Real-time highway road condition detection method and system based on Internet of Vehicles and Internet of Things

Through the combination of Internet of Vehicles and Internet of Things, real-time acquisition and sharing of vehicle and traffic facilities data is solved, the problem of single information acquisition path in the existing technology is solved, the accuracy and interactivity of real-time road conditions detection is realized, and the intelligence of modern highways is met. need.

CN118609372BActive Publication Date: 2024-11-01POWER CHINA KUNMING ENG CORP LTD +1
View PDF 13 Cites 0 Cited by

Patent Information

Application Number
CN202411074294.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-11-01
Estimated Expiration
2044-08-07

AI Technical Summary

Technical Problem

In the prior art, there is a single information acquisition path for real-time road conditions detection, which leads to a lack of status information for obtaining vehicles and traffic facilities, and cannot meet the intelligent requirements of real-time road conditions of modern highways.

Method used

Using the Internet of Vehicles and Internet of Things method, the vehicle data is obtained through the vehicle-mounted equipment on the vehicle, the road facility data is obtained by using the Internet of Things sensor equipment, and the wireless communication is transmitted to the central server for storage and processing, and the data analysis algorithm is used to calculate and Predict real-time road conditions information to realize real-time communication and information sharing between vehicles.

Benefits of technology

Real-time acquisition and sharing of vehicle and traffic facilities status information is realized, the accuracy and information interactivity of real-time road conditions detection are improved, the intelligent needs of modern highways are met, and traffic management and user experience are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118609372B_ABST
    Figure CN118609372B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of road condition detection, and discloses a method and system for detecting real-time road conditions of highways based on vehicle networking and the Internet of Things. The method comprises: acquiring vehicle data and road facility data, transmitting the collected vehicle data and road facility data to a central server through wireless communication, and storing them; removing abnormal data from the collected vehicle data and road facility data, and correcting them to obtain fused vehicle data and road facility data; calculating and predicting real-time road condition information based on the fused vehicle data and road facility data using a data analysis algorithm; displaying the calculated and predicted real-time road condition information to vehicle users through a mobile application, providing real-time road condition information and suggestions; verifying and adjusting the real-time road condition; and the system comprises: a data acquisition module, a data processing module, and a result output module. The present invention achieves the purpose of improving traffic efficiency, safety, and user travel experience.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of road condition detection, and in particular to a method and system for real-time highway road condition detection based on Internet of Vehicles and Internet of Things. Background Art

[0002] Real-time highway traffic conditions refer to the acquisition and provision of information on current road traffic conditions through real-time data acquisition and processing technology. Real-time highway traffic conditions can reduce traffic congestion, improve traffic safety, optimize route planning, and provide real-time information services. However, existing real-time highway traffic condition detection has poor real-time communication and information sharing capabilities between vehicles, and is unable to obtain real-time status information of vehicles and traffic facilities, resulting in low accuracy in the acquisition of real-time highway traffic conditions.

[0003] Prior art 1, application number: CN 202310949826.1 discloses a highway road condition warning system based on road network big data, including a monitoring center, the monitoring center is connected to an information collection module, an information processing module and a road condition warning module; the information processing module includes an information analysis unit, an accident location unit and an information storage unit; the real-time road condition information of the highway is obtained through the information collection module, and uploaded to the information processing module for processing, the accident information is obtained, the warning information database is constructed, and the specific location of the accident is determined, and the accident information is uploaded to the road condition warning module for warning; although it realizes warning of congested sections and bad weather conditions, so that drivers can obtain accurate real-time highway road condition information, and choose smooth and safe sections to drive in time, greatly improving the ability of the highway road condition warning system. However, it only involves accident information warning, and the information acquisition path is single, resulting in a lack of status information of vehicles and traffic facilities, and cannot meet the intelligent requirements of modern highway real-time road conditions.

[0004] Prior art 2, application number: CN 202211571415.5 discloses a traffic flow collection, prediction, control and information push system and method based on smart highways, which inputs the current traffic flow and the current average vehicle speed into the speed prediction model to obtain the predicted average vehicle speed, and uses the predicted average vehicle speed to divide the traffic congestion situation, and formulates a collaborative control strategy based on the congestion situation and the real-time road conditions of the highway; the real-time situation of the highway where the vehicle is currently located is matched with the collaborative control strategy, and the vehicle terminal controls the vehicle driving according to the prompts corresponding to the collaborative control strategy. Although customized push services are provided to drivers for vehicle information, thereby alleviating highway traffic congestion and improving inter-city travel efficiency. However, information sharing between vehicles cannot be achieved, resulting in poor information interactivity and poor user experience.

[0005] Prior art three, application number: CN 202311570061.7 discloses a vehicle lane change guidance method, device, equipment and storage medium, including: obtaining the current lane condition information of the target vehicle's driving lane; when the current lane condition information indicates that there is a no-driving zone in the target vehicle's driving lane, determining the current driving area of ​​the target vehicle; when the target vehicle's current driving area is a lane change guidance area, sending lane change request information to the roadside equipment; receiving the guidance status information returned by the roadside equipment; and controlling the target vehicle to drive from the current driving lane to the target driving lane based on the guidance status information. Although judging whether the target vehicle needs to change lanes based on the real-time road conditions of the highway and the driving area of ​​the target vehicle, and controlling the target vehicle to drive into the target driving lane based on the guidance status information when the vehicle needs to change lanes, it can effectively reduce the impact of accidents on multiple lanes and improve road traffic efficiency on the basis of ensuring driving safety. However, it is limited to the guidance of a single vehicle, and cannot realize information sharing of multiple vehicles, resulting in poor functionality, and cannot effectively reflect the real-time road conditions of multiple sections of highways, and the actual application scope is limited.

[0006] At present, the existing technologies 1, 2 and 3 have a single information acquisition path, resulting in a lack of information on the status of vehicles and traffic facilities, and cannot meet the intelligent requirements of modern highway real-time road conditions. Therefore, the present invention provides a real-time road condition detection method and system for highways based on the Internet of Vehicles and the Internet of Things. Through the combination of the Internet of Vehicles and the Internet of Things, the status information of vehicles and traffic facilities, such as traffic flow, road conditions and parking spaces, can be obtained in real time; the combination of the Internet of Vehicles and the Internet of Things can realize real-time communication and information sharing between vehicles, for example, through communication between vehicles, functions such as collaborative driving and collision avoidance can be realized; at the same time, the Internet of Things technology can monitor the status of vehicles in real time, such as vehicle failures and fatigue driving, to provide early warning and safety protection. Summary of the invention

[0007] The main purpose of the present invention is to provide a method and system for real-time highway road condition detection based on the Internet of Vehicles and the Internet of Things, so as to solve the problem that the information acquisition path in the prior art is single, resulting in a lack of status information of vehicles and traffic facilities, and failing to meet the intelligent requirements of modern highway real-time road conditions.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A method for detecting real-time road conditions on a highway based on an Internet of Vehicles and an Internet of Things, the method comprising:

[0010] Obtain vehicle data and road facility data. Vehicle data is obtained through on-board equipment installed on the vehicle; road facility data is obtained through wireless communication using IoT sensor devices arranged on the highway; the collected vehicle data and road facility data are transmitted to the central server through wireless communication and stored;

[0011] The collected vehicle data and road facility data are subjected to abnormal data removal and correction to obtain fused vehicle data and road facility data; based on the fused vehicle data and road facility data, real-time road condition information is calculated and predicted using data analysis algorithms;

[0012] The calculated and predicted real-time traffic information is displayed to vehicle users through mobile applications, providing real-time traffic information and suggestions; the real-time traffic conditions are verified and adjusted through data feedback from vehicle users.

[0013] As a further improvement of the present invention, the interaction process between the vehicle-mounted device and the IoT sensor device includes:

[0014] As a publisher, the vehicle-mounted device registers and publishes data types and topics including the registered vehicle location and speed with the central server, and specifies a corresponding topic for each data type; as a subscriber, the IoT sensor device registers and subscribes to data types and topics including traffic flow and meteorological data with the central server, and specifies a corresponding topic for each data type;

[0015] The vehicle-mounted device publishes the collected vehicle data to the central server according to the topic specified during registration; the data type and topic information need to be attached to the data when publishing; the IoT sensor device subscribes to the data published on the central server according to the topic specified during registration, and specifies the data type and topic when subscribing;

[0016] After the central server receives the data published by the publisher, it sends the data to the subscriber using the push method according to the subscriber's registration information, or the subscriber actively requests the data using the pull method.

[0017] As a further improvement of the present invention, the process of sending data to subscribers by push mode includes:

[0018] After receiving the data published by the publisher, the central server determines which subscribers need to be pushed to according to the data type and topic;

[0019] The central server packages the data and adds identification information of the data type and subject;

[0020] The central server generates a symmetric encryption key through the established communication connection, and the key is used to encrypt the data packet; the central server uses the symmetric encryption key to encrypt the data packet to generate an encrypted data packet; the central server obtains the subscriber's public key, and the public key is used to decrypt the symmetric encryption key; the central server uses the subscriber's public key to asymmetrically encrypt the symmetric encryption key to generate an encrypted symmetric encryption key; the central server sends the encrypted symmetric encryption key and the encrypted data packet to the subscriber, and sends the data packet to the subscriber;

[0021] The subscriber receives the encrypted data packet and the encrypted symmetric encryption key. The subscriber uses his own private key to asymmetrically decrypt the encrypted symmetric encryption key to obtain the symmetric encryption key. The subscriber uses the obtained symmetric encryption key to symmetrically decrypt the encrypted data packet, restore the original data, and parse and process it according to the data type and topic.

[0022] As a further improvement of the present invention, the process of calculating and predicting real-time traffic information using a data analysis algorithm includes:

[0023] Clean and preprocess the collected vehicle data and road facility data, remove outliers and noise, correct the data, match and associate them through timestamps, and align the corrected vehicle data and road facility data;

[0024] After the vehicle data and the road facility data are aligned, the vehicle data and the road facility data are fused by weighted averaging; target features related to vehicle speed, traffic volume and road congestion are extracted from the fused vehicle data and road facility data;

[0025] A recurrent neural network data model is established based on the vehicle data and road facility data extracted from the target features; the established recurrent neural network data model is trained using the historical vehicle data and road facility data; and the real-time vehicle data and road facility data are predicted using the trained recurrent neural network data model to obtain real-time road condition information.

[0026] As a further improvement of the present invention, the process of aligning the corrected vehicle data and road facility data includes:

[0027] Sort the timestamps in the vehicle data and road facility data; define a target timestamp, that is, to find the closest timestamp, initialize two pointers, one pointing to the starting position of the sorted vehicle data, and the other pointing to the starting position of the sorted road facility data; enter a loop traversal until all data are traversed or the closest timestamp is found;

[0028] In each cycle, the difference between the timestamp of the current vehicle data and road facility data and the target timestamp is compared, and the minimum difference is recorded; based on the recorded minimum difference, the data corresponding to the closest timestamp is determined, namely, the vehicle data and road facility data;

[0029] A data association table is established to associate the matched vehicle data and road facility data to ensure that they correspond at the same time point.

[0030] As a further improvement of the present invention, if the timestamp of the current vehicle data is less than the target timestamp, the pointer of the vehicle data is moved back one position and the comparison with the next vehicle data is continued; if the timestamp of the current road facility data is less than the target timestamp, the pointer of the road facility data is moved back one position and the comparison with the next road facility data is continued; if the timestamp of the current vehicle data is greater than the target timestamp, or the timestamp of the current road facility data is greater than the target timestamp, it means that the closest timestamp has been found and the loop ends.

[0031] As a further improvement of the present invention, the process of establishing a data association table includes:

[0032] Create a data association table of a hash table to store the association information of vehicle data and road facility data; traverse the matched vehicle data and road facility data and correspond them according to the timestamp;

[0033] For each matched vehicle data, its timestamp is used as a keyword and the vehicle data is stored in the association table; for each matched road facility data, its timestamp is used as a keyword;

[0034] Find the corresponding vehicle data in the association table and associate the road facility data with the vehicle data.

[0035] As a further improvement of the present invention, the process of fusing vehicle data and road facility data by weighted averaging includes:

[0036] Determine the target features of the vehicle data and road facility data that need to be fused, and normalize the vehicle data and road facility data;

[0037] The correlation index between vehicle data and road facility data is calculated based on the Pearson correlation coefficient. The value range of the correlation coefficient is -1 to 1. The closer the value is to 1, the stronger the positive correlation is. The closer the value is to -1, the stronger the negative correlation is. The closer the value is to 0, the no correlation is. According to the degree of correlation, weights are set for vehicle data and road facility data.

[0038] For each target feature, the vehicle data and road facility data are weighted averaged according to the weight, that is, each data is multiplied by the corresponding weight, and then the sum is taken and divided by the sum of the weights; based on the result of the weighted average, the fused data is obtained, that is, the comprehensive information obtained by fusing the vehicle data and the road facility data.

[0039] As a further improvement of the present invention, the process of establishing a recurrent neural network data model includes:

[0040] Prepare a data set for establishing training data and validation data for a recurrent neural network, the data set including historical vehicle data, road facility data, and corresponding target features; preprocess the prepared data;

[0041] Design the architecture of a recurrent neural network with a gated recurrent unit. The recurrent neural network consists of multiple recurrent layers, each with a certain number of neurons.

[0042] The preprocessed data set is divided into a training set and a validation set. The training set is used to train the recurrent neural network model, and the validation set is used to evaluate the performance of the recurrent neural network model and adjust hyperparameters. The designed recurrent neural network model is trained using the training set. During the training process, the weights and biases of the recurrent neural network model are updated through the back-propagation algorithm.

[0043] Use the validation set to evaluate the trained recurrent neural network model, calculate the performance indicators of the recurrent neural network model on the validation set, and adjust the recurrent neural network based on the evaluation results.

[0044] To achieve the above object, the present invention also provides the following technical solutions:

[0045] A real-time highway road condition detection system based on the Internet of Vehicles and the Internet of Things, which is applied to the real-time highway road condition detection method based on the Internet of Vehicles and the Internet of Things, and the real-time highway road condition detection system based on the Internet of Vehicles and the Internet of Things includes:

[0046] The data acquisition module is used to acquire vehicle data and road facility data. The vehicle data is acquired through the on-board equipment carried by the vehicle; the road facility data is acquired through the IoT sensor equipment arranged on the road through wireless communication; the collected vehicle data and road facility data are transmitted to the central server through wireless communication and stored;

[0047] The data processing module is used to remove abnormal data and correct the collected vehicle data and road facility data to obtain fused vehicle data and road facility data; based on the fused vehicle data and road facility data, the data analysis algorithm is used to calculate and predict real-time road condition information;

[0048] The result output module is used to display the calculated and predicted real-time traffic information to vehicle users through mobile applications, providing real-time traffic information and suggestions; and verifying and adjusting the real-time traffic conditions through data feedback from vehicle users.

[0049] The present invention obtains vehicle data and road facility data, and transmits and stores data, realizes data collection, transmission and storage, and provides a data basis for subsequent data processing and analysis; obtains real-time vehicle and road status information, and provides data support for real-time road condition detection. The vehicle data and road facility data are processed, corrected and fused, and then the real-time road condition information is calculated and predicted based on the fused data. Through data processing and analysis algorithms, effective road condition features are extracted to calculate and predict real-time road conditions; accurate real-time road condition detection is achieved, and accurate road condition information is provided to drivers to help them make travel decisions. The calculated and predicted real-time road condition information is displayed to vehicle users through mobile applications, and verified and adjusted through user feedback, and the real-time road condition information is displayed to users in an intuitive and easy-to-understand manner, providing real-time road condition information and suggestions, improving the travel safety and efficiency of drivers, and helping users choose the best route and avoid congested areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A schematic diagram of the steps of an embodiment of a method for detecting real-time road conditions of a highway based on the Internet of Vehicles and the Internet of Things of the present invention;

[0051] Figure 2 A schematic flow chart of the interaction steps between a vehicle-mounted device and an Internet of Things sensor device in an embodiment of a method for detecting real-time road conditions based on the Internet of Vehicles and the Internet of Things of the present invention;

[0052] Figure 3 A schematic flow chart of the steps of calculating and predicting real-time road condition information by using a data analysis algorithm in one embodiment of a method for detecting real-time road conditions on a highway based on the Internet of Vehicles and the Internet of Things of the present invention;

[0053] Figure 4 A schematic flow chart of the steps of verifying and adjusting the real-time road condition according to an embodiment of the method for detecting the real-time road condition of a highway based on the Internet of Vehicles and the Internet of Things of the present invention;

[0054] Figure 5 It is a functional module diagram of an embodiment of a highway real-time road condition detection system based on the Internet of Vehicles and the Internet of Things of the present invention;

[0055] Figure 6 It is a structural schematic diagram of an embodiment of an electronic device of the present invention;

[0056] Figure 7 It is a schematic structural diagram of an embodiment of the storage medium of the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] The terms "first", "second" and "third" in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first", "second" and "third" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0059] Reference to an "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0060] like Figure 1 As shown, this embodiment provides an embodiment of a method for detecting real-time road conditions based on the Internet of Vehicles and the Internet of Things. In this embodiment, the method for detecting real-time road conditions based on the Internet of Vehicles and the Internet of Things specifically includes the following steps:

[0061] Step S1: Acquire vehicle data and road facility data. The vehicle data is acquired through the on-board equipment carried by the vehicle; the road facility data is acquired through the IoT sensor equipment arranged on the highway through wireless communication; the collected vehicle data and road facility data are transmitted to the central server through wireless communication and stored;

[0062] Step S2: removing abnormal data from the collected vehicle data and road facility data, and correcting them to obtain fused vehicle data and road facility data; based on the fused vehicle data and road facility data, using data analysis algorithms to calculate and predict real-time road condition information, such as traffic congestion level, vehicle flow, average driving speed, etc.;

[0063] Step S3: Displaying the calculated and predicted real-time traffic information to vehicle users through a mobile application, providing real-time traffic information and suggestions; verifying and adjusting the real-time traffic conditions through data fed back by vehicle users;

[0064] Among them, the following methods can be adopted to display real-time traffic information and suggestions to users through applications:

[0065] Map display: On the map interface of the application, different symbols or colors are used to indicate the traffic conditions in different areas. For example, green indicates smooth traffic, yellow indicates slow traffic, and red indicates congestion. Users can zoom in and out and drag the map to view the real-time traffic conditions in different areas;

[0066] Traffic list: Provide a traffic list in the application, showing the real-time traffic information of the current road according to the distance from the user's current location or the destination set by the user. Users can browse the list to understand the traffic conditions of different roads so as to make corresponding driving decisions;

[0067] Traffic push notifications: The app can send users push notifications of traffic information on a regular basis. Users can set areas or roads of interest in the app and receive relevant traffic push notifications. This way, users can learn about the real-time traffic conditions in the area of ​​interest in a timely manner so that they can make corresponding adjustments.

[0068] Real-time route planning: The application can provide real-time route planning and navigation suggestions based on real-time traffic information. When the user enters the starting point and destination, the application will calculate the best driving route based on real-time traffic information and provide navigation guidance to avoid congested areas and peak hours, allowing users to choose a faster and more efficient driving route;

[0069] Driving suggestions: The application can give corresponding driving suggestions based on real-time traffic information. For example, in congested areas, the application can remind users to choose a detour or change travel time. In road construction areas, the application can remind users to slow down or choose other roads. In the above ways, the application can show users real-time traffic information and suggestions, help users understand road conditions in a timely manner, make appropriate driving decisions, and improve travel safety and efficiency.

[0070] Preferably, step S1 of this embodiment acquires vehicle data and road facility data, and transmits and stores data, realizes data collection, transmission and storage, and provides a data basis for subsequent data processing and analysis; acquires real-time vehicle and road status information, and provides data support for real-time road condition detection. Step S2 processes, corrects and fuses vehicle data and road facility data, and then calculates and predicts real-time road condition information based on the fused data, extracts effective road condition features through data processing and analysis algorithms, and performs real-time road condition calculation and prediction; realizes accurate real-time road condition detection, and provides accurate road condition information for drivers to help them make travel decisions. Step S3 displays the calculated and predicted real-time road condition information to vehicle users through mobile applications, and verifies and adjusts it through user feedback, and displays the real-time road condition information to users in an intuitive and easy-to-understand manner, providing real-time road condition information and suggestions, improving the travel safety and efficiency of drivers, and helping users choose the best route and avoid congested areas.

[0071] In summary, this embodiment obtains vehicle data and road facility data, processes and analyzes the data, and finally displays real-time road condition information to users, thereby improving traffic efficiency, safety and user travel experience. At the same time, it also provides real-time traffic situation awareness for traffic management departments and provides decision support for traffic scheduling and planning.

[0072] This embodiment can obtain the status information of vehicles and traffic facilities in real time, such as traffic flow, road conditions, parking spaces, etc., through the combination of the Internet of Vehicles and the Internet of Things; this information can be used by the traffic management department for real-time scheduling and optimization, reduce congestion, and improve traffic efficiency. Improved vehicle safety: The combination of the Internet of Vehicles and the Internet of Things can realize real-time communication and information sharing between vehicles. For example, through communication between vehicles, functions such as collaborative driving and collision avoidance can be realized. At the same time, the Internet of Things technology can monitor the status of vehicles in real time, such as vehicle failures, fatigue driving, etc., to provide early warning and safety guarantees. Reduce energy consumption and environmental pollution: The combination of the Internet of Vehicles and the Internet of Things can realize the monitoring and management of vehicle energy consumption, such as obtaining the fuel consumption and driving conditions of the vehicle in real time, optimizing driving strategies, reducing fuel consumption and emissions, and reducing pollution to the environment. Provide intelligent travel services: Through the combination of the Internet of Vehicles and the Internet of Things, functions such as intelligent navigation, positioning and route planning can be realized, and personalized travel services can be provided to drivers. For example, according to the travel needs and traffic conditions of the driver, intelligent recommendations for the best routes, parking spaces and other services can be made to improve the convenience and efficiency of travel. Realizing smart city transportation: The combined use of Internet of Vehicles and Internet of Things can provide support for the construction of smart city transportation. By connecting and integrating transportation facilities, traffic management systems, vehicles and drivers, the sharing and intelligent management of traffic information can be realized, thus improving the operating efficiency and service quality of urban transportation.

[0073] In summary, the combined use of the Internet of Vehicles and the Internet of Things in this embodiment can achieve multiple benefits, including traffic management optimization, vehicle safety improvement, energy consumption and environmental pollution reduction, intelligent travel services, and smart city traffic construction. By fully utilizing the advantages of these two technologies, a more intelligent, efficient, and sustainable transportation system can be achieved.

[0074] Furthermore, if Figure 2 As shown, the interaction process between the vehicle-mounted device and the IoT sensor device in step S1 specifically includes the following steps:

[0075] Step S11: The vehicle-mounted device, as a publisher, registers and publishes data types and topics including the registered vehicle location and speed data to the central server; and specifies a corresponding topic for each data type; the IoT sensor device, as a subscriber, registers and subscribes to data types and topics including traffic flow and meteorological data with the central server, and specifies a corresponding topic for each data type;

[0076] Step S12: The vehicle-mounted device publishes the collected vehicle data to the central server according to the subject specified during registration; the data type and subject information need to be attached to the data during publication; the IoT sensor device subscribes to the data published on the central server according to the subject specified during registration, and specifies the data type and subject during subscription;

[0077] Step S13: After receiving the data published by the publisher, the central server sends the data to the subscriber using a push method according to the subscriber's registration information, or the subscriber actively requests the data using a pull method.

[0078] Preferably, in step S11 of this embodiment, the vehicle-mounted device and the IoT sensor device register the data types and topics to be published and subscribed with the central server, realize the recording and management of the registration information, and establish associations for subsequent data publishing and subscription; through the accurate recording of the registration information, ensure that the vehicle-mounted device and the IoT sensor device can correctly publish and subscribe to the data types and topics of interest. Step S12 The vehicle-mounted device publishes the collected vehicle data to the central server according to the topic specified during registration; the IoT sensor device subscribes to the data published on the central server according to the topic specified during registration, realizes the publication of vehicle data and the subscription of IoT sensor devices, and ensures the transmission and reception of data; through the publication and subscription of data, realizes the data interaction between the vehicle-mounted device and the IoT sensor device, and provides a data basis for real-time road condition detection. Step S13 After the central server receives the data published by the publisher, according to the registration information of the subscriber, it uses the push method to send data to the subscriber, or the subscriber uses the pull method to actively request data, realizes the transmission and reception of data, and ensures that the data can be accurately delivered to the subscriber; through the transmission and reception of data, realizes the data interaction between the vehicle-mounted device and the IoT sensor device, and provides data support for real-time road condition detection.

[0079] In summary, the interaction process between the vehicle-mounted device and the IoT sensor device of this embodiment realizes data transmission and reception through registration, publishing and subscription, providing data basis and support for real-time road condition detection; by accurately publishing and subscribing to data types and topics of interest, data interaction between the vehicle-mounted device and the IoT sensor device is realized, providing data support for calculating and predicting real-time road condition information, improving traffic efficiency, safety and user travel experience. Through the above publish-subscribe mode interaction process, the vehicle-mounted device publishes vehicle data as a publisher, the IoT sensor device subscribes to the data of interest as a subscriber, and the central server serves as a transfer station for information, realizing data interaction between the vehicle-mounted device and the IoT sensor device; wireless transmission and storage of vehicle data and road facility data can be realized, providing a data basis for subsequent data processing and analysis, and then calculating and predicting real-time road condition information, and displaying the results to vehicle users through mobile applications. At the same time, the feedback data of vehicle users can also be uploaded to the central server through wireless communication for verification and adjustment of real-time road conditions.

[0080] Furthermore, the process of sending data to the subscriber by push in step S13 specifically includes the following steps:

[0081] Step S131: After receiving the data published by the publisher, the central server determines which subscribers need to be pushed according to the data type and subject;

[0082] Step S132: The central server packages the data and adds identification information of the data type and subject;

[0083] Step S133: The central server generates a symmetric encryption key through the established communication connection, and the key is used to encrypt the data packet; the central server uses the symmetric encryption key to encrypt the data packet to generate an encrypted data packet; the central server obtains the subscriber's public key, and the public key is used to decrypt the symmetric encryption key; the central server uses the subscriber's public key to asymmetrically encrypt the symmetric encryption key to generate an encrypted symmetric encryption key; the central server sends the encrypted symmetric encryption key and the encrypted data packet to the subscriber, and sends the data packet to the subscriber;

[0084] Step S134: The subscriber receives the encrypted data packet and the encrypted symmetric encryption key, and uses his own private key to asymmetrically decrypt the encrypted symmetric encryption key to obtain the symmetric encryption key; the subscriber uses the obtained symmetric encryption key to symmetrically decrypt the encrypted data packet to restore the original data. Parse and process according to the data type and subject.

[0085] In addition, the process of actively requesting data by the subscriber using the pull method (the encryption and decryption process refers to the push method) specifically includes the following steps:

[0086] The subscriber actively sends data requests to the central server, including information such as data type and subject;

[0087] After receiving the request, the central server searches for the corresponding data according to the data type and subject of the request;

[0088] The central server packages the requested data and adds identification information of the data type and subject;

[0089] The central server sends the data packet to the subscriber through the established communication connection;

[0090] After receiving the data packet, the subscriber parses and processes it according to the data type and topic.

[0091] Preferably, step S131 of this embodiment determines which subscribers need to be pushed according to the data type and subject, and screens and matches the data according to the attributes of the data and the needs of the subscribers to ensure that the data is sent only to the relevant subscribers, which can reduce unnecessary data transmission and processing and improve the efficiency and performance of the system. Step S132 packages the data, adds the identification information of the data type and subject, organizes and encapsulates the data, and adds the identification information to facilitate the subscriber to parse and process the data after receiving it; through the identification information, the subscriber can identify the type and subject of the data and process it in a targeted manner. Step S133 sends the encrypted data packet and the encrypted symmetric encryption key to the subscriber through the established communication connection, realizes the transmission of data through the established communication connection, symmetric encryption of the data packet and asymmetric encryption of the symmetric encryption key, which can ensure the confidentiality of the data and the security of transmission. Step S134 The subscriber receives the encrypted data packet and the encrypted symmetric encryption key, uses the private key to asymmetrically decrypt the symmetric encryption key, and obtains the symmetric encryption key; then uses the symmetric encryption key to symmetrically decrypt the data packet to restore the original data. Parsing and processing are performed according to the data type and subject. Subscribers can restore and parse the encrypted data by decrypting the symmetric encryption key and symmetric decrypting the data packet. Through data parsing and processing, subscribers can obtain the required information and perform corresponding business operations.

[0092] In summary, the purpose of these steps in this embodiment is to ensure the secure transmission, confidentiality and integrity of data, and at the same time, through data screening, packaging and parsing, subscribers can obtain the required data to meet their business needs; sending data to subscribers in a push manner can deliver data in real time and efficiently, improving the response speed of the system and user experience. At the same time, encryption and decryption technology is used to protect the security of data and prevent unauthorized visitors from obtaining data content.

[0093] In this embodiment, the difference between the push mode and the pull mode in the interaction process is that the push mode is that the central server actively sends data to the subscriber, while the pull mode is that the subscriber actively requests data from the central server; the push mode is suitable for scenarios with high real-time requirements, and data can be pushed to subscribers in a timely manner; while the pull mode is suitable for scenarios where subscribers need to obtain data regularly or on demand, and data can be actively pulled as needed. Through the push mode and the pull mode, vehicle-mounted devices and IoT sensor devices can flexibly interact with data according to actual needs, provide data support for real-time road condition detection, and improve traffic efficiency, safety and user travel experience.

[0094] Furthermore, if Figure 3 As shown, the process of calculating and predicting real-time traffic information using the data analysis algorithm in step S2 specifically includes the following steps:

[0095] Step S21: cleaning and preprocessing the collected vehicle data and road facility data, removing outliers and noise, and correcting the data, matching and associating through timestamps, and aligning the corrected vehicle data and road facility data;

[0096] Step S22: after the vehicle data and the road facility data are aligned, the vehicle data and the road facility data are fused by weighted averaging; target features related to vehicle speed, traffic volume, road congestion, etc. are extracted from the fused vehicle data and road facility data;

[0097] Step S23: Establish a recurrent neural network data model based on the vehicle data and road facility data extracted from the target features; use historical vehicle data and road facility data to train the established recurrent neural network data model; use the trained recurrent neural network data model to predict the real-time vehicle data and road facility data to obtain real-time road condition information.

[0098] Preferably, step S21 of this embodiment cleans, preprocesses and aligns the collected vehicle data and road facility data to ensure the accuracy and consistency of the data, remove outliers and noise, correct the data, and align the vehicle data and road facility data; the quality of the data can be improved, providing a reliable basis for subsequent data fusion and analysis. Step S22 performs weighted average fusion on the aligned vehicle data and road facility data, and extracts data related to the target features; fuses information from different data sources, comprehensively considers the weights of vehicle data and road facility data, and obtains more comprehensive and accurate data; extracting target features helps in subsequent data analysis and model building. Step S23 establishes a recurrent neural network (RNN) data model, and uses historical vehicle data and road facility data for training. Through the RNN model, real-time vehicle data and road facility data are predicted to obtain real-time road condition information; the patterns and rules of historical data are used to predict future road conditions, and the calculation and prediction of real-time road condition information is realized.

[0099] In step S23, the vehicle data parameters change:

[0100] Vehicle speed: the vehicle's current speed, usually in kilometers per hour (km / h);

[0101] Traffic volume: the number of vehicles passing through a certain section of road per unit time, usually expressed in vehicles / hour;

[0102] Location data: vehicle GPS location information, providing the specific location of the vehicle on the road (longitude, latitude);

[0103] Driving direction: The specific direction in which the vehicle is traveling, usually expressed as an angle or direction;

[0104] Acceleration: The rate at which a vehicle's speed changes per unit time, reflecting the acceleration or deceleration of the vehicle;

[0105] License Plate Recognition Data: A unique identifier used to identify a specific vehicle;

[0106] Travel time: the time it takes a vehicle to travel a specific road section;

[0107] Road facility data parameter changes:

[0108] Road type: classification of roads, such as expressways, urban roads, rural roads, etc.;

[0109] Road width: The width of the road, which helps to assess vehicle passing capacity;

[0110] Traffic light status: the current working status of the traffic light (such as red light, green light, yellow light);

[0111] Road congestion level: road utilization rate or congestion level based on real-time monitoring;

[0112] Road segment length: The length of the road segment that needs to be predicted for flow calculation;

[0113] Weather conditions: including temperature, precipitation, haze, etc., which may affect road capacity;

[0114] Model training process:

[0115] In steps S21 to S23, a recurrent neural network (RNN) is used to predict road condition information, and the specific process is as follows:

[0116] Data cleaning and preprocessing: Clean the collected vehicle data (such as speed, traffic volume, location, etc.) and road facility data (such as road type, signal light status, etc.) to remove missing values ​​and outliers; align the data associated with the timestamp to ensure the synchronization of vehicle and facility data;

[0117] Data fusion: In step S22, the cleaned vehicle data and road facility data are integrated using weighted averaging or other fusion methods to generate a unified data set. This process takes into account the importance of each data type to maximize the fusion effect; extract important target features such as vehicle speed, flow rate, and road congestion, which will serve as important inputs for model training;

[0118] Model establishment and training: In step S23, the extracted target features are used to construct a recurrent neural network (RNN) model. RNN can process time series data and is suitable for predicting dynamically changing road conditions. The established RNN model is trained using historical vehicle data and road facility data. The process includes dividing the data into a training set and a validation set, and using back propagation to adjust the model weights to minimize the prediction error. Real-time prediction: Once the model training is completed, the trained model is used to predict new real-time vehicle data and road facility data. By inputting the current state (such as real-time vehicle speed and flow), the predicted road condition information is output, such as traffic congestion in the next few minutes.

[0119] In summary, the purpose of these steps in this embodiment is to build a data model and predict real-time traffic information through data cleaning, fusion and analysis; through real-time traffic information, vehicle users can understand the degree of road congestion, vehicle flow and average driving speed, etc., so as to make better driving decisions; for traffic management departments, real-time traffic information can help them better control and optimize traffic, improve road operation efficiency and traffic safety. Therefore, by calculating and predicting real-time traffic information through data analysis algorithms, beneficial traffic services and decision support can be provided to vehicle users and traffic management departments.

[0120] In the whole process of this embodiment, the key is to select appropriate features, establish an effective recurrent neural network data model, and conduct sufficient data training and verification. By continuously optimizing algorithms and models, the accuracy and practicality of real-time traffic information can be improved, thereby providing better traffic services and user experience. Through the above steps, vehicle data and road facility data can be effectively integrated to obtain comprehensive real-time traffic information; accurate and comprehensive traffic information can be provided to vehicle users to help them make better driving decisions and improve the operating efficiency and traffic safety of the road network.

[0121] Furthermore, the process of aligning the corrected vehicle data and road facility data in step S21 specifically includes the following steps:

[0122] Step S211: sort the timestamps in the vehicle data and the road facility data; define a target timestamp, that is, to find the closest timestamp, initialize two pointers, one pointing to the starting position of the sorted vehicle data, and the other pointing to the starting position of the sorted road facility data; enter a loop traversal until all data are traversed or the closest timestamp is found (in each data set, the timestamp with the smallest difference before and after the target timestamp can also be a candidate for multiple timestamps, but the specific implementation logic of the final selection will determine the selected timestamp);

[0123] Step S212: In each cycle, compare the difference between the timestamps of the current vehicle data and road facility data and the target timestamp, and record the minimum difference; according to the recorded minimum difference, determine the data corresponding to the closest timestamp, that is, the vehicle data and road facility data;

[0124] If the timestamp of the current vehicle data is less than the target timestamp, move the pointer of the vehicle data back one position and continue to compare the next vehicle data; if the timestamp of the current road facility data is less than the target timestamp, move the pointer of the road facility data back one position and continue to compare the next road facility data; if the timestamp of the current vehicle data is greater than the target timestamp, or the timestamp of the current road facility data is greater than the target timestamp, it means that the closest timestamp has been found and the loop ends;

[0125] Step S213: Establish a data association table to associate the matched vehicle data and road facility data to ensure that they correspond at the same time point.

[0126] Preferably, step S211 of this embodiment sorts the timestamps in the vehicle data and road facility data, and searches for the data closest to the target timestamp, arranges the data in chronological order, and finds the data closest to the target timestamp, providing an orderly data basis for subsequent data alignment, ensuring the consistency and accuracy of the data. In step S212, in the loop traversal, compares the difference between the timestamp of the current vehicle data and road facility data and the target timestamp, and records the minimum difference, determines the data closest to the target timestamp according to the minimum difference; finds the data closest to the target timestamp, ensures the alignment accuracy of the data, provides accurate alignment data, and provides a reliable basis for subsequent data fusion and analysis. Step S213 establishes a data association table, associates the matched vehicle data and road facility data, corresponds the vehicle data and road facility data at the same time point, ensures the consistency of the data, provides associated data, and provides a reliable basis for subsequent data fusion and analysis.

[0127] In summary, this embodiment aligns the corrected vehicle data and road facility data through the above steps, which can ensure that the data corresponds at the same time point and improve the consistency and accuracy of the data; it can provide a reliable data basis for subsequent data fusion and analysis, and further calculate and predict real-time road condition information; for vehicle users and traffic management departments, accurate alignment data helps to provide more accurate road condition information, helping them make better driving decisions and traffic management decisions.

[0128] Furthermore, the process of establishing the data association table in step S213 includes the following steps:

[0129] Step S2131: Create a data association table of a hash table to store association information of vehicle data and road facility data; traverse the matched vehicle data and road facility data and correspond them according to the timestamp;

[0130] Step S2132: For each matched vehicle data, use its timestamp as a keyword and store the vehicle data in the association table; for each matched road facility data, use its timestamp as a keyword;

[0131] Step S2133: Find the corresponding vehicle data in the association table, and associate the road facility data with the vehicle data.

[0132] Preferably, step S2131 of this embodiment creates a data association table of a hash table for storing association information of vehicle data and road facility data, creates a data structure for storing association information, facilitates subsequent data association and query operations, provides an efficient storage method for establishing associations between vehicle data and road facility data, and speeds up data processing and analysis. Step S2132 uses its timestamp as a keyword for each matched vehicle data, stores the vehicle data in the association table, stores the vehicle data by timestamp, facilitates subsequent query and association operations based on timestamps, establishes a corresponding relationship between the vehicle data and the association table, and provides a basis for subsequent data association and query. Step S2133 finds the corresponding vehicle data in the association table, associates the road facility data with the vehicle data, searches for the corresponding vehicle data in the association table according to the timestamp, associates the road facility data with the vehicle data, and corresponds and associates the matched vehicle data and road facility data at the same time point to ensure data consistency and accuracy.

[0133] In summary, through the above steps, this embodiment establishes a data association table, which can easily find and obtain the matched data, and help further data processing and analysis; it can provide a reliable data basis for subsequent data fusion and analysis, such as calculating real-time road condition information, predicting traffic conditions, etc. For vehicle users and traffic management departments, accurate data association helps to provide more accurate road condition information, helping them make better driving decisions and traffic management decisions. By establishing a data association table, the matched vehicle data and road facility data can be corresponded and associated at the same time point. This ensures the consistency of vehicle data and road facility data, and provides a reliable data basis in subsequent data fusion and analysis. The association table can easily find and obtain the matched data, and help further data processing and analysis, such as calculating real-time road condition information, predicting traffic conditions, etc.

[0134] Furthermore, the process of fusing the vehicle data and the road facility data by weighted averaging in step S22 specifically includes the following steps:

[0135] Step S221: determining target features of vehicle data and road facility data to be fused, and normalizing the vehicle data and road facility data;

[0136] Step S222: Calculate the correlation index between the vehicle data and the road facility data according to the Pearson correlation coefficient, where the value range of the correlation coefficient is from -1 to 1, and the closer the value is to 1, the stronger the positive correlation is, the closer the value is to -1, the stronger the negative correlation is, and the closer the value is to 0, the no correlation is; according to the degree of correlation, set weights for the vehicle data and the road facility data;

[0137] Among them, the Pearson correlation coefficient is calculated. The Pearson correlation coefficient is an indicator to measure the degree of linear correlation between two variables. Its calculation formula is: ;

[0138] in is a variable and The covariance of and They are and The standard deviation of

[0139] Extract target features from vehicle data and road facility data and ;

[0140] Calculate the mean and ;

[0141] Calculating covariance ;

[0142] Calculate standard deviation and ;

[0143] Substitute into the formula to calculate r;

[0144] Based on the value of the Pearson correlation coefficient, the weights are set according to the following rules:

[0145] Positive correlation r) is close to 1,

[0146] If r is between [0.5, 1], it indicates a strong positive correlation, so higher weights are given to vehicle data and road facility data.

[0147] Negative correlation r is close to -1;

[0148] If r is between [-1, -0.5], it indicates a strong negative correlation, and you can consider giving it a lower or opposite weight;

[0149] No correlation r is close to 0;

[0150] If r is between [-0.5, 0.5], it means there is no or weak correlation. In this case, you can consider setting the weights to the same or fixed values.

[0151] Assume that the weight of vehicle data is set to , the weight of road facility data is set to :

[0152] The weighting formula can be set as follows:

[0153] This ensures that the weights sum to 1, i.e. ( + = 1 ).

[0154] For example, if the calculated value is ( r = 0.8 ): ,

[0155] If you find that the set weights do not meet actual needs (for example, the weight values ​​are too concentrated), you can make appropriate adjustments based on the specific business scenario;

[0156] Step S223: For each target feature, perform weighted averaging of the vehicle data and the road facility data according to the weights, i.e., multiply each data by the corresponding weight, then sum and divide by the sum of the weights; based on the result of the weighted average, obtain the fused data, i.e., the comprehensive information obtained by fusing the vehicle data and the road facility data.

[0157] Preferably, step S221 of this embodiment determines the target features of the vehicle data and road facility data to be fused, normalizes the vehicle data and road facility data, determines the target features to be fused, and normalizes the vehicle data and road facility data so that the data have the same dimension and range; provides a unified data basis for subsequent weighted averaging to ensure the consistency of the data during the fusion process. Step S222 calculates the correlation index between the vehicle data and the road facility data according to the Pearson correlation coefficient, and sets weights for the vehicle data and the road facility data according to the degree of correlation; quantifies the degree of correlation between the vehicle data and the road facility data by calculating the correlation index, and sets weights for the data according to the degree of correlation; determines the weight according to the correlation between the data, so that the importance of the data is more reasonably considered during the fusion process, and the accuracy and reliability of data fusion are improved. Step S223 performs weighted averaging of vehicle data and road facility data for each target feature according to weights to obtain fused data; performs weighted averaging of vehicle data and road facility data according to set weights to fuse the two to obtain comprehensive information; comprehensively utilizes the information of vehicle data and road facility data to obtain more accurate and comprehensive data fusion results, providing a reliable data basis for subsequent calculation and prediction of road condition information.

[0158] In summary, through the above steps, this embodiment uses weighted average to fuse vehicle data and road facility data, which can comprehensively utilize the information of both and improve the accuracy and reliability of the data. By setting weights, the importance and relevance of the data can be better considered, making the fusion result more reasonable and accurate. The fused data can be used as input to calculate and predict real-time road condition information, helping traffic management departments and vehicle users to make better driving decisions and traffic management decisions. By using weighted average to fuse vehicle data and road facility data, the information of both can be comprehensively utilized to improve the accuracy and reliability of the data. The weighted average method can be flexibly adjusted according to actual conditions, and different importance can be assigned according to the weight of the data, thereby obtaining a more reasonable fusion result. The fused data can be used as input to establish a data model for calculation and prediction of real-time road condition information.

[0159] Furthermore, the process of establishing the recurrent neural network data model in step S23 includes the following steps:

[0160] Step S231: preparing a data set for establishing training data and verification data including a recurrent neural network, wherein the data set includes historical vehicle data, road facility data, and corresponding target features; preprocessing the prepared data;

[0161] Step S232: designing the architecture of the recurrent neural network of the gated recurrent unit, the recurrent neural network consisting of multiple recurrent layers, each recurrent layer having a certain number of neurons;

[0162] Among them, the recurrent neural network architecture of the gated recurrent unit GRU is composed of the following components:

[0163] Input Gate: The input gate determines how much of the input data at the current moment will be remembered; it uses a sigmoid activation function to output a value between 0 and 1 to represent the importance of the input data;

[0164] Update Gate: The update gate determines how much of the output state of the previous loop layer will be retained and passed to the current moment. It also outputs a value between 0 and 1 through a sigmoid activation function to control the forgetting and retention of the state of the previous moment.

[0165] Candidate State: The candidate state is a temporary state calculated based on the current input and the output state of the previous loop layer, which is used to calculate the output state at the current moment;

[0166] Output State: The output state is the final output of the recurrent neural network at the current moment, which is jointly determined by the input gate, update gate, and candidate state. It is obtained by multiplying the candidate state by the output of the input gate and the output state of the previous moment by the output of the update gate, and then adding them together. The function of GRU is to introduce a gating mechanism into the recurrent neural network to solve the problem of long-term dependencies. Through the control of the input gate and update gate, GRU can decide whether to ignore or retain past information, so as to better capture long-term dependencies in sequence data. Compared with traditional recurrent neural networks, GRU has fewer parameters and better memory capacity, and can better process time series data, including vehicle data and road facility data, and improve the prediction accuracy and generalization ability of the model;

[0167] Step S233: Divide the preprocessed data set into a training set and a validation set, the training set is used to train the recurrent neural network model, and the validation set is used to evaluate the performance of the recurrent neural network model and adjust hyperparameters; use the training set to train the designed recurrent neural network model; during the training process, update the weights and biases of the recurrent neural network model through the back propagation algorithm;

[0168] Step S234: Use the validation set to evaluate the trained recurrent neural network model, calculate the performance indicators of the recurrent neural network model on the validation set, including the root mean square error (RMSE), mean absolute error (MAE), etc., and adjust the recurrent neural network according to the evaluation results.

[0169] Preferably, step S231 of this embodiment prepares training data and verification data for establishing a recurrent neural network model, and preprocesses the data, prepares a data set suitable for training, and preprocesses the data, such as normalization, standardization, feature selection, etc., to provide high-quality, processable data for subsequent recurrent neural network model training, and improve the training effect and model performance. Step S232 designs the architecture of the recurrent neural network of the gated recurrent unit, including determining the number of recurrent layers, the number of neurons, etc., and designs a suitable recurrent neural network architecture, such as a gated recurrent unit (GRU), etc. according to the needs of the problem and the characteristics of the data; provides a recurrent neural network model architecture that can better capture data sequence information and time series relationships, and improves the expression and prediction capabilities of the recurrent neural network model. Step S233 divides the preprocessed data set into a training set and a verification set, uses the training set to train the recurrent neural network model, and uses the back propagation algorithm to update the weights and biases of the model, uses the training set to train the designed recurrent neural network model, and updates the parameters of the model through the back propagation algorithm, so that the recurrent neural network model can better fit the training data, improve the generalization ability of the recurrent neural network model, and enable it to accurately predict unseen data. Step S23 uses the validation set to evaluate the trained recurrent neural network model, calculates the performance indicators of the model on the validation set, such as the root mean square error (RMSE), mean absolute error (MAE), etc., adjusts the recurrent neural network based on the evaluation results, evaluates the performance of the recurrent neural network model on the validation set, understands the prediction accuracy and error degree of the model, adjusts the model based on the evaluation results, optimizes the performance of the model, and improves the prediction and generalization capabilities of the recurrent neural network model.

[0170] In summary, through the above steps, this embodiment establishes a recurrent neural network data model that can be trained according to historical vehicle data and road facility data, and uses the model to predict real-time vehicle data and road facility data, thereby obtaining real-time road condition information; it provides accurate and reliable road condition information for traffic management departments and vehicle users, helping them make better driving decisions and traffic management decisions. By continuously optimizing and adjusting the recurrent neural network model, the prediction accuracy and generalization ability of the recurrent neural network model can be improved, and the accuracy and practicality of road condition information can be improved. Through the above steps, the establishment of a recurrent neural network data model can be trained according to historical vehicle data and road facility data, and the model can be used to predict real-time vehicle data and road facility data, thereby obtaining real-time road condition information; it provides accurate and reliable road condition information for traffic management departments and vehicle users, helping them make better driving decisions and traffic management decisions.

[0171] Furthermore, the process of updating the weights and biases of the recurrent neural network model by the back propagation algorithm in step S233 specifically includes the following steps:

[0172] Step S2331: Using the current weight and bias parameters, input the input data in the training set into the recurrent neural network, and calculate the output value of each neuron layer by layer;

[0173] Step S2332: Calculate the loss function value of the recurrent neural network model based on the output of the recurrent neural network and the target features in the training set to measure the prediction error of the recurrent neural network model on the training set; Starting from the last time step of the recurrent neural network model, back-propagate the error signal based on the gradient of the loss function to calculate the error gradient of each time step;

[0174] Step S2333: According to hyperparameters such as error gradient and learning rate, use gradient descent or other optimization algorithms, use stochastic gradient descent to update the weights and bias parameters in the recurrent neural network, and gradually adjust the parameters by iterating multiple training samples to reduce the value of the loss function.

[0175] Step S2334: Repeat steps S2331 to S2334 until the recurrent neural network model converges or reaches a predetermined number of training rounds.

[0176] Preferably, step S2331 of this embodiment calculates the output value of each neuron through forward propagation, passes the input data through each layer of the recurrent neural network, calculates and transmits information layer by layer, obtains the final output, and uses the current weight and bias parameters to map the input data to the output space, providing a basis for subsequent error calculation and parameter update. Step S2332 calculates the loss function value of the recurrent neural network model, and calculates the error gradient. According to the output of the recurrent neural network model and the target features of the training set, the prediction error of the recurrent neural network model on the training set is measured, and the gradient is calculated for back propagation, the performance of the recurrent neural network model is evaluated, and the parameter update is guided by the loss function value and gradient, so that the recurrent neural network model is gradually optimized. Step S2333 uses the gradient descent method or other optimization algorithms to update the weights and bias parameters of the recurrent neural network model according to hyperparameters such as gradient and learning rate, and adjusts the parameters of the recurrent neural network model according to the error gradient and learning rate, so that the loss function value of the recurrent neural network model is gradually reduced; through parameter update, the recurrent neural network model is better fitted to the training set, and the prediction accuracy and generalization ability of the recurrent neural network model are improved. Step S2334 repeats steps S2331 to S2333 until the recurrent neural network model converges or reaches a predetermined number of training rounds, iteratively trains the recurrent neural network model, and continuously adjusts parameters until the recurrent neural network model converges or reaches a predetermined number of training rounds. Through iterative training, the recurrent neural network model gradually improves its performance and obtains better prediction results.

[0177] In summary, this embodiment uses the above steps to update the weights and biases of the recurrent neural network model through the back propagation algorithm, so that the recurrent neural network model can be gradually optimized, the prediction accuracy and generalization ability can be improved, the recurrent neural network model can better learn the characteristics and patterns of the data, and the prediction ability of the recurrent neural network model for vehicle data and road facility data can be improved, and accurate prediction results can be provided for traffic management and decision-making. Through the above steps, the back propagation algorithm can be used to update the weights and bias parameters in the recurrent neural network model according to the error between the prediction results of the recurrent neural network model and the actual target features, so that the prediction results of the recurrent neural network model are gradually optimized to improve the performance and accuracy of the recurrent neural network model, so that the recurrent neural network model can better learn the characteristics and patterns of the data, and improve the generalization and prediction ability of the recurrent neural network model.

[0178] Furthermore, if Figure 4 As shown, the process of verifying and adjusting the real-time traffic conditions in step S3 specifically includes the following steps:

[0179] Step S31: Collect user feedback on real-time traffic information through the vehicle user mobile application; the user provides feedback information such as actual road conditions, congestion conditions and travel time to verify the accuracy of the current road conditions; use the real-time data collected by the IoT sensor device to verify the real-time road conditions, and compare the actual observed data with the predicted road conditions to verify the accuracy of the prediction;

[0180] Step S32: combining the traffic flow data and camera image data provided by the traffic management department and other third-party data sources to compare and verify with the predicted real-time road conditions;

[0181] Step S33: Based on user feedback, sensor data, and verification results from third-party data sources, the prediction results are gradually adjusted by continuously iterating and updating the recurrent neural model using methods such as online learning.

[0182] Preferably, step S31 of this embodiment collects user feedback and real-time data through vehicle user mobile applications and IoT sensor devices to verify the accuracy of real-time road conditions, collect and integrate user feedback and real-time data for comparison and verification with predicted road conditions; through user feedback and actual observation data, verify whether the prediction results of the recurrent neural network model are consistent with the actual situation, and then judge the accuracy and reliability of the recurrent neural network model. Step S32 combines third-party data sources, such as traffic flow data and camera image data provided by the traffic management department, to compare and verify with the predicted real-time road conditions, and uses independent road condition information provided by the third-party data source to compare and verify with the prediction results of the recurrent neural network model. By comparing with reliable third-party data sources, the accuracy and credibility of the recurrent neural network model are further verified. Step S33 uses online learning and other methods to gradually adjust the prediction results by continuously iterating and updating the recurrent neural network model based on user feedback, sensor data and verification results of third-party data sources; dynamically adjusts the parameters and algorithms of the recurrent neural network model based on the verification results to make it more in line with actual road conditions; improves the accuracy and practicality of the recurrent neural network model through continuous updating and adjustment of the recurrent neural network model, and provides users with more reliable and accurate real-time road condition information.

[0183] In summary, through the above steps, this embodiment can verify and adjust the real-time road conditions, thereby improving the prediction accuracy and reliability of the recurrent neural network model, and can provide vehicle users with more accurate road condition information and suggestions, helping them make more reasonable driving decisions and improve the efficiency and safety of road travel. At the same time, through user feedback and verification of third-party data sources, the recurrent neural network model can be continuously improved and optimized to improve its performance and reliability in practical applications. Through the above method, real-time road conditions can be verified and adjusted, and the accuracy and reliability of the prediction results can be continuously optimized. More accurate and reliable real-time road condition information can be provided, providing vehicle users with more accurate road condition suggestions and navigation services, and improving the efficiency and safety of road travel.

[0184] like Figure 5 As shown, this embodiment also provides an embodiment of a highway real-time road condition detection system based on the Internet of Vehicles and the Internet of Things. In this embodiment, the highway real-time road condition detection system based on the Internet of Vehicles and the Internet of Things is applied to the highway real-time road condition detection method based on the Internet of Vehicles and the Internet of Things in the above embodiment, and the highway real-time road condition detection system based on the Internet of Vehicles and the Internet of Things includes a data acquisition module 1, a data processing module 2, and a result output module 3 that are electrically connected in sequence;

[0185] Among them, the data acquisition module 1 is used to obtain vehicle data and road facility data. The vehicle data is obtained through the on-board equipment installed on the vehicle; the road facility data is obtained by the Internet of Things sensor equipment arranged on the road through wireless communication; the collected vehicle data and road facility data are transmitted to the central server through wireless communication and stored; the data processing module 2 is used to remove abnormal data and correct the collected vehicle data and road facility data to obtain fused vehicle data and road facility data; based on the fused vehicle data and road facility data, the data analysis algorithm is used to calculate and predict real-time road condition information, such as traffic congestion level, vehicle flow, average driving speed, etc.; the result output module 3 is used to display the calculated and predicted real-time road condition information to vehicle users through mobile applications, providing real-time road condition information and suggestions; the real-time road condition is verified and adjusted through the data fed back by vehicle users.

[0186] Preferably, the data acquisition module 1 of this embodiment is used to acquire vehicle data and road facility data, collect data through vehicle-mounted equipment and Internet of Things sensor equipment, and transmit to the central server for storage through wireless communication; acquire vehicle and road data in real time, ensure the timeliness and integrity of data, provide real-time vehicle and road data as the basis for subsequent processing and analysis, and provide data support for the real-time road condition detection system. The data processing module 2 removes and corrects the abnormal data of the collected vehicle data and road facility data to obtain the fused vehicle data and road facility data, pre-processes the original data, removes abnormal data and corrects inaccurate data, and obtains accurate and reliable data; improves the quality and accuracy of the data, and provides an accurate data basis for subsequent real-time road condition calculation and prediction. The result output module 3 displays the calculated and predicted real-time road condition information to the vehicle user through a mobile application, providing real-time road condition information and suggestions; visualizes the calculated real-time road condition information so that users can intuitively understand the road conditions and make corresponding driving decisions, and provides accurate real-time road condition information and suggestions for vehicle users, improving the efficiency and safety of road travel.

[0187] In summary, this embodiment realizes data acquisition, processing and output through the synergy of the above modules, and builds a real-time road condition detection system based on the Internet of Vehicles and the Internet of Things; it realizes accurate detection and prediction of real-time road conditions, provides timely road condition information and suggestions for vehicle users, and improves the efficiency and safety of road travel. At the same time, through user feedback and verification, the system's models and algorithms are continuously optimized and adjusted to improve the accuracy and reliability of real-time road condition information.

[0188] Furthermore, the data acquisition module 1 specifically includes:

[0189] The function allocation submodule is used for the vehicle-mounted device to act as a publisher, register and publish data types and topics including the registered vehicle location and speed with the central server, and specify a corresponding topic for each data type; the IoT sensor device acts as a subscriber, registers and subscribes to data types and topics including traffic flow and meteorological data with the central server, and specifies a corresponding topic for each data type;

[0190] The data publishing submodule is used for the vehicle-mounted device to publish the collected vehicle data to the central server according to the topic specified during registration. The data type and topic information need to be attached to the data during publishing. The IoT sensor device subscribes to the data published on the central server according to the topic specified during registration, and specifies the data type and topic during subscription.

[0191] The data sending submodule is used for the central server to send data to the subscriber using the push method according to the subscriber's registration information after receiving the data published by the publisher, or for the subscriber to actively request data using the pull method.

[0192] Preferably, the vehicle-mounted device of the function allocation submodule of this embodiment acts as a publisher, registers and publishes vehicle location and speed data types and topics with the central server; the IoT sensor device acts as a subscriber, registers and subscribes to traffic flow and meteorological data data types and topics with the central server; realizes the registration and topic allocation of publishers and subscribers to ensure the correct publication and subscription of data; performs effective data distribution and subscription according to different data types and topics, and provides a basis for subsequent data transmission and processing. The vehicle-mounted device of the data publishing submodule publishes the collected vehicle data to the central server according to the topic specified during registration, and attaches the data type and topic information to the data; publishes the collected vehicle data to the central server, and attaches the data type and topic information to the data, to ensure the accurate transmission and identification of the data; transmits the data collected by the vehicle-mounted device to the central server, and provides an accurate data source for subsequent data processing and analysis. After receiving the data published by the publisher, the central server of the data sending submodule sends data to the subscriber using the push method according to the subscriber's registration information, or the subscriber actively requests data using the pull method; according to the subscriber's registration information, the data is pushed or pulled to ensure that the subscriber can obtain the required data in a timely manner; the timely transmission of data and real-time acquisition by subscribers are ensured, providing real-time data support for subsequent data processing and analysis.

[0193] In summary, through the synergy of the above submodules, the data acquisition module of this embodiment realizes the data collection, publishing and subscription of vehicle-mounted devices and IoT sensor devices, ensuring the timeliness and accuracy of data; a data acquisition mechanism is established to ensure that data can be acquired and transmitted in a timely manner, providing an accurate data basis for subsequent data processing and analysis. At the same time, according to the allocation of data types and topics, data classification and directional transmission are realized, improving data efficiency and availability.

[0194] Furthermore, the data processing module 2 specifically includes:

[0195] The data alignment submodule is used to clean and preprocess the collected vehicle data and road facility data, remove outliers and noise, and correct the data, match and associate them through timestamps, and align the corrected vehicle data and road facility data;

[0196] The data fusion submodule is used to fuse the vehicle data and the road facility data by weighted average after the vehicle data and the road facility data are aligned; and to extract target features related to vehicle speed, traffic volume, road congestion, etc. from the fused vehicle data and road facility data;

[0197] The model building submodule is used to build a recurrent neural network data model based on the vehicle data and road facility data extracted from the target features; use the historical vehicle data and road facility data to train the built recurrent neural network data model; use the trained recurrent neural network data model to predict the real-time vehicle data and road facility data to obtain real-time road condition information.

[0198] Preferably, the data alignment submodule of this embodiment cleans, preprocesses and corrects the collected vehicle data and road facility data, removes outliers and noise, and matches and associates through timestamps to achieve data alignment; cleans and corrects data to ensure data accuracy and consistency; improves data quality and accuracy, and provides an accurate data basis for subsequent data fusion and analysis. After aligning the vehicle data and road facility data, the data fusion submodule fuses the two by weighted average, and extracts target features related to vehicle speed, traffic volume, and road congestion from the fused data; fuses the vehicle data and road facility data, and extracts relevant target features; through fusion and feature extraction, more comprehensive and accurate data information is obtained, providing accurate input for subsequent model building and prediction. The model building submodule establishes a recurrent neural network (RNN) data model based on the extracted target features. Use historical vehicle data and road facility data to train the established RNN data model; use the trained RNN data model to predict real-time vehicle data and road facility data to obtain real-time road condition information; establish and train the RNN data model to predict real-time road conditions; by establishing and training the model, effective prediction of real-time road conditions can be achieved, providing real-time road condition information for subsequent result output.

[0199] In summary, through the synergy of the above submodules, the data processing module of this embodiment realizes the cleaning, alignment, fusion and modeling of the collected vehicle data and road facility data, and obtains accurate real-time road condition information; improves the quality and accuracy of the data, and provides an accurate data basis for subsequent road condition prediction and result output. At the same time, by establishing and training models, the prediction and analysis of real-time road conditions are realized, and accurate road condition information and suggestions are provided to vehicle users, thereby improving the efficiency and safety of road travel.

[0200] Furthermore, the result output module 3 specifically includes:

[0201] The information feedback submodule is used to collect user feedback on real-time traffic information through the vehicle user mobile application; users provide feedback information such as actual road conditions, congestion conditions and travel time to verify the accuracy of current road conditions; use real-time data collected by IoT sensor devices to verify real-time traffic conditions, and compare the actual observed data with the predicted traffic conditions to verify the accuracy of the prediction;

[0202] The comparison and verification submodule is used to compare and verify the predicted real-time road conditions with the traffic flow data and camera image data provided by the traffic management department and other third-party data sources;

[0203] The iterative update submodule is used to gradually adjust the prediction results by continuously iterating and updating the recurrent neural model based on user feedback, sensor data, and verification results from third-party data sources, using methods such as online learning.

[0204] Preferably, the information feedback submodule of this embodiment collects user feedback on real-time road condition information through a vehicle user mobile application. The user provides feedback information such as actual road conditions, congestion and driving time to verify the accuracy of the current road conditions; collect and integrate user feedback information for comparison and verification with the predicted road conditions, and verify whether the prediction results of the model are consistent with the actual situation through user feedback, thereby judging the accuracy and credibility of the model. The comparison and verification submodule combines third-party data sources such as traffic flow data and camera image data provided by the traffic management department to compare and verify with the predicted real-time road conditions, and uses independent road condition information provided by the third-party data source to compare and verify with the model prediction results; by comparing with reliable third-party data sources, the accuracy and reliability of the model are further verified. The iterative update submodule uses online learning and other methods based on user feedback, sensor data and the verification results of the third-party data source, and gradually adjusts the prediction results by continuously iterating and updating the recurrent neural model. According to the verification results, the parameters and algorithms of the model are dynamically adjusted to make it more consistent with the actual road conditions; through continuous model updating and adjustment, the accuracy and practicality of the model are improved to provide users with more reliable and accurate real-time road condition information.

[0205] In summary, through the synergy of the above submodules, the result output module of this embodiment realizes the verification and adjustment of real-time road conditions, improves the prediction accuracy and reliability of the model, and provides more accurate road condition information and suggestions for vehicle users, helping them make more reasonable driving decisions and improve the efficiency and safety of road travel. At the same time, through user feedback and verification of third-party data sources, the model can be continuously improved and optimized to improve its performance and reliability in practical applications.

[0206] like Figure 6As shown, this embodiment provides an embodiment of an electronic device. In this embodiment, the electronic device 4 includes a processor 41 and a memory 42 coupled to the processor 41.

[0207] The memory 42 stores program instructions for implementing the real-time highway road condition detection method based on the Internet of Vehicles and the Internet of Things according to any of the above embodiments.

[0208] The processor 41 is used to execute program instructions stored in the memory 42 to perform communication automation control.

[0209] The processor 41 may also be referred to as a CPU (Central Processing Unit). The processor 41 may be an integrated circuit chip having the ability to process signals. The processor 41 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0210] Further, Figure 7 The schematic diagram of the structure of the storage medium of an embodiment of the present application is shown in FIG. The storage medium 5 of the embodiment of the present application stores program instructions 51 capable of implementing all the above methods, wherein the program instructions 51 can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0211] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and 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.

[0212] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units. The above is only an implementation mode of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present invention.

[0213] The specific implementation methods of the invention are described in detail above, but they are only examples, and the invention is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the invention, therefore, the equalization, modification, improvement, etc. made without departing from the spirit and principle of the invention should be included in the scope of the invention.

Claims

1. A method for real-time highway road condition detection based on Internet of Vehicles and Internet of Things, characterized in that: The method for detecting real-time road conditions based on the Internet of Vehicles and the Internet of Things includes: Obtain vehicle data and road facility data. Vehicle data is obtained through on-board equipment installed on the vehicle; road facility data is obtained through wireless communication using IoT sensor devices arranged on the highway; the collected vehicle data and road facility data are transmitted to the central server through wireless communication and stored; The collected vehicle data and road facility data are subjected to abnormal data removal and correction to obtain fused vehicle data and road facility data; based on the fused vehicle data and road facility data, real-time road condition information is calculated and predicted using data analysis algorithms; The calculated and predicted real-time traffic information is displayed to vehicle users through mobile applications, providing real-time traffic information and suggestions; the real-time traffic conditions are verified and adjusted through the data fed back by vehicle users; The process of using data analysis algorithms to calculate and predict real-time traffic information includes: Clean and preprocess the collected vehicle data and road facility data, remove outliers and noise, correct the data, match and associate them through timestamps, and align the corrected vehicle data and road facility data; After the vehicle data and the road facility data are aligned, the vehicle data and the road facility data are fused by weighted averaging; target features related to vehicle speed, traffic volume and road congestion are extracted from the fused vehicle data and road facility data; A recurrent neural network data model is established based on the vehicle data and road facility data extracted from the target features; the established recurrent neural network data model is trained using the historical vehicle data and road facility data; the real-time vehicle data and road facility data are predicted using the trained recurrent neural network data model to obtain real-time road condition information; The interaction process between the vehicle-mounted device and the IoT sensor device includes: As a publisher, the vehicle-mounted device registers and publishes data types and topics including the registered vehicle location and speed with the central server, and specifies a corresponding topic for each data type; as a subscriber, the IoT sensor device registers and subscribes to data types and topics including traffic flow and meteorological data with the central server, and specifies a corresponding topic for each data type; The vehicle-mounted device publishes the collected vehicle data to the central server according to the topic specified during registration; the data type and topic information need to be attached to the data when publishing; the IoT sensor device subscribes to the data published on the central server according to the topic specified during registration, and specifies the data type and topic when subscribing; After receiving the data published by the publisher, the central server sends the data to the subscriber using the push method according to the subscriber's registration information, or the subscriber actively requests the data using the pull method; The process of aligning the corrected vehicle data and road facility data includes: Sort the timestamps in the vehicle data and road facility data; define a target timestamp, that is, to find the closest timestamp, initialize two pointers, one pointing to the starting position of the sorted vehicle data, and the other pointing to the starting position of the sorted road facility data; enter a loop traversal until all data are traversed or the closest timestamp is found; In each cycle, the difference between the timestamp of the current vehicle data and road facility data and the target timestamp is compared, and the minimum difference is recorded; based on the recorded minimum difference, the data corresponding to the closest timestamp is determined, namely, the vehicle data and road facility data; Establish a data association table to associate the matched vehicle data with the road facility data to ensure that they correspond at the same time point; If the timestamp of the current vehicle data is less than the target timestamp, move the pointer of the vehicle data back one position and continue to compare the next vehicle data; if the timestamp of the current road facility data is less than the target timestamp, move the pointer of the road facility data back one position and continue to compare the next road facility data; if the timestamp of the current vehicle data is greater than the target timestamp, or the timestamp of the current road facility data is greater than the target timestamp, it means that the closest timestamp has been found and the loop ends; The process of building a recurrent neural network data model includes: Prepare a data set for establishing training data and validation data for a recurrent neural network, the data set including historical vehicle data, road facility data, and corresponding target features; preprocess the prepared data; Design the architecture of a recurrent neural network with a gated recurrent unit. The recurrent neural network consists of multiple recurrent layers, each with neurons. The preprocessed data set is divided into a training set and a validation set. The training set is used to train the recurrent neural network model, and the validation set is used to evaluate the performance of the recurrent neural network model and adjust hyperparameters. The designed recurrent neural network model is trained using the training set. During the training process, the weights and biases of the recurrent neural network model are updated through the back-propagation algorithm. Use the validation set to evaluate the trained recurrent neural network model, calculate the performance indicators of the recurrent neural network model on the validation set, and adjust the recurrent neural network based on the evaluation results.

2. The method for real-time highway road condition detection based on Internet of Vehicles and Internet of Things according to claim 1 is characterized in that: The process of sending data to subscribers using push includes: After receiving the data published by the publisher, the central server determines which subscribers need to be pushed based on the data type and topic; The central server packages the data and adds identification information of the data type and subject; The central server generates a symmetric encryption key through the established communication connection, and the key is used to encrypt the data packet; the central server uses the symmetric encryption key to encrypt the data packet to generate an encrypted data packet; the central server obtains the subscriber's public key, and the public key is used to decrypt the symmetric encryption key; the central server uses the subscriber's public key to asymmetrically encrypt the symmetric encryption key to generate an encrypted symmetric encryption key; the central server sends the encrypted symmetric encryption key and the encrypted data packet to the subscriber, and sends the data packet to the subscriber; The subscriber receives the encrypted data packet and the encrypted symmetric encryption key. The subscriber uses his own private key to asymmetrically decrypt the encrypted symmetric encryption key to obtain the symmetric encryption key. The subscriber uses the obtained symmetric encryption key to symmetrically decrypt the encrypted data packet, restore the original data, and parse and process it according to the data type and topic.

3. The method for real-time highway road condition detection based on Internet of Vehicles and Internet of Things according to claim 1 is characterized in that: The process of establishing a data association table includes: Create a data association table of a hash table to store the association information of vehicle data and road facility data; traverse the matched vehicle data and road facility data and correspond them according to the timestamp; For each matched vehicle data, use its timestamp as the keyword and store the vehicle data in the association table; For each matched road facility data, its timestamp is used as the keyword; Find the corresponding vehicle data in the association table and associate the road facility data with the vehicle data.

4. The method for real-time highway road condition detection based on Internet of Vehicles and Internet of Things according to claim 1 is characterized in that: The process of fusing vehicle data and road facility data using weighted average includes: Determine the target features of the vehicle data and road facility data that need to be fused, and normalize the vehicle data and road facility data; The correlation index between vehicle data and road facility data is calculated based on the Pearson correlation coefficient. The value range of the correlation coefficient is -1 to 1. The closer the value is to 1, the stronger the positive correlation is. The closer the value is to -1, the stronger the negative correlation is. The closer the value is to 0, the no correlation is. According to the degree of correlation, weights are set for vehicle data and road facility data. For each target feature, the vehicle data and road facility data are weighted averaged according to the weight, that is, each data is multiplied by the corresponding weight, and then the sum is taken and divided by the sum of the weights; based on the result of the weighted average, the fused data is obtained, that is, the comprehensive information obtained by fusing the vehicle data and the road facility data.

5. A real-time highway traffic condition detection system based on the Internet of Vehicles and the Internet of Things, which is applied to the real-time highway traffic condition detection method based on the Internet of Vehicles and the Internet of Things as claimed in any one of claims 1 to 4, characterized in that: The highway real-time road condition detection system based on the Internet of Vehicles and the Internet of Things includes: The data acquisition module is used to acquire vehicle data and road facility data. The vehicle data is acquired through the on-board equipment carried by the vehicle; the road facility data is acquired through the IoT sensor equipment arranged on the road through wireless communication; the collected vehicle data and road facility data are transmitted to the central server through wireless communication and stored; The data processing module is used to remove abnormal data and correct the collected vehicle data and road facility data to obtain fused vehicle data and road facility data; based on the fused vehicle data and road facility data, the data analysis algorithm is used to calculate and predict real-time road condition information; The result output module is used to display the calculated and predicted real-time traffic information to vehicle users through mobile applications, providing real-time traffic information and suggestions; and verifying and adjusting the real-time traffic conditions through data feedback from vehicle users.

Citation Information

Patent Citations

  • Traffic flow collection, prediction, management and control and informatization pushing system and method based on intelligent high speed

    CN116206440A

  • Highway road condition early warning system based on road network big data

    CN116740974A

  • Vehicle lane changing guiding method, device and equipment and storage medium

    CN117456757A

  • Correlation analyzing method based on vehicle, road conditions and safe travel big data in network of vehicles

    CN105844913A

  • Multi-vehicle collaborative planning method based on distributed crowd-sourcing learning

    CN114283607A