Real-time traffic flow prediction method and system based on deep learning

Abstract

The present application is applicable to the field of traffic prediction technology, and in particular to a real-time traffic flow prediction method and system based on deep learning, the method comprising: obtaining traffic data in real time; inputting the traffic data of the current time period into the prediction model, and obtaining the road congestion value of each time period within the preset time period after the current time period output by the prediction model; obtaining the location data of the second electronic device; when the geographical location reflected by the location data is within the preset geographical column, and the road congestion value of the current time period is higher than the first preset value, generating first road condition information, and sending the road condition information to the second electronic device. The real-time traffic flow prediction method based on deep learning provided by the present application can solve the problem that it is impossible to provide users with comprehensive traffic congestion information during the traffic flow prediction process, resulting in inaccurate predictions and inability to automatically help users deal with traffic congestion, thereby affecting the user's travel experience.

Description

Technical Field

[0001] The present application belongs to the field of traffic prediction technology, and in particular, relates to a real-time traffic flow prediction method and system based on deep learning. Background Art

[0002] Traffic flow forecasting refers to the use of traffic monitoring equipment, cameras, sensors and other equipment to collect vehicle operation data on the road, and analyze and predict vehicle flow on roads, intersections or sections.

[0003] Traditional traffic flow forecasts often only consider the traffic conditions of the road itself and are unable to provide users with comprehensive traffic congestion information, resulting in inaccurate forecasts and thus being unable to automatically help users cope with traffic congestion, which in turn affects users' travel experience. Summary of the invention

[0004] The embodiments of the present application provide a real-time traffic flow prediction method and system based on deep learning, which can solve the problem that only the traffic conditions of the road itself are considered during the traffic flow prediction process, and comprehensive traffic congestion information cannot be provided to users, resulting in inaccurate predictions, and thus unable to automatically help users deal with traffic congestion, thereby affecting the user's travel experience.

[0005] In a first aspect, an embodiment of the present application provides a real-time traffic flow prediction method based on deep learning, comprising:

[0006] Acquire traffic data in real time; wherein the traffic data is used to reflect vehicle flow and road conditions;

[0007] Input the traffic data of the current time period into a prediction model to obtain the road congestion value of each time period within a preset time period after the current time period output by the prediction model; wherein the prediction model is obtained by deep learning;

[0008] Acquiring location data of the second electronic device;

[0009] When the geographical location reflected by the location data is within a preset geographical column and the road congestion value in the current time period is higher than a first preset value, first road condition information is generated and sent to the second electronic device; wherein the road condition information is used to indicate the congestion level of the road within the preset geographical column, so that the user of the second electronic device can select a travel road;

[0010] When no road adjustment response is received from the second electronic device within the first time period, when it is detected that the road congestion value in the next time period is higher than the first preset value and the geographical location reflected by the location data is within the preset geographical column, second road condition information is generated and sent to the second electronic device.

[0011] The above technical solutions in the embodiments of the present application have at least the following technical effects:

[0012] The real-time traffic flow prediction method of deep learning provided by the present application, first, by acquiring traffic data including vehicle flow and road conditions in real time, more diverse and multi-dimensional information can be provided, which is helpful for the data used in the subsequent model training and prediction process to be the latest, and can timely reflect the actual situation of traffic conditions, and improve the prediction accuracy. Secondly, the traffic data of the current time period is input into the prediction model, and the road congestion value of each time period within the preset time period after the current time period output by the prediction model is obtained, which can comprehensively consider the impact of various factors on traffic flow and road congestion, and provide more reference data for real-time traffic flow prediction, which is helpful for the subsequent deep learning model to more comprehensively learn and understand the correlation and regularity between traffic data, so as to improve the prediction accuracy. Then, the location data of the second electronic device is obtained, which can provide additional feature information, such as the driving direction, speed, and dwell time of the vehicle connected to the second electronic device, which is helpful for a comprehensive understanding of the vehicle operation status and road conditions, and improve the prediction accuracy and reliability. Then, when the geographical location reflected by the location data is within the preset geographical column, and the road congestion value of the current time period is higher than the first preset value, the first road condition information is generated and sent to the second electronic device, which can provide the user with the current road congestion status in a timely manner, helping the user to more accurately select a travel route, thereby reducing travel time and avoiding traffic congestion (for example, user usage feedback and behavior data can be collected and used to improve the prediction performance of the deep learning model, thereby continuously optimizing the traffic flow prediction system), thereby improving the prediction accuracy. Finally, when the road adjustment response of the second electronic device is not received in the first time period, when it is detected that the road congestion value of the next time period is higher than the first preset value, and the geographical location reflected by the location data is within the preset geographical column, the second road condition information is generated and sent to the second electronic device, which can help the user make better travel decisions in the next time period and avoid traffic congestion, which helps to improve the user's travel efficiency and comfort and enhance the user experience.

[0013] In a possible implementation manner of the first aspect, before inputting the traffic data of the current time period into a prediction model to obtain a road congestion value for each time period within a preset time period after the current time period output by the prediction model, the method further includes:

[0014] Generate a first sample based on empirical data; wherein the empirical data includes the traffic volume and road conditions of the road acquired in the past time period, the road conditions include weather information and construction road condition information, the weather information includes information on whether it is the rainy season, and the construction road condition information includes information on whether there is a road construction plan and information on the schedule of the construction period; each of the first samples includes a road congestion value corresponding to each time period of the day, and the road congestion value is determined according to the congestion time acquired every day;

[0015] Adding identification information to the traffic data; wherein the identification information is used to indicate a current road congestion value corresponding to the traffic data in a current time period;

[0016] Sending the traffic data and the first sample after adding the identification information to a server;

[0017] Receiving an updated prediction model sent by the server; wherein the updated prediction model is obtained by the server through training based on the received traffic data after adding the identification information and the first sample;

[0018] The current prediction model is updated to the updated prediction model.

[0019] In a possible implementation manner of the first aspect, when the geographical location reflected by the location data is within a preset geographical column and the road congestion value in the current time period is higher than a first preset value, after generating first road condition information and sending the first road condition information to the second electronic device, the method further includes:

[0020] determining the usage state of the second electronic device when the first electronic device does not receive a road adjustment response from the second electronic device; wherein the road adjustment response carries road selection information corresponding to the road condition information;

[0021] When the usage state is a first motion state, the first electronic device sends the first road condition information to the second electronic device within a first preset time period; wherein the user is using a navigation application of the second electronic device when the second electronic device is in the first motion state;

[0022] When the usage state is the second motion state, the first electronic device does not send the first road condition information to the second electronic device within a second preset time period; wherein, when the second electronic device is in the second motion state, the user does not use the navigation application of the second electronic device.

[0023] In a possible implementation manner of the first aspect, when the first electronic device does not receive a road adjustment response from the second electronic device, determining a usage status of the second electronic device includes:

[0024] If the first electronic device does not receive a road adjustment response corresponding to the road condition information from the second electronic device within a second preset time period, the first electronic device obtains selection data and / or motion data of the second electronic device; wherein the selection data is used to indicate whether the second electronic device uses the navigation application of the second electronic device to select a road, and the motion data is used to indicate whether the driving path of the vehicle connected to the second electronic device deviates from the preset driving path;

[0025] When the selection data indicates that the second electronic device is using a navigation application of the second electronic device to select a road and / or the motion data indicates that a driving path of a vehicle connected to the second electronic device does not deviate from a preset driving path, determining that the use state of the second electronic device is the first motion state;

[0026] When the selection data indicates that the second electronic device does not use the navigation application of the second electronic device to select a road and / or the motion data indicates that the driving path of the vehicle connected to the second electronic device is zero, the usage state of the second electronic device is determined to be the second motion state.

[0027] In a possible implementation manner of the first aspect, after inputting the traffic data of the current time period into a prediction model to obtain a road congestion value for each time period within a preset time period after the current time period output by the prediction model, the method further includes:

[0028] Based on the traffic data of the current time period, determining whether the road condition of the current time period is a first road condition; wherein the first road condition is a road condition including at least two adjacent intersections within a preset distance;

[0029] If the road condition is the first road condition, obtaining a first signal device list; wherein the first signal device list includes at least one signal device;

[0030] Acquire signal device cycle information of each of the signal devices in the first signal device list;

[0031] Based on the signal device cycle information and the first road condition, synchronization is performed to obtain green wave band control information; wherein the green wave band control information is used to control the signal devices between adjacent intersections so that vehicles can pass through the adjacent intersections continuously and smoothly.

[0032] In a possible implementation manner of the first aspect, the synchronous setting based on the signal device cycle information and the first road condition to obtain green band control information includes:

[0033] When the first road condition indicates that the vehicle flow is greater than the first threshold value and the duration value in the signal device cycle information is the first duration value, obtaining the first control information in the green wave band control information; wherein the first control information is used to control the signal device to extend the green light duration;

[0034] When the first road condition indicates that the traffic volume is less than the first threshold value and the duration value in the signal device cycle information is a second duration value, the second control information in the green wave band control information is obtained; wherein the second control information is used to control the signal device to shorten the cycle duration of the signal device.

[0035] In a possible implementation manner of the first aspect, after performing synchronization setting based on the signal device cycle information and the first road condition to obtain green band control information, the method further includes:

[0036] Acquire bus travel data in the first road condition; wherein the bus travel data is used to indicate the daily departure schedule and average speed data of the bus;

[0037] According to the departure schedule data and the average vehicle speed data, third control information in the green band control information is obtained; wherein the third control information is used to adjust the phase difference of the signal device;

[0038] Traffic management decision information is obtained according to the first control information, the second control information and the third control information.

[0039] In a possible implementation manner of the first aspect, at least two signal devices are included on the road, and the first electronic device is communicatively connected to the signal device. After performing synchronization setting based on the signal device cycle information and the first road condition to obtain green wave band control information, the method further includes:

[0040] Based on the green band control information, controlling the first signal device to enter a signal adjustment state; wherein the signal adjustment state is used to remind the user of the time or priority of the signal device;

[0041] If the first electronic device cannot receive a control response fed back by any one of the two signal devices, when the first signal device enters a signal adjustment state, the first signal device receives a first adjustment signal sent by the second signal device; wherein the first adjustment signal is used to determine a signal duration adjustment condition of the second signal device;

[0042] The first signal device sends a first adjustment response to the second signal device; wherein the first adjustment response is used to instruct the second signal device to maintain the signal adjustment state after receiving the first adjustment response, and the second signal device is in the signal adjustment state when sending the first adjustment signal.

[0043] In a possible implementation manner of the first aspect, after performing synchronization setting based on the signal device cycle information and the first road condition to obtain green band control information, the method further includes:

[0044] If the first electronic device cannot receive a control response fed back by any one of the two signal devices, when the first signal device enters a non-signal adjustment state, the first signal device receives a second adjustment signal sent by the second signal device; wherein the second signal device is in the signal adjustment state when sending the second adjustment signal;

[0045] In response to receiving the second adjustment signal, the first signal device does not send a second adjustment response to the second signal device, so that the second signal device exits the signal adjustment state.

[0046] In a possible implementation manner of the first aspect, when the geographical location reflected by the position data is within a preset geographical column and the road congestion value in the current time period is higher than a first preset value, generating the first road condition information includes:

[0047] Dividing the road sections within the preset geographic column to obtain traffic flow data within each area;

[0048] Acquiring vehicle speed data from the traffic data;

[0049] The first road condition information in each area is generated based on the vehicle speed data and the vehicle flow data.

[0050] In a second aspect, an embodiment of the present application provides a real-time traffic flow prediction system based on deep learning, comprising:

[0051] A first acquisition unit is used to acquire traffic data in real time; wherein the traffic data is used to reflect the traffic volume and road conditions;

[0052] A determination unit, configured to input the traffic data of the current time period into a prediction model, and obtain a road congestion value of each time period within a preset time period after the current time period output by the prediction model; wherein the prediction model is obtained by deep learning;

[0053] A second acquiring unit, configured to acquire location data of the second electronic device;

[0054] a first generating unit, configured to generate first road condition information when the geographical location reflected by the location data is within a preset geographical column and the road congestion value in the current time period is higher than a first preset value, and send the road condition information to the second electronic device; wherein the first road condition information is used to indicate the congestion level of the road within the preset geographical column, so that a user using the second electronic device can select a travel road;

[0055] The second generating unit is used to generate second road condition information when no road adjustment response is received from the second electronic device within the first time period, and when it is detected that the road congestion value in the next time period is higher than the first preset value and the geographical location reflected by the location data is within the preset geographical column, and send the second road condition information to the second electronic device.

[0056] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements a method as described in any one of the first aspects above.

[0057] It can be understood that the beneficial effects of the second to third aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0059] Figure 1 It is a flowchart of a real-time traffic flow prediction method based on deep learning provided in an embodiment of the present application;

[0060] Figure 2 It is a schematic diagram of the implementation process before step S200 in the real-time traffic flow prediction method based on deep learning provided in an embodiment of the present application;

[0061] Figure 3 It is a schematic diagram of the implementation process after step S400 in the real-time traffic flow prediction method based on deep learning provided in an embodiment of the present application;

[0062] Figure 4 It is a flowchart of step S410 in the real-time traffic flow prediction method based on deep learning provided in one embodiment of the present application;

[0063] Figure 5 It is a schematic diagram of the implementation process after step S200 in the real-time traffic flow prediction method based on deep learning provided in an embodiment of the present application;

[0064] Figure 6 It is a schematic diagram of the implementation process of step S290 in the real-time traffic flow prediction method based on deep learning provided in an embodiment of the present application;

[0065] Figure 7 It is a schematic diagram of the implementation process after step S290 in the real-time traffic flow prediction method based on deep learning provided in one embodiment of the present application;

[0066] Figure 8 It is a schematic diagram of the implementation process after step S290 in the real-time traffic flow prediction method based on deep learning provided in another embodiment of the present application;

[0067] Fig. 9 This is a schematic diagram of the implementation process after step S290 in the real-time traffic flow prediction method based on deep learning provided in another embodiment of the present application;

[0068] Fig.10 It is a schematic diagram of the implementation process of step S400 in the real-time traffic flow prediction method based on deep learning provided by the present application;

[0069] Fig.11 It is a structural diagram of a real-time traffic flow prediction system based on deep learning provided in an embodiment of the present application;

[0070] Fig.12 It is a schematic diagram of the structure of a first electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0071] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0072] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0073] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0074] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0075] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0076] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0077] Traffic flow forecasting refers to the use of traffic monitoring equipment, cameras, sensors and other equipment to collect vehicle operation data on the road, and analyze and predict vehicle flow on roads, intersections or sections.

[0078] Traditional traffic flow forecasts often only consider the traffic conditions of the road itself, and it is difficult to comprehensively consider the impact of external factors (such as weather, road construction activities, etc.) on traffic flow, resulting in inaccurate forecasts, which in turn affects users' travel experience.

[0079] To solve the above problems, an embodiment of the present application provides a real-time traffic flow prediction method and system based on deep learning.

[0080] In this method, firstly, by acquiring traffic data including vehicle flow and road conditions in real time, more diversified and multi-dimensional information can be provided, which is helpful for the data used in the subsequent model training and prediction process to be the latest, and to be able to timely reflect the actual situation of traffic conditions, thereby improving the prediction accuracy. Secondly, the traffic data of the current time period is input into the prediction model, and the road congestion value of each time period within the preset time period after the current time period output by the prediction model is obtained, which can comprehensively consider the impact of various factors on traffic flow and road congestion, provide more reference data for real-time traffic flow prediction, and help the subsequent deep learning model to more comprehensively learn and understand the correlation and regularity between traffic data, so as to improve the prediction accuracy. Then, obtaining the location data of the second electronic device can provide additional feature information, such as the driving direction, speed, and dwell time of the vehicle connected to the second electronic device, which helps to fully understand the vehicle operation status and road conditions, and improve the prediction accuracy and reliability. Then, when the geographical location reflected by the location data is within the preset geographical column, and the road congestion value of the current time period is higher than the first preset value, the first road condition information is generated and sent to the second electronic device, which can provide the user with the current road congestion status in a timely manner, helping the user to more accurately select a travel route, thereby reducing travel time and avoiding traffic congestion (for example, user usage feedback and behavior data can be collected and used to improve the prediction performance of the deep learning model, thereby continuously optimizing the traffic flow prediction system), thereby improving the prediction accuracy. Finally, when the road adjustment response of the second electronic device is not received in the first time period, when it is detected that the road congestion value of the next time period is higher than the first preset value, and the geographical location reflected by the location data is within the preset geographical column, the second road condition information is generated and sent to the second electronic device, which can help the user make better travel decisions in the next time period and avoid traffic congestion, which helps to improve the user's travel efficiency and comfort and enhance the user experience.

[0081] The real-time traffic flow prediction method based on deep learning provided in the embodiment of the present application can be applied to a first electronic device. In this case, the first electronic device is the executor of the real-time traffic flow prediction method based on deep learning provided in the embodiment of the present application. The embodiment of the present application does not impose any restrictions on the specific type of the first electronic device.

[0082] For example, the first electronic device can be a terminal device such as a tablet computer, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a desktop computer, a smart large screen, a computing device with wireless communication function or other processing devices connected to a wireless modem, a vehicle networking terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, and other mobile terminals.

[0083] The second electronic device may be a tablet computer, a laptop computer, a netbook, a handheld communication device, a mobile terminal such as a vehicle-mounted device, etc. The first electronic device is in communication connection with the second electronic device.

[0084] In order to better understand the real-time traffic flow prediction method based on deep learning provided in the embodiment of the present application, the specific implementation process of the real-time traffic flow prediction method based on deep learning provided in the embodiment of the present application is exemplarily introduced below.

[0085] Figure 1 A schematic flow chart of a real-time traffic flow prediction method based on deep learning provided in an embodiment of the present application is shown. The real-time traffic flow prediction method based on deep learning includes:

[0086] S100, real-time acquisition of traffic data, wherein the traffic data is used to reflect vehicle flow and road conditions.

[0087] For example, traffic monitoring cameras can be set up at traffic intersections and roads in the city to obtain information such as traffic volume, vehicle speed, traffic density, etc. through real-time monitoring of image and video data, and the obtained information can be sent to the first electronic device via wired or wireless means. Alternatively, traffic sensor equipment such as ground sensing coils, microwave radars, infrared sensors, etc. can also be installed on the road. These sensors can detect the traffic volume and speed of vehicles passing by in real time, help monitor the traffic volume and road conditions of the road, and send the obtained information to the first electronic device via wired or wireless means.

[0088] Such a setting can provide more diverse and multi-dimensional information by acquiring traffic data including vehicle flow and road conditions in real time, which will help ensure that the data used in subsequent model training and prediction processes are up to date and can reflect the actual traffic conditions in a timely manner, thereby improving prediction accuracy.

[0089] S200, inputting the traffic data of the current time period into the prediction model, and obtaining the road congestion value of each time period within a preset time period after the current time period output by the prediction model. The prediction model is obtained by deep learning.

[0090] For example, the traffic data in the current time period is used as a training set, and features are extracted from the traffic data, such as time, location, traffic flow, vehicle speed and other important features. These features will be used as inputs to the deep learning model, and these features are input into the deep learning model for training. The model weights are updated through the back propagation algorithm so that it learns the patterns and rules in the existing data to obtain a prediction model. The traffic data in the current time period is input into the prediction model for prediction. The prediction model will predict the road congestion value for each time period within the preset time period after the current time period based on the patterns and rules learned from the traffic data. For example, the current time is 8 o'clock in the morning, and the road congestion value for each 30-minute section in the next 2 hours is predicted. The road congestion value between 8:30-9:00 is 75, indicating that the urban highway section is lightly congested, and the road congestion value between 9:30-10:00 is 85, indicating that the city center road is moderately congested. For another example, the current time is 5 p.m. on Friday, and the road congestion is predicted for each 15-minute period in the next hour. The road congestion value between 5:45 and 6:00 is 95, indicating that the main roads leading into the city are severely congested.

[0091] With such a setting, the traffic data of the current time period is input into the prediction model, and the road congestion value of each time period within the preset time period after the current time period is obtained by the output of the prediction model. This can comprehensively consider the impact of various factors on traffic flow and road congestion, provide more reference data for real-time traffic flow prediction, and help subsequent deep learning models to more comprehensively learn and understand the correlation and regularity between traffic data, so as to improve prediction accuracy.

[0092] In one possible implementation, see Figure 2 In step S200, the traffic data of the current time period is input into the prediction model, and before the road congestion value of each time period within a preset time period after the current time period output by the prediction model is obtained, the real-time traffic flow prediction method based on deep learning also includes:

[0093] S210, generating a first sample based on the empirical data. The empirical data includes the traffic volume and road conditions of the road obtained in the past period, the road conditions include weather information and construction road condition information, the weather information includes information on whether it is the rainy season, and the construction road condition information includes information on whether there is a road construction plan and the arrangement information of the construction period in the schedule. Each first sample includes a road congestion value corresponding to each time period of the day, and the road congestion value is determined according to the congestion time obtained every day.

[0094] It is understandable that empirical data including road traffic flow data, weather information and construction road condition information over a period of time in the past can be collected. These data can be obtained from relevant institutions such as traffic management departments, meteorological bureaus, construction companies, or collected through cameras. Features are extracted from the collected data to generate the first sample. These features may include: Time period: various time periods of the day, such as morning rush hour, evening rush hour, etc. Road information: road name or ID. Traffic flow: the number of vehicles on the road in each time period. Weather information: such as whether it is raining, whether there is haze, etc. Construction road condition information: such as whether there is a road construction plan, construction cycle schedule, etc.

[0095] With this setting, generating the first sample based on empirical data can provide data support for the training of subsequent models and help the model predict traffic conditions more accurately. This feature engineering that comprehensively considers multiple features helps improve the prediction effect.

[0096] S220, adding identification information to the traffic data, wherein the identification information is used to indicate the current road congestion value corresponding to the traffic data in the current time period.

[0097] For example, the road congestion value can be divided into several levels, such as smooth, slow, and congested levels, which can be represented by A, B, and C, and these identification information can be added to the traffic data. For example, if the traffic volume of a road during the morning rush hour exceeds twice the average value, it is marked as congested (A), and if the traffic volume during the evening period is below the average value, it is marked as smooth (C).

[0098] Such a setting and adding identification information to traffic data can enable the model to more accurately understand the road congestion situation in the current time period during the subsequent model training process, which helps the model to better capture the traffic characteristics and patterns in a specific period of time, thereby improving the accuracy of the prediction.

[0099] S230, sending the traffic data with the identification information added and the first sample to the server.

[0100] It is understood that a connection between the first electronic device and the server is established using an appropriate communication protocol (such as HTTP, WebSocket, etc.). The traffic data with added identification information is sent to the server. Once the server receives the data, it returns a response to confirm receipt or provide further processing results to ensure that the first electronic device can process the response returned by the server to determine whether the server has received it successfully.

[0101] With this configuration, the traffic data with added identification information and the first sample are sent to the server, which can realize real-time traffic data update and prediction, thus ensuring that the deep learning model can obtain the latest traffic information in a timely manner, thereby realizing more accurate real-time traffic prediction.

[0102] S240, receiving an updated prediction model sent by the server, wherein the updated prediction model is obtained by the server through training based on the received traffic data with added identification information and the first sample.

[0103] It can be understood that the first electronic device sends a request to the server, requesting an updated prediction model. After receiving the request, the server will train according to the received traffic data with added identification information and the first sample, and generate a new prediction model. The server packages the new model into response data and sends it back to the first electronic device. After the first electronic device receives the response from the server, it needs to parse the response data and extract the updated prediction model. Once the first electronic device successfully parses the updated prediction model, it can apply it to the actual traffic flow prediction task and use it to perform real-time traffic flow prediction.

[0104] With this setting, real-time traffic data updates and predictions can be achieved by receiving the updated prediction model sent by the server, which ensures that the deep learning model can obtain the latest traffic information in a timely manner, thereby achieving more accurate real-time traffic flow predictions.

[0105] S250: Update the current prediction model to an updated prediction model.

[0106] It can be understood that the first electronic device needs to download the updated prediction model file sent by the server through a network request. The server usually sends the model file to the first electronic device in a specific format (such as .h5, .pt, etc.). The first electronic device needs to use appropriate tools or libraries to load the downloaded updated model file. For example, in the field of deep learning, libraries such as TensorFlow and PyTorch can be used to load neural network models. According to the format and type of the model file, the model is loaded into memory using the corresponding loading function. Once the updated model is loaded successfully, the first electronic device can deploy it into the application for use in tasks such as real-time traffic flow prediction, which can include replacing the current prediction model to ensure that the new model can be called by the application and perform actual prediction operations.

[0107] With such a setting, updating the current prediction model to the updated prediction model can be continuously improved and optimized, and the prediction accuracy and effect can be gradually improved, which is conducive to establishing a continuously improving intelligent traffic management system and providing better support for urban traffic flow management.

[0108] S300: Acquire location data of a second electronic device.

[0109] For example, the location data of the second electronic device may be acquired through a GPS chip built into the second electronic device, or the real-time location data of the second electronic device may be acquired through a positioning service on the second electronic device.

[0110] With such a configuration, obtaining the location data of the second electronic device can provide additional feature information, such as the driving direction, speed, and dwell time of the vehicle connected to the second electronic device, which helps to fully understand the vehicle's operating status and road conditions and improve prediction accuracy and reliability.

[0111] S400, when the geographical location reflected by the location data is within the preset geographical column and the road congestion value in the current time period is higher than the first preset value, first road condition information is generated and sent to the second electronic device. The road condition information is used to indicate the congestion level of the road within the preset geographical column, so that the user of the second electronic device can select a travel road.

[0112] For example, a preset geographic column range can be set in the system to determine whether the location of the second electronic device is within the range, and the geographic location in the location data of the second electronic device is compared with the preset geographic column. If the geographic location is within the preset geographic column range and the road congestion value of the current time period is higher than the first preset value, the first road condition information is generated (for example, the user of the second electronic device is within the preset geographic column range, indicating that the user's travel range is within the preset geographic column range. When the user's travel range is congested, that is, the road congestion value of the current time period is higher than the first preset value, the name of the congested road in the current time period, the degree of congestion, or the information of selecting other non-congested roads is sent to the second electronic device). For example, the first road condition information may include the road name, congestion level (such as mild, moderate, severe), and estimated delay time of the current time period. The generated first road condition information is sent to the second electronic device in the form of push notification, SMS, in-app message, etc.

[0113] With such a configuration, when the geographical location reflected by the location data is within the preset geographical column and the road congestion value in the current time period is higher than the first preset value, first road condition information is generated and the road condition information is sent to the second electronic device, so that the current road congestion status can be provided to the user in a timely manner, helping the user to more accurately select a travel route, thereby reducing travel time and avoiding traffic congestion (for example, user usage feedback and behavior data can be collected and used to improve the prediction performance of the deep learning model, thereby continuously optimizing the traffic flow prediction system), thereby improving the prediction accuracy.

[0114] In one possible implementation, see Figure 3In step S400, when the geographical location reflected by the location data is within the preset geographical column and the road congestion value in the current time period is higher than the first preset value, first road condition information is generated and the road condition information is sent to the second electronic device, the real-time traffic flow prediction method based on deep learning also includes:

[0115] S410: If the first electronic device does not receive a road adjustment response from the second electronic device, determine the usage status of the second electronic device, wherein the road adjustment response carries road selection information corresponding to the road condition information.

[0116] It can be understood that the first electronic device can be provided with a responder to obtain the usage status of the second electronic device, which can be achieved by sending a request message to the second electronic device and regularly checking the response of the second electronic device. If the first electronic device does not receive the road adjustment response of the second electronic device within a certain period of time, the usage status of the second electronic device is determined. For example, the first electronic device sends the first road condition information of the current time period to the second electronic device, and the first road condition information includes road selection information, which can provide the user with a non-congested road selection. The second electronic device is interactive, and the user can input information or perform operations on it, and the user's input status can be checked to understand its current usage. For example, if the user stops inputting or operating the device, it means that they have made road adjustments before receiving the first road condition information, and are driving on a non-congested road or do not need to make road adjustments temporarily. For another example, if the first electronic device does not receive the road adjustment response of the second electronic device, it means that the first road condition information fails to be sent, resulting in the first electronic device not receiving the road adjustment response of the second electronic device.

[0117] With this arrangement, when the first electronic device does not receive the road adjustment response from the second electronic device, determining the usage status of the second electronic device can provide data support for subsequent faster corresponding processing and adjustment, thereby improving the system's operating efficiency and response speed.

[0118] In one possible implementation, see Figure 4 S410, when the first electronic device does not receive a road adjustment response from the second electronic device, determining a usage state of the second electronic device includes:

[0119] S411, if the first electronic device does not receive a road adjustment response corresponding to the road condition information from the second electronic device within the second preset time period, the first electronic device obtains the selection data and / or motion data of the second electronic device. The selection data is used to indicate whether the second electronic device uses the navigation application of the second electronic device to select a road, and the motion data is used to indicate whether the driving path of the vehicle connected to the second electronic device deviates from the preset driving path.

[0120] It can be understood that after the first electronic device sends the road condition information, a second preset time is set as the waiting time. If no response is received from the second electronic device within the time, it means that the second electronic device has not received the road condition information. When the second electronic device has not received the road condition information, the selection data of the second electronic device is obtained to indicate whether the second electronic device uses the navigation application of the second electronic device to select the road. The selection data can be obtained through a communication method or an API. At the same time, the motion data of the second electronic device is obtained to indicate whether the driving path of the vehicle connected to the second electronic device deviates from the preset driving path. The motion data may include acceleration, position information, etc., which can be obtained through the sensor or positioning function of the second electronic device.

[0121] With such a configuration, if the first electronic device does not receive a road adjustment response corresponding to the road condition information from the second electronic device within the second preset time period, the first electronic device obtains the selection data and / or motion data of the second electronic device, and can understand the status of the second electronic device in real time, such as the selection status and movement status, which helps the first electronic device to make corresponding operations in a timely manner to cope with changes in traffic conditions and improve real-time performance and accuracy.

[0122] S412, when the selection data indicates that the second electronic device is using a navigation application of the second electronic device to select a road and / or the motion data indicates that a driving path of a vehicle connected to the second electronic device deviates from a preset driving path, it is determined that the usage state of the second electronic device is the first motion state.

[0123] For example, if the selection data indicates that the second electronic device is using the navigation application of the second electronic device to select a road, it means that the user has a travel need after the road selection is completed, so the use state of the second electronic device is determined to be the first motion state. For example, the road selection can be performed using the navigation application on the second electronic device, or the road selection can be performed using the navigation application on the vehicle connected to the second electronic device. Or if the motion data indicates that the driving path of the vehicle connected to the second electronic device deviates from the preset driving path, it means that the user's travel path is different from the preset travel path. For example, the preset travel path can be a travel route that the user has planned before the trip, but the user modified the travel path midway, so the vehicle's driving path deviates from the preset driving path. Therefore, it is determined that the user using the second electronic device needs new first road condition information, so the user can choose to take corresponding actions, such as resending the first road condition information.

[0124] With such a configuration, when the selection data indicates that the second electronic device is using a navigation application of the second electronic device to select a road and / or the motion data indicates that a driving path of a vehicle connected to the second electronic device deviates from a preset driving path, the usage state of the second electronic device is determined to be the first motion state, which helps to timely track information such as the vehicle's position, speed, and direction in traffic management and intelligent transportation systems, so as to more effectively manage and optimize traffic flow.

[0125] S413, when the selection data indicates that the second electronic device does not use the navigation application of the second electronic device to select a road and / or the motion data indicates that the driving path of the vehicle connected to the second electronic device is zero, determining that the usage state of the second electronic device is the second motion state.

[0126] For example, if the selection data indicates that the second electronic device does not use the navigation application of the second electronic device to select a road, it means that the user does not have a travel demand, so the use state of the second electronic device is determined to be the second motion state. Or if the motion data indicates that the driving path of the vehicle connected to the second electronic device is zero, it means that the user has not traveled, for example, the user parks the car in the parking lot, so the driving path of the vehicle is zero, thus, it is determined that the user using the second electronic device does not need the first road condition information, so the user can choose to take corresponding actions, such as not sending the first road condition information.

[0127] With such a configuration, when the selection data indicates that the second electronic device has not used the navigation application of the second electronic device to select a road and / or the motion data indicates that the driving path of the vehicle connected to the second electronic device is zero, the usage state of the second electronic device is determined to be the second motion state, so as to understand the user's travel needs, help the user to plan the route more accurately and avoid congestion or other road problems, thereby improving the user experience.

[0128] S420, when the usage state is the first motion state, the first electronic device sends road condition information to the second electronic device within a first preset time period, wherein the user is using a navigation application of the second electronic device when the second electronic device is in the first motion state.

[0129] For example, the first electronic device can communicate with the second electronic device to obtain information about the currently active application in the second electronic device by calling the relevant API of the second electronic device to determine whether it is a navigation application. For example, if a navigation application activity is obtained (for example, it may be that the user frequently operates the screen of the second electronic device or interacts with the navigation application, such as the user frequently operates the screen of the second electronic device and interacts with the navigation application to select a road in the navigation application), the user is using the navigation application of the second electronic device; if the navigation application is obtained to be dormant, the user is not using the navigation application of the second electronic device. When it is confirmed that the second electronic device is in the first motion state, that is, the user is using the navigation application, the first electronic device can send the first road condition information to the second electronic device within a first preset time period.

[0130] With such a configuration, when the usage state is the first motion state, the first electronic device sends the first road condition information to the second electronic device within the first preset time period, which can help the user plan the route more accurately and avoid congestion or other road problems, thereby improving the user experience.

[0131] S430: When the usage state is the second motion state, the first electronic device does not send the first road condition information to the second electronic device within a second preset time period. When the second electronic device is in the second motion state, the user does not use the navigation application of the second electronic device.

[0132] It can be understood that the first electronic device can communicate with the second electronic device to obtain information about the currently active application in the second electronic device by calling the relevant API to determine whether it is a navigation application. For example, if the navigation application activity is obtained (for example, the user frequently operates the screen of the second electronic device or interacts with the navigation application), the user is using the navigation application of the second electronic device; if the navigation application is obtained to be dormant, the user is not using the navigation application of the second electronic device. When it is confirmed that the second electronic device is in the second motion state, that is, the user is not using the navigation application, the first electronic device may not send the first road condition information to the second electronic device within the second preset time.

[0133] With such a configuration, when the usage state is the second motion state, the first electronic device does not send the first road condition information to the second electronic device within the second preset time period, which can help users plan routes more accurately and avoid congestion or other road problems, thereby improving user experience.

[0134] In one possible implementation, see Figure 5 In step S200, after inputting the traffic data of the current time period into the prediction model and obtaining the road congestion value of each time period within a preset time period after the current time period output by the prediction model, the real-time traffic flow prediction method based on deep learning also includes:

[0135] S260: Based on the traffic data of the current time period, determine whether the road condition of the current time period is a first road condition, wherein the first road condition is a road condition including at least two adjacent intersections within a preset distance.

[0136] It can be understood that based on the collected traffic data, the intersections on the current road are identified, and when the road conditions of at least two adjacent intersections are included within the preset distance, it is determined that the road condition in the current time period is the first road condition.

[0137] With this setting, based on the traffic data of the current time period, it is determined whether the road condition of the current time period is the first road condition. This means that what is evaluated is not just the situation of a single road section, but the traffic conditions of multiple areas. This allows a more comprehensive understanding of the traffic conditions in the area, so that appropriate measures can be taken to intervene, without having to wait for traffic problems to worsen before taking action.

[0138] S270: If the road condition is the first road condition, obtain a first signal device list, wherein the first signal device list includes at least one signal device.

[0139] It is understood that after determining that the road condition is the first condition, the relevant traffic management database or traffic planning system can be queried to obtain the first signal equipment list of the road section. The first signal equipment list records at least one signal equipment on the road section, for example, the signal equipment can be a traffic light.

[0140] With such a setting, if the road condition is the first condition, obtaining the first signal equipment list will help to subsequently use the signal equipment to control the order of vehicle passage and optimize traffic signal timing, thereby improving traffic efficiency and reducing congestion. By accurately obtaining the first signal equipment list, traffic management and optimization measures can be formulated in a more targeted manner.

[0141] S280: Acquire signal device cycle information of each signal device in the first signal device list.

[0142] It can be understood that for each signal device recorded in the first signal device list, its signal device cycle can be obtained. The signal device cycle refers to the length of time it takes for a signal light to complete a complete cycle from green light to yellow light and then to red light. For example, it can be determined by measuring the duration of the signal light in different states.

[0143] With such a setting, obtaining the signal device cycle information of each signal device in the first signal device list can provide data support for the subsequent adjustment of the green, yellow and red light times of the traffic lights according to road flow conditions and traffic demand, so as to improve traffic smoothness, reduce congestion and enhance user experience.

[0144] S290: Synchronize and set based on the signal device cycle information and the first road condition to obtain green wave band control information, wherein the green wave band control information is used to control the signal devices between adjacent intersections so that vehicles can pass through the adjacent intersections continuously and smoothly.

[0145] It can be understood that, first, the cycle information of each signal device is analyzed, including the green light time, the yellow light time, and the red light time. Then, the first road condition, that is, the traffic flow and vehicle speed at adjacent intersections, can be analyzed, which can be obtained through data provided by traffic monitoring equipment, sensors, or traffic management departments. The first road condition information can help determine parameters such as the time and vehicle density required for vehicles to pass through adjacent intersections. Synchronous settings are performed based on the signal device cycle information and the first road condition to obtain green wave band control information. For example, the beat, green light time, and transition time of the signal lights at adjacent intersections can be adjusted to ensure that vehicles can pass through adjacent intersections continuously and smoothly. For example, a vehicle is traveling on a 500-meter road with three adjacent intersections at a speed of 30 km / h, that is, (30 km / h×1000 m / km÷3600 s)=8.33 m / s. The time required for the vehicle to travel on the 500-meter road is 500 m÷8.33 m / s=60 seconds. Based on the time required for the vehicle to travel on the 500-meter road, the traffic light cycle is reset. Since there are three adjacent intersections, the vehicle needs to pass through all three intersections, so the traffic light cycle at each intersection needs to take into account the time it takes for the vehicle to pass and the waiting time. Assuming that the travel time at each intersection is 60 seconds, the entire green wave band requires 6 intersection cycles, that is, 6×60 seconds = 360 seconds. 360 seconds are evenly distributed to 3 intersections, and the signal light cycle at each intersection is 360 seconds / 3=120 seconds. According to the recalculated signal light cycle, the signal light control of the intersection is adjusted: the green light time is set to 60 seconds and the red light time is set to 60 seconds at each intersection to ensure that the vehicle passes through the road within the specified time. In this way, the setting is synchronized based on the signal equipment cycle information and the first road condition to obtain the green wave band control information, which can enable vehicles to pass continuously and smoothly between adjacent intersections, reduce the pause and waiting time, and improve traffic efficiency.

[0146] In one possible implementation, see Figure 6 , S290, performing synchronization setting based on the signal device cycle information and the first road condition to obtain green wave band control information, including:

[0147] S291, when the first road condition indicates that the traffic volume is greater than the first threshold value and the duration value in the signal device cycle information is the first duration value, obtain the first control information in the green wave band control information, wherein the first control information is used to control the signal device to extend the green light duration.

[0148] It can be understood that when the first road condition indicates that the traffic volume is greater than the first threshold, indicating that the first road condition is congested, and the duration value in the signal device cycle information is the first duration value, it means that the duration value in the signal device cycle information cannot satisfy the vehicle to continuously pass through the two intersections, and the first control information in the green wave band control information is generated. For example, the first control information can be to increase the green light time or reduce the red light time so that the vehicle can pass through the road within the signal light cycle.

[0149] In this way, when the first road condition indicates that the vehicle flow is greater than the first threshold value and the duration value in the signal device cycle information is the first duration value, the first control information in the green band control information is obtained, so that the passing time of vehicles passing through the signal device is extended, thereby increasing the number of vehicles passing through the signal device per unit time. This helps to improve the traffic capacity of the signal device, reduce the parking waiting time of vehicles, and alleviate traffic pressure.

[0150] S292, when the first road condition indicates that the vehicle flow is less than the first threshold value and the duration value in the signal device cycle information is the second duration value, obtain the second control information in the green band control information, wherein the second control information is used to control the signal device to shorten the cycle duration of the signal device.

[0151] It can be understood that when the first road condition indicates that the traffic volume is less than the first threshold value, indicating that the first road condition is unobstructed, and the duration value in the signal device cycle information is the second duration value, it means that the duration value in the signal device cycle information has extra time when the vehicle continuously passes through the two intersections, and the second control information in the green wave band control information is generated. For example, the second control information can be to shorten the cycle duration of the signal device so that the vehicle can pass through the road within the signal light cycle while matching the cycle duration of the signal device with the unobstructed road condition.

[0152] With such a setting, when the first road condition indicates that the traffic volume is less than the first threshold and the duration value in the signal device cycle information is the second duration value, the second control information in the green wave band control information is obtained, and vehicles can pass through the signal device more smoothly, which helps to optimize traffic flow, reduce traffic jams and congestion, and improve the overall traffic efficiency of the road.

[0153] In one possible implementation, see Figure 7 In step S290, after synchronization is performed based on the signal device cycle information and the first road condition to obtain green wave band control information, the real-time traffic flow prediction method based on deep learning further includes:

[0154] S293, obtaining bus driving data in the first road condition, wherein the bus driving data is used to indicate the daily departure schedule and average speed data of the bus.

[0155] Understandably, modern buses are equipped with GPS positioning systems that can track the location and operation status of buses in real time. By collecting these GPS data, we can get information such as the driving path, speed and departure time of each bus, and thus calculate the daily departure frequency and average speed data.

[0156] With such a setting, obtaining the bus driving data in the first road condition can provide an important reference for traffic planning, which helps to optimize traffic flow, reduce traffic jams and congestion, and improve the overall traffic efficiency of the road.

[0157] S294, obtaining third control information in the green band control information according to the departure schedule data and the average vehicle speed data, wherein the third control information is used to adjust the phase difference of the signal device.

[0158] It can be understood that by analyzing the collected departure data and average speed data, the signal phase difference can be calculated using methods and algorithms in traffic engineering. The signal phase difference refers to the difference between the phases of adjacent traffic lights (i.e., the duration of the green light). By adjusting the signal phase difference, the coordination of traffic signals can be achieved to form green wave band control. For example, the signal phase difference can be adjusted according to the departure schedule and average speed data of buses, as well as the traffic flow conditions. For example, when the bus density on a certain section of road is high or the traffic speed is slow, the phase difference of the signal lights can be appropriately adjusted to extend the green light time so that the buses can pass through the intersection quickly.

[0159] With such a setting, the third control information in the green wave band control information is obtained according to the departure schedule data and the average vehicle speed data, so as to evaluate the effect of the green wave band control, and then continuously optimize and adjust the strategy to improve traffic efficiency and service quality.

[0160] S295, obtaining traffic management decision information according to the first control information, the second control information and the third control information.

[0161] It can be understood that the first control information and the second control information may include a control strategy for a traffic light, such as green wave band control, which includes the phase difference setting of the traffic signal, the duration of the green light, the timing scheme of the traffic light, etc. The collected data may be the signal control parameters and timing plans of each intersection. The third control information may be a control strategy for a specific traffic demand or event, such as bus priority, emergency vehicle passage, etc. This information may be an adjustment strategy based on bus schedule data, average speed data, and priority passage for special vehicles. The collected first, second, and third control information are comprehensively analyzed. This includes integrating and evaluating data on traffic flow, signal light control schemes, and special traffic demand. By analyzing the data, traffic bottlenecks, congestion points, and traffic management decision information can be identified. For example, the first control information: real-time traffic flow data of Road A and Road B, including hourly traffic flow statistics. Traffic speed data of Road A and Road B. Vehicle queue length and time at each intersection. Second control information: traffic signal light timing scheme at the intersection of Road A and Road B, including green and red light durations. Green wave band control scheme for intersections. Third control information: Bus schedule data, including bus routes and schedules that pass through the intersection of Road A and Road B. Special traffic demand data, such as emergency vehicle priority, pedestrian crossing time, etc. Analyze the traffic flow data of Road A and Road B to identify peak hours and congestion periods. Determine the time periods with the highest traffic flow at the intersection and the reasons for vehicle delays. Evaluate the timing plan of the intersection lights and check whether the green light duration and red light duration match the actual traffic flow. Evaluate the traffic conditions of buses and special vehicles and determine the extent to which traffic congestion affects them. Adjust the green and red light duration of the intersection lights based on traffic flow and peak time data to better match traffic flow. Adjust the bus schedule for passing through the intersection based on the bus schedule data to reduce the time the bus stays at the intersection. Provide priority for emergency vehicles, such as making temporary signal light adjustments during congestion to ensure that emergency vehicles pass quickly.

[0162] With such arrangement, traffic management decision information is obtained according to the first control information, the second control information and the third control information, so as to optimize the operation and service quality of the traffic system.

[0163] In one possible implementation, see Figure 8 , the road includes at least two signal devices. In step S290, after synchronous setting is performed based on the signal device cycle information and the first road condition to obtain green wave band control information, the real-time traffic flow prediction method based on deep learning also includes:

[0164] S296, based on the green band control information, control the first signal device to enter a signal adjustment state, wherein the signal adjustment state is used to remind the user of the time or priority of the signal device.

[0165] It can be understood that the first signal device is controlled according to the green band control information so that the first signal device enters the signal adjustment state. Once the first signal device enters the signal adjustment state, the user can be reminded of the time or priority of the signal device in different ways. For example, a countdown or related prompt information can be displayed on the information panel of the system.

[0166] In this way, based on the green band control information, the first signal device is controlled to enter the signal adjustment state, so that the signal devices can be coordinated within a certain range to form a continuous green light passage time window. In this way, vehicles can pass through multiple adjacent intersections in a smoother way, reducing waiting time and delays, thereby improving overall traffic efficiency.

[0167] S297, if the first electronic device does not receive a control response fed back by any of the two signal devices, when the first signal device enters a signal adjustment state, the first signal device receives a first adjustment signal sent by the second signal device, wherein the first adjustment signal is used to determine the signal duration adjustment of the second signal device.

[0168] It is understandable that the system may have a monitoring mechanism for detecting whether a control response from any one of the first signal device and the second signal device has been received. If the system fails to receive any control response, it may indicate a communication failure or a control failure. When the first signal device enters the signal adjustment state, the first signal device receives a first adjustment signal sent by the second signal device. The first signal device may use the first adjustment signal to determine the signal duration adjustment of the second signal device. The first adjustment signal may include information about the working status of the second signal device, duration adjustment progress, or other relevant information.

[0169] With such a configuration, if the first electronic device cannot receive a control response fed back by any of the two signal devices, when the first signal device enters a signal adjustment state, the first signal device receives a first adjustment signal sent by the second signal device, so that the first signal device and the second signal device can share adjustment data. Through coordination and cooperation between the first signal device and the second signal device, even when the system cannot control the first signal device or the second signal device, collaborative operation of the signal adjustment work can be achieved.

[0170] S298, the first signal device sends a first adjustment response to the second signal device, wherein the first adjustment response is used to instruct the second signal device to maintain a signal adjustment state after receiving the first adjustment response, and the second signal device is in a signal adjustment state when sending the first adjustment signal.

[0171] It can be understood that when the first signal device detects through the first adjustment signal that the second signal device is in a signal adjustment state and needs to send an adjustment response, it will send the first adjustment response to the second signal device in accordance with the communication protocol, and the first adjustment response contains information about the adjustment state and progress of the signal duration of the first signal device. After receiving the first adjustment response, the second signal device needs to parse and process it. It can adjust its own adjustment state according to the information in the first adjustment response to ensure coordination and synchronization with the first signal device, for example, perform further signal adjustment work according to the information in the first adjustment response, or adjust its own working state to adapt to the requirements of the first signal device to achieve collaborative work.

[0172] With such a configuration, the first signal device sends the first adjustment response to the second signal device, which can effectively coordinate and synchronize the work of the two signal devices and avoid traffic congestion, which helps to improve the user's travel efficiency and comfort and enhance the user experience.

[0173] In one possible implementation, see Fig. 9 In step S290, after synchronization is performed based on the signal device cycle information and the first road condition to obtain green wave band control information, the real-time traffic flow prediction method based on deep learning further includes:

[0174] S2991, if the first electronic device cannot receive a control response fed back by any one of the two signal devices, and the first signal device enters a non-signal adjustment state, the first signal device receives a second adjustment signal sent by the second signal device. The second signal device is in a signal adjustment state when sending the second adjustment signal.

[0175] It is understandable that the system may have a monitoring mechanism for detecting whether a control response from any one of the first signal device and the second signal device has been received. If the system fails to receive any control response, it may indicate a communication failure or a control failure. When the first signal device enters a non-signal adjustment state, the first signal device receives a second adjustment signal sent by the second signal device. This may be the case where the first signal and the second signal do not require signal adjustment, so when the first signal device enters a non-signal adjustment state, the second signal also needs to enter a non-signal adjustment state.

[0176] With such a configuration, if the first electronic device cannot receive a control response fed back by any of the two signal devices, and the first signal device enters a non-signal adjustment state, the first signal device receives a second adjustment signal sent by the second signal device, which can enable the first signal device and the second signal device to share adjustment data. Through coordination and cooperation between the first signal device and the second signal device, even when the system cannot control the first signal device or the second signal device, collaborative operation of the signal adjustment work can be achieved.

[0177] S2992: In response to receiving the second adjustment signal, the first signal device does not send a second adjustment response to the second signal device, so that the second signal device exits the signal adjustment state.

[0178] It can be understood that after receiving the second adjustment signal, the first signal device can determine whether it is necessary to send a second adjustment response according to a pre-set protocol or signal analysis mechanism. If it is necessary to make the second signal device exit the signal adjustment state, the first signal device can choose not to send the second adjustment response. In the case where the first signal device does not send the second adjustment response, the second signal device can set a timeout mechanism. If the second adjustment response of the first signal device is not received within a certain period of time, the second signal device automatically exits the adjustment state, so that the first signal and the second signal can be made without signal adjustment.

[0179] With such a configuration, in response to receiving the second adjustment signal, the first signal device does not send a second adjustment response to the second signal device, so that the second signal device exits the signal adjustment state, which can effectively coordinate and synchronize the work of the two signal devices and improve traffic management efficiency.

[0180] In one possible implementation, see Fig.10 In step S400, when the geographical location reflected by the location data is within the preset geographical column and the road congestion value in the current time period is higher than the first preset value, the first road condition information is generated, including:

[0181] S440, dividing the road sections within the preset geographic column to obtain traffic flow data in each area.

[0182] It is understood that the road section range contained in the geographic column can be determined. The road section in the geographic column is divided into multiple areas, each of which can be an intersection, a road section or other specific geographic range. The area can be divided according to the change of traffic flow, the structure of the road network and other relevant factors. Vehicle monitoring equipment, such as traffic cameras, vehicle sensors or vehicle counters, is installed in each divided area. These monitoring equipment can be used to monitor the passing of vehicles in real time and record vehicle flow data.

[0183] With this setting, the road sections within the preset geographic columns are divided and the traffic flow data in each area is obtained, which can help decision makers identify traffic bottleneck areas, optimize road layout and timing plans, thereby improving traffic congestion problems and improving overall traffic efficiency.

[0184] S450: Obtain vehicle speed data from traffic data.

[0185] For example, traffic sensors and monitoring equipment, such as traffic cameras, radars, or sensors, can be installed on roads. These devices can calculate the average speed of vehicles by measuring the time and location of vehicles passing by.

[0186] With this setting, by obtaining vehicle speed data from traffic data, data support can be provided for subsequent optimization of road layout and timing plans, thereby improving traffic congestion problems and improving overall traffic efficiency.

[0187] S460: Generate first road condition information in each area based on the vehicle speed data and the vehicle flow data.

[0188] It can be understood that the average speed of vehicles in each area is calculated and combined with the traffic flow data to generate a traffic condition index. For example, the speed data can be divided into several ranges (such as smooth, slow, and congested), and the traffic condition index can be calculated based on the number of vehicles and the speed distribution. This can provide information about the degree of traffic congestion in different areas. Based on the speed and traffic flow data, the traffic congestion hotspots in each area are identified. By analyzing the reduction in vehicle speed and the increase in traffic flow, the location and severity of the congestion can be determined, which can provide information about which areas have congestion problems, that is, the first road condition information. For example, the speed data can be used to calculate the average speed and the minimum speed, and the traffic flow data can be used to determine the vehicle density on the road. If the average speed is lower than a certain threshold (for example, 40 kilometers per hour) and the vehicle density is higher than a certain level, it can be determined that there is traffic congestion in the area.

[0189] With this setting, road condition information in each area is generated based on vehicle speed data and traffic flow data, which can timely identify and monitor road congestion, so as to take corresponding traffic management measures, such as adjusting traffic light duration, optimizing traffic flow distribution, providing traffic diversion services, etc., to effectively alleviate traffic congestion and improve road traffic efficiency.

[0190] S500, when no road adjustment response is received from the second electronic device within the first time period, when it is detected that the road congestion value of the next time period is higher than the first preset value, and the geographical location reflected by the location data is within the preset geographical column, second road condition information is generated and the road condition information is sent to the second electronic device.

[0191] For example, when no road adjustment response is received from the second electronic device within the first time period, it means that the user of the second electronic device has parked the car in the parking lot and has no travel needs, or something has delayed the current travel plan. The road congestion value of the next time period can be obtained, and the location data of the second electronic device can be checked to confirm whether it is within the preset geographic column range. The road congestion value of the next time period is compared with the first preset value to determine whether the road congestion situation meets the conditions for generating the second road condition information. If the location data is within the geographic column range and the road congestion value is higher than the first preset value, the second road condition information is generated according to the actual situation, including the road name, congestion level, estimated travel time, traffic condition description, etc. The generated second road condition information is sent to the second electronic device through push notifications, text messages, in-app messages, etc., to remind the user of the road conditions of the next time period and suggest a suitable travel route.

[0192] With such a configuration, when no road adjustment response is received from the second electronic device within the first time period, when it is detected that the road congestion value in the next time period is higher than the first preset value and the geographical location reflected by the location data is within the preset geographical column, second road condition information is generated and sent to the second electronic device, which can help the user make better travel decisions in the next time period and avoid traffic congestion, which helps to improve the user's travel efficiency and comfort and enhance the user experience.

[0193] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0194] Corresponding to the real-time traffic flow prediction method based on deep learning described in the above embodiment, the embodiment of the present application also provides a real-time traffic flow prediction system based on deep learning, and each unit of the system can implement each step of the real-time traffic flow prediction method based on deep learning. Fig.11 A structural block diagram of a real-time traffic flow prediction system based on deep learning provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0195] Reference Fig.11 , a real-time traffic flow prediction system based on deep learning, including:

[0196] The first acquisition unit is used to acquire traffic data in real time, wherein the traffic data is used to reflect the traffic volume and road conditions.

[0197] The determination unit is used to input the traffic data of the current time period into the prediction model to obtain the road congestion value of each time period within a preset time period after the current time period output by the prediction model. The prediction model is obtained by deep learning.

[0198] The second acquiring unit is used to acquire the location data of the second electronic device.

[0199] The first generating unit is used to generate first road condition information and send the road condition information to the second electronic device when the geographical location reflected by the location data is within the preset geographical column and the road congestion value in the current time period is higher than the first preset value. The first road condition information is used to indicate the congestion level of the road within the preset geographical column so that the user of the second electronic device can select a travel road.

[0200] The second generating unit is used to generate second road condition information when no road adjustment response is received from the second electronic device within the first time period, and when it is detected that the road congestion value of the next time period is higher than the first preset value and the geographical location reflected by the location data is within the preset geographical column, and send the second road condition information to the second electronic device.

[0201] It should be noted that the information interaction, execution process, etc. between the above-mentioned systems / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0202] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units is used as an example for illustration. In practical applications, the above-mentioned functions can be assigned to different functional units as needed, that is, the internal structure of the system can be divided into different functional units to complete all or part of the functions described above. The functional units in the embodiment can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0203] The embodiment of the present application also provides a first electronic device, Fig.12 This is a schematic diagram of the structure of a first electronic device provided in an embodiment of the present application. Fig.12 As shown, the first electronic device 6 of this embodiment includes: at least one processor 60 ( Fig.12Only one is shown), at least one memory 61 ( Fig.12 Only one is shown) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, the first electronic device 6 implements the steps in any of the above-mentioned embodiments of the real-time traffic flow prediction method based on deep learning, or implements the functions of each unit in the above-mentioned system embodiments.

[0204] Exemplarily, the computer program 62 may be divided into one or more units, which are stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 62 in the first electronic device 6.

[0205] For example, the first electronic device may be a tablet computer, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a desktop computer, a smart screen, or other terminal devices, a computing device with a wireless communication function or other processing devices connected to a wireless modem, a vehicle networking terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, or other mobile terminals. The first electronic device 6 may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will appreciate that Fig.12 It is only an example of the first electronic device 6 and does not constitute a limitation on the first electronic device 6. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, buses, etc.

[0206] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0207] In some embodiments, the memory 61 may be an internal storage unit of the first electronic device 6, such as a hard disk or memory of the first electronic device 6. In other embodiments, the memory 61 may also be an external storage device of the first electronic device 6, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the first electronic device 6. Further, the memory 61 may also include both an internal storage unit and an external storage device of the first electronic device 6. The memory 61 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program, etc. The memory 61 may also be used to temporarily store data that has been output or is to be output.

[0208] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0209] An embodiment of the present application provides a computer program product. When the computer program product runs on a first electronic device, the first electronic device implements the steps in any of the above method embodiments.

[0210] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the first electronic device, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electric carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0211] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0212] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0213] In the embodiments provided in the present application, it should be understood that the disclosed first electronic device and the real-time traffic flow prediction method based on deep learning can be implemented in other ways. For example, the first electronic device embodiment described above is only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0214] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0215] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A real-time traffic flow prediction method based on deep learning, characterized in that: Applied to a first electronic device, the first electronic device is communicatively connected with a second electronic device, and the method includes: Acquire traffic data in real time; wherein the traffic data is used to reflect vehicle flow and road conditions; Input the traffic data of the current time period into a prediction model to obtain the road congestion value of each time period within a preset time period after the current time period output by the prediction model; wherein the prediction model is obtained by deep learning; Acquiring location data of the second electronic device; When the geographical location reflected by the location data is within a preset geographical column and the road congestion value in the current time period is higher than a first preset value, first road condition information is generated and sent to the second electronic device; wherein the road condition information is used to indicate the congestion level of the road within the preset geographical column, so that the user of the second electronic device can select a travel road; When no road adjustment response is received from the second electronic device in the first time period, when it is detected that the road congestion value in the next time period is higher than the first preset value and the geographical location reflected by the position data is within the preset geographical column, second road condition information is generated and sent to the second electronic device; Wherein, when the geographical location reflected by the location data is within the preset geographical column and the road congestion value in the current time period is higher than the first preset value, after generating the first road condition information and sending the first road condition information to the second electronic device, the method further includes: determining the usage state of the second electronic device when the first electronic device does not receive a road adjustment response from the second electronic device; wherein the road adjustment response carries road selection information corresponding to the road condition information; When the usage state is a first motion state, the first electronic device sends the first road condition information to the second electronic device within a first preset time period; wherein the user is using a navigation application of the second electronic device when the second electronic device is in the first motion state; When the usage state is the second motion state, the first electronic device does not send the first road condition information to the second electronic device within a second preset time period; wherein, when the second electronic device is in the second motion state, the user does not use the navigation application of the second electronic device.

2. The real-time traffic flow prediction method based on deep learning as claimed in claim 1, characterized in that: Before inputting the traffic data of the current time period into the prediction model to obtain the road congestion value of each time period within a preset time period after the current time period output by the prediction model, the method further includes: Generate a first sample based on empirical data; wherein the empirical data includes the traffic volume and road conditions of the road acquired in the past time period, the road conditions include weather information and construction road condition information, the weather information includes information on whether it is the rainy season, and the construction road condition information includes information on whether there is a road construction plan and information on the schedule of the construction period; each of the first samples includes a road congestion value corresponding to each time period of the day, and the road congestion value is determined according to the congestion time acquired every day; Adding identification information to the traffic data; wherein the identification information is used to indicate a current road congestion value corresponding to the traffic data in a current time period; Sending the traffic data and the first sample after adding the identification information to a server; Receiving an updated prediction model sent by the server; wherein the updated prediction model is obtained by the server through training based on the received traffic data after adding the identification information and the first sample; The current prediction model is updated to the updated prediction model.

3. The real-time traffic flow prediction method based on deep learning as claimed in claim 1, characterized in that: The determining the usage status of the second electronic device when the first electronic device does not receive the road adjustment response from the second electronic device includes: If the first electronic device does not receive a road adjustment response corresponding to the road condition information from the second electronic device within a second preset time period, the first electronic device obtains selection data and / or motion data of the second electronic device; wherein the selection data is used to indicate whether the second electronic device uses the navigation application of the second electronic device to select a road, and the motion data is used to indicate whether the driving path of the vehicle connected to the second electronic device deviates from the preset driving path; When the selection data indicates that the second electronic device is using a navigation application of the second electronic device to select a road and / or the motion data indicates that a driving path of a vehicle connected to the second electronic device does not deviate from a preset driving path, determining that the use state of the second electronic device is the first motion state; When the selection data indicates that the second electronic device does not use the navigation application of the second electronic device to select a road and / or the motion data indicates that the driving path of the vehicle connected to the second electronic device is zero, the usage state of the second electronic device is determined to be the second motion state.

4. The real-time traffic flow prediction method based on deep learning as claimed in claim 1, characterized in that: After inputting the traffic data of the current time period into the prediction model and obtaining the road congestion value of each time period within a preset time period after the current time period output by the prediction model, the method further includes: Based on the traffic data of the current time period, determining whether the road condition of the current time period is a first road condition; wherein the first road condition is a road condition including at least two adjacent intersections within a preset distance; If the road condition is the first road condition, obtaining a first signal device list; wherein the first signal device list includes at least one signal device; Acquire signal device cycle information of each of the signal devices in the first signal device list; Based on the signal device cycle information and the first road condition, synchronization is performed to obtain green wave band control information; wherein the green wave band control information is used to control the signal devices between adjacent intersections so that vehicles can pass through the adjacent intersections continuously and smoothly.

5. The real-time traffic flow prediction method based on deep learning as claimed in claim 4, characterized in that: The synchronous setting based on the signal device cycle information and the first road condition to obtain green band control information includes: When the first road condition indicates that the vehicle flow is greater than the first threshold value and the duration value in the signal device cycle information is the first duration value, obtaining the first control information in the green wave band control information; wherein the first control information is used to control the signal device to extend the green light duration; When the first road condition indicates that the traffic volume is less than the first threshold value and the duration value in the signal device cycle information is a second duration value, the second control information in the green wave band control information is obtained; wherein the second control information is used to control the signal device to shorten the cycle duration of the signal device.

6. The real-time traffic flow prediction method based on deep learning as claimed in claim 5, characterized in that: After performing synchronization setting based on the signal device cycle information and the first road condition to obtain green band control information, the method further includes: Acquire bus travel data in the first road condition; wherein the bus travel data is used to indicate the daily departure schedule and average speed data of the bus; According to the departure schedule data and the average vehicle speed data, third control information in the green band control information is obtained; wherein the third control information is used to adjust the phase difference of the signal device; Traffic management decision information is obtained according to the first control information, the second control information and the third control information.

7. The real-time traffic flow prediction method based on deep learning as claimed in claim 4, characterized in that: The road includes at least two signal devices, the first electronic device is in communication with the signal device, and after performing synchronization setting based on the signal device cycle information and the first road condition to obtain green wave band control information, the method further includes: Based on the green band control information, controlling the first signal device to enter a signal adjustment state; wherein the signal adjustment state is used to remind the user of the time or priority of the signal device; If the first electronic device cannot receive a control response fed back by any one of the two signal devices, when the first signal device enters a signal adjustment state, the first signal device receives a first adjustment signal sent by the second signal device; wherein the first adjustment signal is used to determine a signal duration adjustment condition of the second signal device; The first signal device sends a first adjustment response to the second signal device; wherein the first adjustment response is used to instruct the second signal device to maintain the signal adjustment state after receiving the first adjustment response, and the second signal device is in the signal adjustment state when sending the first adjustment signal.

8. The real-time traffic flow prediction method based on deep learning as claimed in claim 7, characterized in that: After performing synchronization setting based on the signal device cycle information and the first road condition to obtain green band control information, the method further includes: If the first electronic device cannot receive a control response fed back by any one of the two signal devices, when the first signal device enters a non-signal adjustment state, the first signal device receives a second adjustment signal sent by the second signal device; wherein the second signal device is in the signal adjustment state when sending the second adjustment signal; In response to receiving the second adjustment signal, the first signal device does not send a second adjustment response to the second signal device, so that the second signal device exits the signal adjustment state.

9. A real-time traffic flow prediction system based on deep learning, characterized in that: For implementing the method according to any one of claims 1 to 8, the real-time traffic flow prediction system based on deep learning comprises: A first acquisition unit is used to acquire traffic data in real time; wherein the traffic data is used to reflect the traffic volume and road conditions; A determination unit, configured to input the traffic data of the current time period into a prediction model, and obtain a road congestion value of each time period within a preset time period after the current time period output by the prediction model; wherein the prediction model is obtained by deep learning; A second acquiring unit, configured to acquire location data of the second electronic device; a first generating unit, configured to generate first road condition information when the geographical location reflected by the location data is within a preset geographical column and the road congestion value in the current time period is higher than a first preset value, and send the road condition information to the second electronic device; wherein the first road condition information is used to indicate the congestion level of the road within the preset geographical column, so that a user using the second electronic device can select a travel road; The second generating unit is used to generate second road condition information when no road adjustment response is received from the second electronic device within the first time period, and when it is detected that the road congestion value in the next time period is higher than the first preset value and the geographical location reflected by the location data is within the preset geographical column, and send the second road condition information to the second electronic device.

Images