within a preset time period by date; dividing the dates into a specified number of categories; determining a target category corresponding to the target date among these categories; and predicting the future traffic flow on the target date based on the current traffic information on the target date and the traffic information on the dates included in the target category.

[0013] Another aspect of the present disclosure provides an apparatus and method for predicting traffic flow on a road, which can predict the afternoon traffic flow on a target date with high accuracy in the following manner: storing morning traffic information and afternoon traffic information during a preset time period by date, dividing the dates into a specified number of categories, determining a target category corresponding to the target date among these categories, and predicting the afternoon traffic flow on the target date based on the morning traffic information on the target date and the morning traffic information of the dates included in the target category.

[0014] Another aspect of the present disclosure provides an apparatus and method for predicting the traffic flow of a road, which can predict the afternoon traffic flow of a target date with high accuracy in the following manner: storing morning traffic information and afternoon traffic information during a preset time period by date, dividing the dates into a specified number of categories, determining a target category corresponding to the target date among the categories, determining a representative date that is most similar to the morning traffic information of the target date among the morning traffic information of the dates included in the target category, and predicting the afternoon traffic flow of the target date based on the afternoon traffic information of the representative date.

[0015] The technical problems to be solved by the present disclosure are not limited to the above problems, and those skilled in the art to which the present disclosure belongs will clearly understand any other technical problems not mentioned herein from the following description. In addition, it can be easily understood that the objects and advantages of the present disclosure can be achieved by the units and combinations thereof described in the claims.

[0016] According to one aspect of the present disclosure, a device for predicting traffic flow on a road includes: a memory that stores traffic information within a preset time period by date; and a controller that divides the date into a specified number of categories, determines a target category corresponding to a target date among the categories, and predicts the traffic flow on the target date based on the traffic information on the date included in the target category.

[0017] In some implementations, the controller may obtain traffic information of the target date from a traffic information server; and predict future traffic flow of the target date based on the traffic information of the target date and traffic information by dates included in the target category.

[0018] In some implementations, the memory may store morning traffic information and afternoon traffic information during a preset time period by date.

[0019] In some implementations, the controller may obtain morning traffic information of the target date from a traffic information server, and predict afternoon traffic flow of the target date based on the morning traffic information of the target date and morning traffic information of dates included in the target category.

[0020] In some implementations, the controller may obtain morning traffic information of the target date from a traffic information server, determine a representative date that is most similar to the morning traffic information of the target date among the morning traffic information of the dates included in the target category, and predict the afternoon traffic flow of the target date based on the afternoon traffic information of the representative date.

[0021] In some implementations, the controller may determine, based on the comparative learning model, a representative date that is most similar to the morning traffic information of the target date among the morning traffic information of the dates included in the target category.

[0022] In some implementations, the traffic information may include at least one of the following: traffic volume on each road, traffic speed on each road, travel time on each road, and combinations thereof.

[0023] In some implementations, the categories may include at least one of the following: a non-weekend public holiday, the day before a non-weekend public holiday, the day after a non-weekend public holiday, the day before two consecutive public holidays, the first day of two consecutive public holidays, the last day of two consecutive public holidays, the day after two consecutive public holidays, the day before five consecutive public holidays, the first day of five consecutive public holidays, the last day of five consecutive public holidays, the day after five consecutive public holidays, a weekday, and combinations thereof.

[0024] According to another aspect of the present disclosure, a method for predicting traffic flow on a road includes: a memory storing traffic information within a preset time period by date; a controller dividing the date into a specified number of categories; the controller determining a target category corresponding to a target date among the categories; and the controller predicting the traffic flow on the target date based on the traffic information on the date included in the target category.

[0025] In some implementations, predicting the traffic flow on the target date may include: obtaining, by the controller, traffic information on the target date from a traffic information server, and predicting, by the controller, future traffic flow on the target date based on the traffic information on the target date and traffic information on dates included in the target category.

[0026] In some implementations, storing the traffic information by date may include: storing, by the memory, the morning traffic information and the afternoon traffic information during a preset time period by date.

[0027] In some implementations, predicting the traffic flow on the target date may include: obtaining, by a controller, morning traffic information on the target date from a traffic information server; and predicting, by the controller, afternoon traffic flow on the target date based on the morning traffic information on the target date and morning traffic information on dates included in the target category.

[0028] In some implementations, predicting the traffic flow on a target date may include: obtaining, by a controller, morning traffic information on the target date from a traffic information server; determining, by the controller, a representative date that is most similar to the morning traffic information on the target date among the morning traffic information on dates included in the target category; and predicting, by the controller, the afternoon traffic flow on the target date based on the afternoon traffic information on the representative date.

[0029] In some implementations, determining the representative date that is most similar to the morning traffic information of the target date may include: determining, by the controller based on a comparative learning model, a representative date that is most similar to the morning traffic information of the target date among the morning traffic information of the dates included in the target category. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a block diagram showing an embodiment of an apparatus for predicting traffic flow of a road.

[0031] Figure 2 is a diagram showing an embodiment of a comparative learning model provided in an apparatus for predicting a traffic flow of a road.

[0032] Figure 3 : is a diagram showing an embodiment of a learning process of a comparative learning model provided in an apparatus for predicting a traffic flow of a road.

[0033] Figure 4 : is a diagram showing an embodiment of a similarity prediction process of a comparative learning model provided in an apparatus for predicting a traffic flow of a road.

[0034] Figure 5 : is a diagram showing an embodiment of traffic information by date of a first category stored in a memory provided in an apparatus for predicting a traffic flow of a road.

[0035] Figure 6 : is a diagram showing an embodiment of traffic information by date of a second category stored in a memory provided in an apparatus for predicting a traffic flow of a road.

[0036] Figure 7 : is a diagram showing an embodiment of traffic information by date of a third category stored in a memory provided in an apparatus for predicting a traffic flow of a road.

[0037] Figure 8 : is a diagram showing an embodiment of traffic information by date of a fourth category stored in a memory provided in an apparatus for predicting a traffic flow of a road.

[0038] Fig. 9 is a flow chart illustrating an exemplary method of predicting traffic flow for a road.

[0039] Fig.10 is a block diagram illustrating an embodiment of a computing system for performing a method of predicting traffic flow for a road. DETAILED DESCRIPTION

[0040] Figure 1 is a block diagram showing an apparatus for predicting traffic flow on a road.

[0041] like Figure 1 As shown, the apparatus 100 for predicting the traffic flow of a road may include a memory 10, a communication device 20, an output device 30, and a controller 40. In this case, depending on a scheme for implementing the apparatus 100 for predicting the traffic flow of a road, the components may be combined with each other to be implemented as one, or some components may be omitted.

[0042] Regarding each component, first, the memory 10 may store traffic information within a preset time period by date.

[0043] The memory 10 may store various logics, algorithms, and programs required in the following processing: dividing dates into a specified number of categories, determining a target category corresponding to a target date among these categories, and predicting the traffic flow of the target date based on traffic information of the dates included in the target category.

[0044] The memory 10 can store various logics, algorithms and programs required in the following processing: dividing dates into a specified number of categories, determining a target category corresponding to a target date among these categories, and predicting the future traffic flow of the target date based on the traffic information of the dates included in the target category and the current traffic information of the target date.

[0045] The memory 10 may store morning traffic information and afternoon traffic information by date during a preset time period.

[0046] The memory 10 can store various logics, algorithms and programs required in the following processing: dividing dates into a specified number of categories, determining a target category corresponding to a target date among these categories, and predicting the afternoon traffic flow on the target date based on the morning traffic information on the target date and the morning traffic information of the dates included in the target category.

[0047] The memory 10 can store various logics, algorithms and programs required in the following processing: dividing dates into a specified number of categories, determining a target category corresponding to a target date among these categories, determining a representative date that is most similar to the morning traffic information of the target date among the morning traffic information of the dates included in the target category, and predicting the afternoon traffic flow of the target date based on the afternoon traffic information of the representative date.

[0048] The communication device 20 , which is a module for providing a communication interface with the traffic information server 200 , may include at least one of a mobile communication module, a wireless Internet module, and a short-range communication module.

[0049] The mobile communication module can communicate with the traffic information server 200 through a mobile communication network, which is constructed according to the technical standard or communication scheme of mobile communication (for example, Global System for Mobile Communications (GSM), Code Division Multiple Access (CDMA), Code Division Multiple Access 2000 (CDMA2000), Enhanced Voice Data Optimization or Enhanced Voice Data Only (EV-DO), Wideband CDMA (WCDMA), High Speed ​​Downlink Packet Access (HSDPA), High Speed ​​Uplink Packet Access (HSUPA), Long Term Evolution (LTE), Advanced Long Term Evolution (LTEA), etc.).

[0050] The wireless Internet module, which is a module for wireless Internet access, can communicate with the traffic information server 200 through wireless LAN (WLAN), wireless fidelity (Wi-Fi), Wi-Fi direct, Digital Living Network Alliance (DLNA), wireless broadband (WiBro), Worldwide Interoperability for Microwave Access (WiMAX), High-Speed ​​Downlink Packet Access (HSDPA), High-Speed ​​Uplink Packet Access (HSUPA), Long Term Evolution (LTE), Advanced Long Term Evolution (LTE-A), etc.

[0051] Short-range communication module can be used by using Bluetooth TM , Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, Near Field Communication (NFC), and Wireless Universal Serial Bus (USB) technologies to support short-range communication with the traffic information server 200.

[0052] The output device 30 may output in video or audio the traffic flow on the target date predicted by the controller 40. To this end, the output device 30 may include a display module and an audio output module.

[0053] Such a display module can display (output) information processed in the vehicle multimedia system. For example, in navigation mode, a user interface (UI) or a graphical user interface (GUI) related to driving, such as a map, speed, direction, distance indication, etc. related to the current location, destination, route, etc., can be displayed. In black box mode or shooting mode, a captured image, UI or GUI can be displayed.

[0054] In addition, the display module may include at least one of a liquid crystal display (LCD), a thin film transistor liquid crystal display (TFT LCD), an organic light emitting diode (OLED), a flexible display, and a 3D display.

[0055] When the display module and the touch sensor form a mutual layer structure (hereinafter, referred to as a 'touch screen'), the display module may be used as an input device in addition to an output device.

[0056] The audio output module can output audio data in a multimedia file playback mode or a broadcast reception mode, or output audio data stored in a storage. The audio output module can output audio signals related to functions performed in the vehicle multimedia system (e.g., warning sounds, notification sounds, route guidance voices, etc.). For example, the audio output module can include a receiver, a speaker, a buzzer, etc.

[0057] The controller 40 may perform overall control so that each component performs its function. The controller 40 may be implemented in the form of hardware or software, or may be implemented in a combination of hardware and software. Preferably, the controller 40 may be implemented as a microprocessor, but is not limited thereto.

[0058] The controller 40 may classify the date into a specified number of categories based on the traffic information by date within a preset time period stored in the memory 10, determine a target category corresponding to a target date (i.e., a date to be predicted) among the categories, and predict the traffic flow of the target date based on the traffic information of each date included in the target category. In this case, the traffic information may include the traffic volume on each road, the traffic speed on each road, the travel time on each road, etc. In addition, the road refers to a road on which the entry and exit of vehicles are monitored, such as a highway, a toll road, etc.

[0059] In this case, the controller 40 can divide the dates within the preset time period into the following categories: such as a public holiday on a non-weekend (Saturday and Sunday), the day before the non-weekend public holiday, the day after the non-weekend public holiday, the day before two consecutive public holidays, the first day of two consecutive public holidays, the last day of two consecutive public holidays, the day after two consecutive public holidays, the day before five consecutive public holidays, the first day of five consecutive public holidays, the last day of five consecutive public holidays, the day after five consecutive public holidays, working days, and the like.

[0060] In addition, the controller 40 can obtain the current traffic information of the target date from the traffic information server 200 through the communication device 20, and predict the future traffic flow of the target date based on the traffic information of the date included in the target category and the current traffic information of the target date. In this case, the current traffic information of the target date refers to the traffic information of the target date stored in the traffic information server 200.

[0061] Meanwhile, the controller 40 may classify the date into a specified number of categories for the morning traffic information and the afternoon traffic information of the date within the preset time period stored in the memory 10, determine a target category corresponding to the target date among the categories, and predict the afternoon traffic flow of the target date based on the morning traffic information of the target date and the morning traffic information of the date included in the target category. In this case, the controller 40 may obtain the morning traffic information of the target date from the traffic information server 200.

[0062] In this case, the controller 40 may determine a representative date that is most similar to the morning traffic information of the target date among the morning traffic information of the dates included in the target category, and predict the afternoon traffic flow of the target date based on the afternoon traffic information of the representative date. Figure 2 As shown, the controller 40 may determine a representative date that is most similar to the morning traffic information of the target date among the morning traffic information of the dates included in the target category based on the comparative learning model.

[0063] Figure 2 is a diagram showing an embodiment of a comparative learning model provided in an apparatus for predicting a traffic flow of a road.

[0064] like Figure 2 As shown, the contrastive learning model 11, which is an artificial neural network model that outputs the similarity between two samples (i.e., the probability of similarity), can be trained to output "1" for two samples that are similar to each other (positive pair), and output "0 (zero)" for two different samples (negative pair), thereby determining the similarity between the two samples.

[0065] Figure 3 : is a diagram showing an embodiment of a learning process of a comparative learning model provided in an apparatus for predicting a traffic flow of a road.

[0066] like Figure 3 As shown, the controller 40 can convert two similar samples is input to the contrastive learning model 11 and the contrastive learning model 11 is trained to output a probability of "1". In addition, the controller 40 can n ,x m ) is input into the contrastive learning model 11, and the contrastive learning model 11 is trained to output the probability "0 (zero)".

[0067] Figure 4 : is a diagram showing an embodiment of a similarity prediction process of a comparative learning model provided in an apparatus for predicting a traffic flow of a road.

[0068] like Figure 4 As shown, the controller 40 can predict the similarity between two samples based on the comparative learning model 11 that has completed learning. That is, the controller 40 can input the morning traffic information of the target date and the morning traffic information of the dates included in the target category into the comparative learning model 11.

[0069] For example, when it is assumed that the morning traffic information by the date included in the target category is A, B, and C, and the morning traffic information of the target date is D, the controller 40 can obtain three similarities by sequentially inputting a total of three pairs (AD, BD, and CD) into the contrastive learning model 11. In this case, the contrastive learning model 11 can predict the similarity between A and D by comparing A and D, predict the similarity between B and D by comparing B and D, and predict the similarity between C and D by comparing C and D.

[0070] Figure 5 : is a diagram showing an embodiment of traffic information by date of a first category stored in a memory provided in an apparatus for predicting a traffic flow of a road. Figure 5 Traffic speed is shown among traffic information.

[0071] exist Figure 5 , reference numeral 510 represents the afternoon traffic speed on the first road on October 10, 2019, reference numeral 520 represents the afternoon traffic speed on the first road on December 26, 2019, and reference numeral 530 represents the afternoon traffic speed on the first road on January 2, 2020. In this case, the first category refers to the day after a public holiday that is not a weekend day, and includes October 10, 2019, December 26, 2019, and January 2, 2020.

[0072] like Figure 5 As shown, since the traffic speed patterns 511, 521, and 531 included in the first category are similar to each other, it can be understood that the traffic speed patterns 511, 521, and 531 are well matched with each other according to the properties of the first category.

[0073] Figure 6 is a diagram showing an embodiment of traffic information by date of the second category stored in a memory provided in an apparatus for predicting a traffic flow of a road, and shows a traffic speed among the traffic information.

[0074] exist Figure 6, reference numeral 610 represents the afternoon traffic speed on the first road on August 15, 2019, reference numeral 620 represents the afternoon traffic speed on the first road on October 3, 2019, and reference numeral 630 represents the afternoon traffic speed on the first road on October 9, 2019. In this case, the second category refers to a public holiday on a non-weekend day, and includes August 15, 2019, October 3, 2019, and October 9, 2019.

[0075] like Figure 6 As shown, since the traffic speed patterns 611, 621, and 631 included in the second category are similar to each other, it can be understood that the traffic speed patterns 611, 621, and 631 are well matched with each other according to the properties of the second category.

[0076] Figure 7 is a diagram showing an embodiment of traffic information by date of the third category stored in a memory provided in an apparatus for predicting a traffic flow of a road, and shows a traffic speed among the traffic information.

[0077] exist Figure 7 , reference numeral 710 represents the afternoon traffic speed on the second road on February 2, 2019, reference numeral 720 represents the afternoon traffic speed on the second road on September 30, 2020, and reference numeral 730 represents the afternoon traffic speed on the second road on September 18, 2021. In this case, the third category refers to the first day of five consecutive public holidays, and includes February 2, 2019, September 30, 2020, and September 18, 2021.

[0078] like Figure 7 As shown, since the traffic speed patterns 711, 721, and 731 included in the third category are similar to each other, it can be understood that the traffic speed patterns 711, 721, and 731 are well matched with each other according to the properties of the third category.

[0079] Figure 8 is a diagram showing an embodiment of traffic information by date of a fourth category stored in a memory provided in an apparatus for predicting a traffic flow of a road, and shows a traffic speed among the traffic information.

[0080] exist Figure 8 , reference numeral 810 represents the morning traffic speed on the third road on October 4, 2020, and reference numeral 820 represents the morning traffic speed on the third road on September 22, 2021. In this case, the fourth category refers to the last day of five consecutive public holidays, and includes October 4, 2020 and September 22, 2021.

[0081] like Figure 8 As shown, since the traffic speed patterns 811 and 821 included in the fourth category are similar to each other, it can be understood that the traffic speed patterns 811 and 821 are well matched to each other according to the properties of the fourth category.

[0082] Fig. 9 is a flowchart showing a method of predicting traffic flow on a road.

[0083] First, at 901 , the memory 10 stores traffic information by date within a preset time period.

[0084] Then, at 902, the controller 40 divides the dates into a specified number of categories.

[0085] Then, at 903 , the controller 40 determines a target category corresponding to the target date among the categories.

[0086] Then, at 904 , the controller 40 predicts the traffic flow of the target date based on the traffic information by date included in the target category.

[0087] Fig.10 is a block diagram illustrating a computing system for executing a method for predicting traffic flow of a road.

[0088] Reference Fig.10 , the method for predicting the traffic flow of a road can be implemented by a computing system 1000. The computing system 1000 may include at least one processor 1100, a storage 1300, a user interface input device 1400, a user interface output device 1500, a memory 1600, and a network interface 1700 connected via a bus 1200.

[0089] The processor 1100 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in the storage 1300 and / or the memory 1600. The storage 1300 and the memory 1600 may include various types of volatile or non-volatile storage media. For example, the storage 1300 may include a ROM (Read Only Memory) 1310 and a RAM (Random Access Memory) 1320.

[0090] Therefore, the process of the method or algorithm described in conjunction with the implementation of the present disclosure may be directly implemented by hardware, software modules, or a combination thereof executed by the processor 1100. The software module may reside in a storage medium (i.e., storage 1300 and / or memory 1600), such as RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, solid state drive (SSD), detachable disk, or CD-ROM. An exemplary storage medium is coupled to the processor 1100, and the processor 1100 can read information from the storage medium and can write information in the storage medium. In another approach, the storage medium may be integrated with the processor 1100. The processor 1100 and the storage medium may reside in an application specific integrated circuit (ASIC). The ASIC may reside in a user terminal. In another approach, the processor and the storage medium may reside in a user terminal as separate components.

[0091] According to an implementation of the present disclosure, the traffic flow on a target date can be predicted with high accuracy by storing traffic information within a preset time period by date, dividing these dates into a specified number of categories, determining a target category corresponding to the target date among these categories, and predicting the traffic information on the target date based on the traffic information on the dates included in the target category.

Claims

1. A device for predicting traffic flow on a road, the device comprising: a memory configured to store traffic information classified by predetermined date; as well as The controller is configured as: Divide the date into a specified number of categories, determining, among said categories, a target category corresponding to a target date, and Traffic volume for the target date is determined based on traffic information within the traffic information associated with one or more dates included in the target category.

2. The device according to claim 1, wherein: The controller is configured to: receiving traffic information on the target date from a traffic information server, and A future traffic volume for the target date is determined based on the traffic information for the target date and the traffic information associated with the one or more dates included in the target category.

3. The device according to claim 1, wherein: The memory is configured to store morning traffic information and afternoon traffic information classified by the predetermined date.

4. The device according to claim 3, wherein: The controller is configured to: The morning traffic information of the target date is received from a traffic information server, and the afternoon traffic flow of the target date is determined based on the morning traffic information of the target date and the morning traffic information associated with the one or more dates included in the target category within the morning traffic information.

5. The device according to claim 3, wherein: The controller is configured to: receiving the morning traffic information of the target date from a traffic information server, determining a representative date from the one or more dates included in the target category, the representative date being associated with information most similar to the morning traffic information of the target date among the morning traffic information associated with the one or more dates included in the target category, and The afternoon traffic flow of the target date is determined based on the afternoon traffic information of the representative date.

6. The device according to claim 5, wherein: The controller is configured to determine the representative date based on a contrastive learning model.

7. The device according to claim 1, wherein: The traffic information includes at least one of the following: a traffic volume on each road, a traffic speed on each road, and a travel time on each road.

8. The device according to claim 1, wherein: The categories include at least one of the following: a non-weekend public holiday, the day before a non-weekend public holiday, the day after a non-weekend public holiday, the day before two consecutive public holidays, the first day of two consecutive public holidays, the last day of two consecutive public holidays, the day after two consecutive public holidays, the day before five consecutive public holidays, the first day of five consecutive public holidays, the last day of five consecutive public holidays, the day after five consecutive public holidays, and a working day.

9. A method for predicting traffic flow on a road, the method comprising: The memory stores the traffic information classified by predetermined date; The controller divides the date into a specified number of categories; determining, by the controller, a target category corresponding to a target date among the categories; as well as Traffic volume for the target date is determined by the controller based on traffic information within the traffic information associated with one or more dates included in the target category.

10. The method according to claim 9, wherein: Determining traffic volumes for the target date includes: receiving, by the controller, traffic information on the target date from a traffic information server; and A future traffic volume for the target date is determined by the controller based on the traffic information for the target date and the traffic information associated with the one or more dates included in the target category.

11. The method according to claim 9, wherein: Storing the traffic information includes: The memory stores morning traffic information and afternoon traffic information classified by the predetermined date.

12. The method according to claim 11, wherein: Determining traffic volumes for the target date includes: receiving, by the controller, morning traffic information of the target date from a traffic information server; and The controller determines afternoon traffic flow on the target date based on the morning traffic information on the target date and morning traffic information within the morning traffic information associated with the one or more dates included in the target category.

13. The method according to claim 11, wherein: Determining traffic volumes for the target date includes: The controller receives the morning traffic information of the target date from a traffic information server; determining, by the controller, a representative date from the one or more dates included in the target category, the representative date being associated with information most similar to the morning traffic information of the target date among morning traffic information associated with the one or more dates included in the target category; and The afternoon traffic flow of the target date is determined by the controller based on the afternoon traffic information of the representative date.

14. The method according to claim 13, wherein: Determining said representative date includes: The representative date is determined by the controller based on a contrastive learning model.

15. The method according to claim 9, wherein: The traffic information includes at least one of the following: a traffic volume on each road, a traffic speed on each road, and a travel time on each road.

16. The method according to claim 9, wherein: The categories include at least one of the following: a non-weekend public holiday, the day before a non-weekend public holiday, the day after a non-weekend public holiday, the day before two consecutive public holidays, the first day of two consecutive public holidays, the last day of two consecutive public holidays, the day after two consecutive public holidays, the day before five consecutive public holidays, the first day of five consecutive public holidays, the last day of five consecutive public holidays, the day after five consecutive public holidays, and a working day.

Images