Apparatus and method for providing real-time traffic information

Abstract

The present disclosure relates to an apparatus and method for providing real-time traffic information. The apparatus includes a traffic information collection unit configured to collect traffic information including real-time travel speed and traffic congestion cause in real time as a user vehicle travels along a travel route. The apparatus further includes a traffic congestion section determination unit configured to determine a normal speed range based on the traffic information, and determine a section in which the real-time travel speed is lower than a lower limit of the normal speed range as a traffic congestion section. The apparatus further includes a traffic congestion section information provision unit configured to provide traffic congestion section information including at least one of a location of the traffic congestion section, a traffic congestion cause, and an expected delay time based on the traffic information and the traffic congestion section included in the travel route of the user vehicle.

Description

[0001] Cross-reference to related applications

[0002] This application claims the benefit and priority of Korean Patent Application No. 10-2024-0186674, filed on December 16, 2024, with the Korean Intellectual Property Office, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to apparatus and methods for providing real-time traffic information, and more specifically, to apparatus and methods for providing real-time traffic information collected in real time. Background Technology

[0004] Traffic information can typically be categorized as traffic flow information or accident information.

[0005] Traffic flow information can include current traffic conditions, such as vehicle speed on each road segment, vehicle density within the segment, and total travel time.

[0006] Drivers can view this traffic flow information through visual display features such as route colors displayed on navigation systems and mobile apps.

[0007] Accident information may include information about traffic accidents, broken-down vehicles, road construction, assemblies, gatherings, etc., and may also include emergency traffic information that affects traffic flow.

[0008] Drivers can view information about such accidents through CCTV footage provided by the traffic management system, posts on the traffic information center website, traffic information social media, and traffic radio.

[0009] Traffic broadcasts are typically used to receive real-time information about accidents and other incidents.

[0010] However, traffic radio provides information focused on major roads based on a specific day of the week and a specific time of day, such as highways on weekends, congested areas during weekday rush hours, and work areas during the day.

[0011] Existing methods for providing traffic information via navigation systems and mobile applications only display traffic flow information, making it difficult for drivers to determine what is causing traffic congestion in congested areas.

[0012] Furthermore, existing methods for providing traffic information primarily provide information about major roads, which prevents drivers from receiving accident information relevant to their routes or forces them to listen to information about road segments unrelated to their routes, resulting in fatigue or wasted time.

[0013] The subjects described in this Background section are intended to facilitate an understanding of the background of this disclosure and may therefore include subjects unknown to those skilled in the art. The statements in this section provide only background information in relation to this disclosure and may not constitute prior art. Summary of the Invention

[0014] This disclosure aims to provide devices and methods for providing real-time traffic information, collecting traffic information including the causes of traffic congestion in real time, and providing traffic information in real time.

[0015] This disclosure also aims to provide devices and methods for providing real-time traffic information, selecting traffic congestion information from traffic congestion information corresponding to the user's vehicle's travel route, and providing traffic information in real time.

[0016] This disclosure also aims to provide devices and methods for providing real-time traffic information and traffic congestion information by using traffic information retrieval to enhance generative large language models.

[0017] The purpose of this disclosure is not limited to the above-described purposes, and other purposes and advantages not mentioned in this disclosure should be understood from the following description and should become apparent from the implementation of this disclosure. It should also be understood that the purposes and advantages of this disclosure can be achieved by the means set forth in the claims and combinations thereof.

[0018] An apparatus for providing real-time traffic information according to one embodiment of the present disclosure includes: a traffic information collection unit configured to collect traffic information, including real-time driving speed and causes of traffic congestion, in real time as a user vehicle travels along a driving route. The apparatus further includes: a traffic congestion segment determination unit configured to determine a normal speed range based on the traffic information, and to determine segments where the real-time driving speed is below the lower limit of the normal speed range as traffic congestion segments. The apparatus further includes a traffic congestion segment information providing unit configured to provide traffic congestion segment information, including at least one of the following: the location of the traffic congestion segment, the cause of traffic congestion, and the expected delay time, based on the traffic information and the traffic congestion segments included in the user vehicle's driving route.

[0019] The traffic information collection unit can use image analysis models to analyze real-time traffic images of each road segment provided by the intelligent transportation system. Traffic information may include at least one of the following: real-time traffic accident information, assembly information, event information, traffic control information, and road construction information.

[0020] The traffic congestion segment identification unit can use a traffic pattern prediction model to determine the normal speed range according to a preset period. The traffic pattern prediction model uses real-time driving speed as training data.

[0021] A method for providing real-time traffic information according to another embodiment of this disclosure includes: collecting traffic information, including real-time driving speed and causes of traffic congestion, in real time as a user vehicle travels along a driving route. The method further includes calculating a normal speed range based on the traffic information. The method also includes identifying road segments where the real-time driving speed is below the lower limit of the normal speed range as traffic congested road segments. The method further includes providing traffic congested road segment information, based on the traffic information and the traffic congested road segments included in the user vehicle's driving route, including at least one of the following: the location of the traffic congested road segment, the cause of traffic congestion, and the expected delay time.

[0022] Real-time traffic information collection may include analyzing real-time traffic images of each road segment provided by the intelligent transportation system using image analysis models. Traffic information may include at least one of the following: real-time traffic accident information, assembly information, event information, traffic control information, and road construction information.

[0023] Identifying a road segment as a traffic congestion segment can also include: using a traffic pattern prediction model to calculate the normal speed range based on a preset period, with the traffic pattern prediction model using real-time driving speed as training data.

[0024] According to this disclosure, by collecting traffic information, including the causes of traffic congestion, in real time, users can be provided with information on congested road sections, including the causes of traffic congestion.

[0025] In addition, by selecting the traffic congestion information corresponding to the user's vehicle's route from the traffic congestion information, real-time traffic congestion information based on the selected route of the user's vehicle can be provided.

[0026] Furthermore, by leveraging retrieval-enhanced generative large language models, traffic congestion information can be provided in real time with reduced cost and time. Attached Figure Description

[0027] The foregoing and other aspects, features, and advantages, as well as the following detailed description of embodiments, should be better understood when read in conjunction with the accompanying drawings. However, this disclosure is not intended to be limited to the details shown in the drawings, and various modifications and structural changes may be made without departing from the spirit of this disclosure and within the scope and limits of the equivalents of the claims. In the various drawings, the same reference numerals and symbols indicate the same elements.

[0028] Figure 1 This is a block diagram illustrating an apparatus for providing real-time traffic information according to one embodiment of the present disclosure.

[0029] Figure 2 This is a flowchart illustrating a method for providing real-time traffic information according to one embodiment of the present disclosure.

[0030] Figure 3 This is a diagram illustrating a method for providing real-time traffic information according to one embodiment of the present disclosure. Detailed Implementation

[0031] The embodiments disclosed in this disclosure are described in more detail with reference to the accompanying drawings, and throughout the drawings, the same reference numerals are used to designate the same or similar components, and redundant descriptions are omitted. As used herein, for ease of explanation, the terms "module" and "unit" used to refer to components are used interchangeably, and therefore the terms themselves should not be considered to have different meanings or functions. Regarding the description of this disclosure, detailed descriptions of related known technologies may be omitted when they are determined to unnecessarily obscure the gist of this disclosure. Furthermore, it should be understood that the drawings are intended only to aid in understanding the embodiments disclosed in this disclosure and do not limit the technical principles and scope of this disclosure. Rather, it should be understood that the drawings include all modifications, equivalents, or substitutions described by the technical principles and falling within the technical scope of this disclosure.

[0032] Although the terms first, second, third, etc., may be used herein to describe various elements, components, regions, layers, and / or parts, these elements, components, regions, layers, and / or parts should not be limited by these terms. These terms are used only to distinguish one element from another.

[0033] When a component or layer is referred to as being "on," "attached to," "connected to," or "coupled to" another component or layer, it may be directly on, attached to, connected to, or coupled to the other component or layer, or there may be intermediate components or layers between the components or layers. Conversely, when a component is referred to as being "directly on," "directly attached to," "directly connected to," or "directly coupled to" another component or layer, there are no intermediate components or layers between the components or layers.

[0034] When the controllers, units, modules, components, devices, elements, etc. of this disclosure are described as having a purpose or performing an operation or function, the controllers, units, modules, components, devices, elements, etc., shall be regarded herein as being "configured" to satisfy that purpose or perform that operation or function. Each controller, unit, module, component, device, element, etc. may be embodied individually or included as part of a device together with a processor and memory (such as a non-transitory computer-readable medium).

[0035] In this disclosure, each of the phrases such as “A or B”, “at least one of A and B”, “at least one of A or B”, “A, B or C”, “at least one of A, B and C”, “at least one of A, B or C”, and “at least one of A, B or C or a combination thereof” may include any or all possible combinations of the items listed together in the corresponding phrase.

[0036] In the following text, see references Figures 1 to 3 The present disclosure provides a detailed description of the apparatus and method for providing real-time traffic information.

[0037] Figure 1 This is a block diagram illustrating an apparatus for providing real-time traffic information according to one embodiment of the present disclosure, and Figure 2 This is a flowchart illustrating a method for providing real-time traffic information according to one embodiment of the present disclosure.

[0038] refer to Figure 1 According to one embodiment of the present disclosure, a device 100 for providing real-time traffic information may include a traffic information collection unit 110, a traffic congestion segment determination unit 120, and a traffic congestion segment information providing unit 130.

[0039] Traffic information collection unit 110 collects real-time traffic information, including real-time speed and causes of traffic congestion, as the user's vehicle travels along the route (see [link]). Figure 2 (S210 in the middle).

[0040] For example, the traffic information collection unit 110 can collect information on real-time driving speed, real-time traffic accidents, gatherings, events, traffic control, road construction, etc., for each road segment provided by the intelligent transportation system.

[0041] For example, the traffic information collection unit 110 can collect traffic information through the Seoul Traffic Operations and Information Service, Road Plus, the City Traffic Information Center, the Traffic Broadcasting Network (TBN), and Twitter. Furthermore, the traffic information collection unit 110 can collect traffic accident information through CCTV, the National Traffic Information Center, local police stations, and local police stations. The traffic information collection unit 110 can collect building and road construction information through local government documents, the National Police Agency, and periodic survey results. It can also collect information on natural disasters (such as forest fires, landslides, floods, river overflows, rainfall, and heavy snow) through CCTV, national disaster text messages, the National Police Agency, and local government disaster text messages. The traffic information collection unit 110 can collect information on rallies through CCTV and the National Police Agency. Finally, the traffic information collection unit 110 can collect event information through the Korea Tourism Organization, portal news releases, local government announcements, and marathon online platforms.

[0042] The traffic information collection unit can use image analysis model 110a to analyze real-time traffic images of each road segment provided by the intelligent transportation system, and can collect the causes of traffic congestion in the relevant road segments in real time. Here, intelligent transportation system (ITS) refers to a system that provides traffic information and services by integrating electronic, control and communication technologies.

[0043] Traffic information collection unit 110 can collect large amounts of traffic information using public APIs provided in formats such as Extensible Markup Language (XML) and JavaScript Object Notation (JSON). Traffic information collection unit 110 can extract traffic information from, for example, text-converted traffic broadcasts, text messages (SMS) that include traffic information and social networking service (SNS) streams.

[0044] Furthermore, the traffic information collection unit 110 can use a text summarization model 110b to classify the collected traffic information according to road segments. The traffic information collection unit 110 can generate summarized traffic information, including the location of congested road segments, the cause of traffic congestion, and expected delay times. Here, the expected delay time can be calculated based on the normal speed range and the real-time driving speed, as will be described below. For example, the expected delay time can be calculated as the difference between the upper limit of the normal speed range and the real-time driving speed.

[0045] Finally, the traffic information collection unit 110 can store the summarized traffic information in the database 110c. At this time, the data stored in the database 110c can be the raw data before preprocessing.

[0046] For example, traffic information collection unit 110 can collect data via an API request module, such as an HTTP client (e.g., a request library in Python) or a WebSocket client (e.g., the socket.10-client library in Node.js), executed on a server equipped with a processor and storage. Traffic information collection unit 110 can perform API communication via a network interface (e.g., a standard Ethernet interface).

[0047] In another embodiment, the traffic information collection unit 110 can collect real-time driving speeds based on GPS, acceleration, and rotation information generated from each vehicle. Therefore, the traffic information collection unit 110 can publish the real-time driving speed v at point i based on time t. i (t). Based on this, the traffic congestion segment determination unit 120 can determine the traffic congestion segment based on the accumulated real-time driving speed v. i (t), using statistical models or artificial intelligence regression models, calculate normal speed data v according to time periods. avg_i (t). This can be calculated as an average or range. Finally, the traffic congestion segment determination unit can be based on v i (t) and v avg_i (t) determines the congestion data at a specific point x i (t). The traffic information collection unit can be implemented by subscribing to x from the unit determined by traffic congestion sections. i (t), and can be published in x i At time (t), the database is updated via the scheduler. When v i (t) compared to v avg_i (t) When the preset threshold is small, the traffic congestion segment determination unit can determine x i (t) is determined to be 1 (considered a congested situation compared to normal speed), and otherwise, the traffic congestion segment determination unit can determine x. i (t) is determined to be 0.

[0048] In another embodiment, when the text collected by the traffic information collection unit includes information about the expected end of lane control due to a traffic accident, or a future time point t that has not yet occurred. f When planning for traffic congestion (such as marathons and rallies), you can record the congestion data at the corresponding time points. i (t) f ).

[0049] Simultaneously, the traffic information collection unit 110 can be configured to determine the number of vehicles included in each link, and to determine a road congestion index proportional to the number of vehicles in each link, the frequency of adaptive cruise control operation, and the frequency of blind spot warning system operation. The traffic congestion segment determination unit 120 can be configured to determine congested road segments based on the congestion index.

[0050] Traffic information can be collected in the form of organized HTML tables or sentences, such as, "The [unavailable lane] of [road name] from [congestion start point] to [congestion end point] is under [control status] due to [congestion cause]. Please drive carefully." In this case, the congestion time t included in the traffic information can be explicitly published on the site or determined based on the time collected by the traffic information collection unit 110. Furthermore, traffic information can be HTML files downloaded periodically from a pre-set website. For example, the traffic information collection unit 110 can assign JSON identifier key-value pairs to the collected text in the order they are collected. When the traffic congestion cause for each point is published on the traffic information site as a table or structured sentence, and information that is structurally located in the same position can be extracted using the same selector, they can be structured into JSON format. The text separated by JSON key-value pairs can be used as metadata for vectors and as filtering conditions during vector searches. For example, these filtering conditions can be implemented to filter by criteria such as within a few minutes before or after a specific time, or within kilometers of the congestion location.

[0051] Furthermore, even when no text regarding congested road sections has been published, multiple images can be acquired by capturing real-time CCTV footage of the corresponding road section, based on the location of the congested road section determined by the traffic congestion determination unit disclosed herein. The traffic information collection unit can also be implemented to identify the captured multiple images using a known machine learning model describing road conditions and output text including the cause of the congestion.

[0052] The traffic congestion segment identification unit 120 calculates the normal speed range for each road segment based on traffic information accumulated in the database, and identifies road segments where the real-time driving speed is below the lower limit of the normal speed range as traffic congestion segments (see [link]). Figure 2 (S220 in the middle).

[0053] For example, the traffic congestion segment determination unit can use the traffic pattern prediction model 120a to output normal speed according to a preset cycle. The traffic pattern prediction model 120a uses the accumulated driving speed data (e.g., driving speed on a certain day of the week and at a certain time of day) stored in the database 110c as training data to generate normal speed patterns for each segment.

[0054] For example, a traffic pattern prediction model can use driving speed data categorized according to a preset period (such as by road, a day of the week, and a time of day) as training data, generate prediction data for each preset period, which has a similar pattern to the training data of driving speed by road, a day of the week, and a time of day, and calculate the normal speed range corresponding to the road, the day, and the time of day.

[0055] At the same time, when there is traffic congestion (e.g., a traffic accident), the real-time driving speed may be lower than the lower limit of the normal speed range.

[0056] Therefore, the traffic congestion segment identification unit will compare the normal speed with the real-time driving speed for each segment, and identify the segments where the real-time driving speed is lower than the lower limit of the normal speed range as traffic congestion segments.

[0057] For example, a traffic congestion segment determination unit can use a speed pattern learning model and a congestion segment determination model executed on a server equipped with a processor and storage device to output congested segments.

[0058] The traffic congestion information providing unit 130 provides traffic congestion information based on traffic information and traffic congestion sections included in the user's vehicle's travel route, including at least one of the following: the location of the traffic congestion section, the cause of the traffic congestion, and the expected delay time (see [link]). Figure 2 (S230 in the middle).

[0059] For example, when a specific road segment is identified as a traffic congestion segment and the causes of congestion in the relevant road segment are collected, the traffic congestion segment information providing unit 130 can provide this information to the user's vehicle until the traffic congestion in the relevant road segment is cleared or the cause is updated.

[0060] Here, the causes of traffic congestion can include, for example, accidents, gatherings, events, traffic control and road construction, and the expected delay time can be the time that is delayed compared to normal conditions due to traffic congestion.

[0061] The traffic congestion information provision unit includes a vector generation unit 130a, a vector search unit 130b, and a text generation unit 130c.

[0062] The vector generation unit 130a can segment and expand traffic information based on at least one of the preset word counting units, sentence boundaries, and paragraph boundaries, and perform vectorization on the segmented and expanded traffic information to generate traffic information vectors.

[0063] In addition, the vector generation unit 130a can segment and expand the traffic congestion segments included in the user vehicle's driving route based on at least one of the preset word counting units, sentence boundaries, and paragraph boundaries, and perform vectorization on the segmented and expanded traffic congestion segments to generate route vectors.

[0064] The generated traffic information vectors and route vectors can be stored in a vector database.

[0065] The vector search unit 130b retrieves traffic information vectors that are close to the route vector from the vector database.

[0066] The vector database is updated using real-time traffic information for each road segment and is configured to allow vector search units to refer to the vector database.

[0067] Existing large language model techniques are limited in providing real-time traffic information because they generate responses from the data they have learned.

[0068] In contrast, this disclosure uses a retrieval-enhanced generative (RAG) large language model (LLM) to generate sentences that provide traffic information, thereby enabling real-time explanation of the causes of traffic congestion.

[0069] Here, because it performs the search based on real-time external data, retrieval-enhanced generative large language models can alleviate the illusion of large language models and provide more accurate real-time information.

[0070] For example, vector generation unit 130a and vector search unit 130b can use a vector search module (such as Facebook FAISS, Milvus, or Annoy) that executes on a server equipped with a processor and storage device to perform similarity searches. For example, vector generation unit 130a and vector search unit 130b can use the vector search module to convert input queries (e.g., routes including congested road sections) into vectors and perform nearest neighbor searches in a vector database.

[0071] The text generation unit 130c generates a sentence based on the retrieved traffic information vector and a preset sentence template, which includes at least one of the following: the location of the traffic congestion section, the cause of the traffic congestion, and the expected delay time.

[0072] Here, the text generation unit can use a retrieval-enhanced generative large language model to generate sentences.

[0073] Therefore, a device for providing real-time traffic information can select congested road segments from the road segments included in the user's vehicle's driving route, perform a similarity search in a vector database to select traffic information related to route vectors containing the selected congested road segments, and generate a sentence by combining the searched keywords. The device can then provide this sentence to the user's vehicle.

[0074] Figure 3 This is a diagram illustrating an example of a method for providing real-time traffic information according to one embodiment of the present disclosure.

[0075] For example, refer to Figure 3 The selected congested road segment is Hakik Bridge, the cause of congestion is a traffic accident, the expected delay time is 10 minutes, and the preset sentence template is: "(selected congested road segment) traffic congestion due to (traffic congestion cause), with an estimated delay of (expected delay time) compared to normal conditions."

[0076] As described above, when the traffic congestion information providing unit 130 is designed to retrieve only traffic congestion sections along the driving route, it can provide information about sections classified as real-time traffic congestion sections.

[0077] Conversely, when the traffic congestion information providing unit 130 is designed to retrieve information on all road segments included in the travel route, if traffic congestion causes (e.g., planned assemblies, events, traffic control, or road construction) are retrieved for the user's estimated arrival time for road segments not currently classified as traffic congestion segments, traffic congestion information for those segments can be provided. The user's estimated arrival time can be calculated based on the required travel time for each road segment, and the required travel time for each road segment can be calculated based on the normal speed range of each road segment and the expected delay time of previous stops, etc.

[0078] Furthermore, the traffic congestion information providing unit 130 can be designed to provide traffic congestion information limited to the user's frequently used routes, thereby restricting the provision of unnecessary information. For example, when a user's vehicle deviates from its driving pattern (e.g., route, time, day of the week), the output of traffic congestion information providing unit can be restricted.

[0079] The device for providing real-time traffic information according to this disclosure can provide users with detour routes that allow their vehicles to avoid traffic congestion by giving them advance notice of congested road sections. Furthermore, it can provide users with expected delay times compared to normal conditions, thereby providing users with information related to their departure time.

[0080] The device for providing real-time traffic information according to this disclosure vectorizes a user's driving route and real-time traffic information, performs a similarity search in a vector database, and then selects real-time traffic information related to the user's driving route. The selected real-time traffic information is appropriately arranged and processed in a preset template, input into a large language model prompt, and outputs a sentence based on the input traffic information.

[0081] Therefore, it not only provides users with the location of traffic congestion sections, but also provides explanations of the reasons for the congestion, thereby enhancing users' understanding of detour guidance based on congested sections, thus improving driving satisfaction and preventing them from entering congested areas.

[0082] As used in this disclosure (especially in the appended claims), the terms “an” and “the” include both singular and plural meanings, unless the context clearly indicates otherwise. Furthermore, it should be understood that any numerical range listed in this disclosure is intended to include all subranges contained therein (unless otherwise explicitly indicated), and therefore, the disclosed numerical range includes each individual value between the minimum and maximum values ​​of the numerical range.

[0083] The steps of the method according to this disclosure can be performed in a suitable order unless a particular order is described or otherwise specified. In other words, this disclosure is not necessarily limited to the enumerated order of steps. All examples or indicative terms (“e.g.,” “such as”) described in this disclosure are used only to describe this disclosure in more detail. Therefore, it should be understood that the scope of this disclosure is not limited to the exemplary embodiments described above or limited by the use of such terms, unless defined by the appended claims. Furthermore, those skilled in the art will understand that various modifications, combinations, and substitutions can be made within the scope of the appended claims or their equivalents, depending on design conditions and factors.

[0084] Therefore, this disclosure is not limited to the exemplary embodiments described above, but is intended to include all modifications, equivalents, and substitutions that fall within the spirit and scope of the appended claims.

Claims

1. A device for providing real-time traffic information, the device comprising: The traffic information collection unit is configured to collect traffic information, including real-time driving speed and causes of traffic congestion, in real time as the user's vehicle travels along the driving route. The traffic congestion section determination unit is configured to determine the normal speed range based on the traffic information, and to determine the road sections where the real-time driving speed is lower than the lower limit of the normal speed range as traffic congestion sections. as well as A traffic congestion information providing unit is configured to provide traffic congestion information based on the traffic information and the traffic congestion sections included in the user vehicle's driving route, including at least one of the following: the location of the traffic congestion section, the cause of the traffic congestion, and the expected delay time.

2. The device according to claim 1, in, The traffic information collection unit is also configured to: collect the causes of traffic congestion based on image analysis models and real-time traffic images provided by the intelligent transportation system, and The traffic information includes at least one of the following: real-time traffic accident information, assembly information, event information, traffic control information, and road construction information.

3. The device according to claim 2, wherein, The traffic congestion segment determination unit is further configured to: use a traffic pattern prediction model to determine the normal speed range according to a preset period, wherein the traffic pattern prediction model uses the real-time driving speed as training data.

4. The device according to claim 3, wherein, The traffic congestion section information providing unit includes a vector generation unit, which is configured to: The traffic information is segmented and expanded based on at least one of the preset word counting units, sentence boundaries, and paragraph boundaries. The segmented and expanded traffic information is then vectorized to generate a traffic information vector. Based on at least one of the preset word counting units, sentence boundaries, and paragraph boundaries, the traffic congestion segments included in the user vehicle's driving route are segmented and expanded, and the segmented and expanded traffic congestion segments are vectorized to generate route vectors.

5. The device according to claim 4, wherein, The traffic congestion information providing unit further includes a vector search unit, which is configured to retrieve traffic information vectors that are close to the route vector.

6. The device according to claim 5, in, The traffic congestion information providing unit further includes a text generation unit, which is configured to: generate a sentence based on the retrieved traffic information vector and a preset sentence template, including at least one of the following: the location of the traffic congestion segment, the cause of the traffic congestion, and the expected delay time. The traffic congestion information providing unit is further configured to provide traffic congestion information based on the generated sentence.

7. The device according to claim 6, wherein, The expected delay time is determined based on the normal speed range and the real-time driving speed.

8. The device according to claim 7, wherein, The text generation unit is also configured to generate the sentence using a retrieval-enhanced generative large language model.

9. A method for providing real-time traffic information, the method comprising: As the user's vehicle travels along the route, real-time traffic information, including real-time driving speed and the causes of traffic congestion, is collected. Based on the traffic information, calculate the normal speed range; Road sections where the real-time driving speed is lower than the lower limit of the normal speed range are identified as traffic congestion sections. as well as Based on the traffic information and the traffic congestion sections included in the user vehicle's driving route, traffic congestion section information is provided, including at least one of the location of the traffic congestion section, the cause of the traffic congestion, and the expected delay time.

10. The method according to claim 9, in, Real-time collection of the traffic information includes: collecting the causes of traffic congestion based on image analysis models and real-time traffic images provided by intelligent transportation systems; and The traffic information includes at least one of the following: real-time traffic accident information, assembly information, event information, traffic control information, and road construction information.

11. The method according to claim 10, wherein, Identifying a road segment as a traffic congestion segment includes: using a traffic pattern prediction model to calculate the normal speed range according to a preset period, wherein the traffic pattern prediction model uses the real-time driving speed as training data.

12. The method according to claim 11, wherein, The information provided regarding the traffic congestion sections includes: The traffic information is segmented and expanded based on at least one of preset word counting units, sentence boundaries, and paragraph boundaries, and the segmented and expanded traffic information is vectorized to generate a traffic information vector; and Based on at least one of the preset word counting units, sentence boundaries, and paragraph boundaries, the traffic congestion segments included in the user vehicle's driving route are segmented and expanded, and the segmented and expanded traffic congestion segments are vectorized to generate route vectors.

13. The method according to claim 12, wherein, Providing the traffic congestion information also includes: retrieving traffic information vectors that are close to the route vector.

14. The method according to claim 13, in, Providing the traffic congestion information further includes: generating a sentence based on the retrieved traffic information vector and a preset sentence template, including at least one of the following: the location of the traffic congestion segment, the cause of the traffic congestion, and the expected delay time. The provision of the traffic congestion information also includes: providing the traffic congestion information based on the generated sentence.

15. The method of claim 14, further comprising: The expected delay time is determined based on the normal speed range and the real-time driving speed.

16. The method according to claim 15, wherein, Generating the sentence includes: using a retrieval-enhanced generative large language model to generate the sentence.

Images