Vehicle monitoring method and system

Abstract

The present invention relates to the technical field of vehicle monitoring methods, and specifically to a vehicle monitoring method and system, comprising the following steps: using a vehicle monitoring system to collect real-time vehicle data including speed, acceleration, and steering angle. In the present invention, by comprehensively utilizing a vehicle monitoring system, a machine learning algorithm, computer vision technology, cloud computing and big data analysis technology, a geographic information system, and a wireless sensor network, all-round monitoring of the real-time driving status of a vehicle, identification and verification of abnormal driving behavior, prediction and analysis of traffic flow, planning of intelligent navigation routes, and sharing of information between vehicles and cooperative driving functions are realized. By collecting vehicle data in real time and identifying abnormal driving behavior, timely discovering and warning of traffic risks, traffic flow prediction and intelligent route planning help drivers avoid congestion points and potential accident risk points, saving driving time, and sharing of information between vehicles and cooperative driving improve driving efficiency and safety.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle monitoring methods, and in particular to a vehicle monitoring method and system. Background Art

[0002] The vehicle monitoring method focuses on using various sensors and communication technologies to track, monitor and manage vehicles. It not only integrates communication technologies such as GPS and GPRS, but also introduces advanced technologies such as GIS geographic information system and wireless sensors. The vehicle monitoring method uses devices such as dashcams and GPS to collect information such as latitude and longitude, running direction, speed and time of vehicles in real time around the world, and transmits it to the back-end server through the GPRS network, and then analyzes and stores the data through management software. GIS geographic information system plays a key role in vehicle monitoring. Through GIS, we can accurately understand the location and movement trajectory of vehicles in geographic space, and achieve more efficient and safe route planning. This spatial analysis can also help managers prevent and respond to various risks at specific times and places. Wireless sensors are used to capture various operating parameters of vehicles, such as fuel volume, tire pressure, engine status, etc. These sensors can send data to the server in real time, which not only reduces data delays and errors, but also increases the multi-dimensional analysis capabilities of the vehicle monitoring system. The main purpose of the vehicle monitoring method is to achieve real-time monitoring and management of vehicles to improve operational efficiency and ensure driving safety. Through full trajectory playback and data analysis, managers can better understand the operating status of the vehicle and prevent and respond to various risks.

[0003] Among the existing vehicle monitoring methods, the existing vehicle monitoring methods mainly rely on traditional on-board instruments or GPS positioning, and their monitoring data may have delays or errors, which is not conducive to accurate analysis and prediction of real-time situations. Existing methods usually only provide basic driving status monitoring and route navigation functions, lacking advanced functions such as abnormal driving behavior recognition, road condition prediction analysis, and vehicle-to-vehicle cooperative driving. Existing methods usually do not make full use of vehicle data, lack in-depth data analysis and machine learning models, and are insufficient to achieve a comprehensive understanding and prediction of driving behavior and road conditions. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a vehicle monitoring method and system.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: a vehicle monitoring method, comprising the following steps:

[0006] S1: Use the vehicle monitoring system to collect real-time vehicle data including speed, acceleration, and steering angle, and transmit it to the server via GPRS for storage and processing to generate a basic vehicle data set;

[0007] S2: Based on the basic vehicle data set, a machine learning algorithm including a support vector machine or a decision tree is used to analyze the driving state of the vehicle, identify abnormal driving behavior, and generate an abnormal driving report;

[0008] S3: Combining the abnormal driving report and computer vision technology, the vehicle images captured by the surveillance camera are processed in real time through target detection and target tracking algorithms to confirm and verify the abnormal driving behavior and generate vehicle behavior verification results;

[0009] S4: using cloud computing and big data analysis technology to conduct in-depth analysis on the basic vehicle data set, combining the vehicle behavior verification results, predicting traffic flow and congestion points, and generating a traffic flow prediction report;

[0010] S5: Based on the traffic flow forecast report and the geographic information system, provide intelligent route planning for the driver to avoid potential traffic congestion and accident risk points, and generate intelligent navigation suggestions;

[0011] S6: Combining the intelligent navigation suggestion and the wireless sensor network, information sharing and cooperative driving between vehicles are realized, and the vehicle adjusts itself according to the driving status and road condition information of surrounding vehicles to generate a cooperative driving strategy.

[0012] As a further solution of the present invention, the real-time vehicle data including speed, acceleration and steering angle are collected by using a vehicle monitoring system and transmitted to a server via GPRS for storage and processing. The steps of generating a basic vehicle data set are specifically as follows:

[0013] S101: Communicate with the vehicle electronic control unit using the OBD-II interface, obtain speed, acceleration, and steering angle parameters through a serial communication protocol, and generate original vehicle data;

[0014] S102: preprocessing the original vehicle data using a data cleaning algorithm including missing value filling and outlier removal to generate preprocessed vehicle data;

[0015] S103: using the GPRS module and the transmission control protocol TCP, sending the pre-processed data to the server to generate a transmission record;

[0016] S104: The server uses a relational database management system to store the transmission records, ensures data integrity through a data quality detection algorithm, and generates a basic vehicle data set.

[0017] As a further solution of the present invention, based on the basic vehicle data set, a machine learning algorithm including a support vector machine or a decision tree is used to analyze the driving state of the vehicle, identify abnormal driving behavior, and generate an abnormal driving report in the following specific steps:

[0018] S201: extracting feature vectors from the basic vehicle data set using a principal component analysis method to generate a feature data set;

[0019] S202: using a support vector machine algorithm with a radial basis function kernel to classify abnormal driving behaviors on the feature data set, and generating an abnormal driving candidate list;

[0020] S203: Using a decision tree algorithm with information gain as a feature selection method, screening the abnormal driving candidate list to generate a screened abnormal driving list;

[0021] S204: Based on the screened abnormal driving list, a template matching algorithm is applied to generate an abnormal driving report.

[0022] As a further solution of the present invention, in combination with the abnormal driving report and computer vision technology, the vehicle images captured by the surveillance camera are processed in real time through target detection and target tracking algorithms to confirm and verify abnormal driving behavior. The specific steps of generating vehicle behavior verification results are as follows:

[0023] S301: Analyze the image captured by the surveillance camera using a target detection algorithm to generate a target detection result;

[0024] S302: Based on the target detection result, a simple online real-time tracking algorithm is used to track the target, and a target tracking result is generated based on the time series data;

[0025] S303: using a data fusion algorithm, specifically a Kalman filter, to fuse the target tracking result and the abnormal driving report to generate fused data;

[0026] S304: Based on the fused data and the preset threshold, a Bayesian reasoning model is applied to confirm and verify the abnormal driving behavior, and a vehicle behavior verification result is generated.

[0027] As a further solution of the present invention, cloud computing and big data analysis technology are used to conduct in-depth analysis on the basic vehicle data set, and combined with the vehicle behavior verification results, traffic flow and congestion points are predicted, and the steps of generating a traffic flow prediction report are specifically as follows:

[0028] S401: Uploading the basic vehicle data set and the vehicle behavior verification result to a cloud server using the Hypertext Transfer Protocol secure protocol to generate a cloud data set;

[0029] S402: Based on the cloud data set, a big data processing framework is used to perform data analysis and feature extraction to generate a traffic analysis report;

[0030] S403: using an autoregressive integrated moving average model to perform time series analysis on the traffic analysis report, predict traffic flow, and generate traffic flow prediction data;

[0031] S404: Integrate the traffic flow prediction data, apply data visualization technology, and use a heat map to generate a traffic flow prediction report.

[0032] As a further solution of the present invention, based on the traffic flow forecast report and the geographic information system, intelligent route planning is provided for the driver to avoid potential traffic congestion and accident risk points. The steps of generating intelligent navigation suggestions are specifically as follows:

[0033] S501: Based on the traffic flow prediction report, using the shortest path algorithm and geographic information system, calculate the optimal route and generate optimal route data;

[0034] S502: Applying the risk assessment model to perform risk assessment on the optimal route through a Markov decision process, and generating a risk assessment report;

[0035] S503: combining the risk assessment report and the optimal route data, using a genetic algorithm to optimize the route and generate an optimized optimal route;

[0036] S504: Based on the generated optimized optimal route, the optimized optimal route is pushed to the vehicle information system in real time by using data wrapper technology and message queue telemetry transmission protocol to generate real-time navigation data and intelligent navigation suggestions.

[0037] As a further solution of the present invention, the intelligent navigation suggestion and the wireless sensor network are combined to realize information sharing and cooperative driving between vehicles. The vehicle adjusts itself according to the driving status and road condition information of the surrounding vehicles. The steps of generating a cooperative driving strategy are specifically as follows:

[0038] S601: Use event-driven architecture and complex event processing algorithms to monitor vehicle behavior, traffic flow, and road conditions in real time and generate real-time monitoring data;

[0039] S602: Based on the real-time monitoring data, natural language processing technology and speech recognition technology are used to generate voice reminders and warnings;

[0040] S603: Using an embedded system and microservice architecture, the real-time monitoring data is connected to an emergency response center to generate a coordinated driving strategy.

[0041] A vehicle monitoring system is used to execute the above vehicle monitoring method, and the vehicle monitoring system is composed of a data collection module, a driving status analysis module, an abnormal behavior verification module, a traffic prediction module, and an intelligent navigation module.

[0042] As a further solution of the present invention, the data collection module communicates with the electronic control unit of the vehicle through the OBD-II interface, processes the collected speed, acceleration, and steering angle data using a data cleaning and preprocessing algorithm, and generates preprocessed vehicle data;

[0043] The driving state analysis module analyzes the driving state of the vehicle based on the preprocessed vehicle data using a support vector machine or a decision tree algorithm to generate an abnormal driving report;

[0044] The abnormal behavior verification module combines the abnormal driving report with the target detection and target tracking algorithm of computer vision to process the vehicle images captured by the surveillance camera in real time and generate the vehicle behavior verification result;

[0045] The traffic prediction module predicts traffic flow and congestion points based on the vehicle behavior verification results and pre-processed vehicle data using cloud computing and big data analysis technology to generate a traffic flow prediction report;

[0046] The intelligent navigation module generates intelligent navigation suggestions and collaborative driving strategies based on traffic flow forecast reports in combination with geographic information systems and wireless sensor networks.

[0047] As a further solution of the present invention, the data collection module includes a data interface submodule, a data cleaning submodule, and a data preprocessing submodule;

[0048] The driving state analysis module includes a data analysis submodule, a machine learning submodule, and an abnormality recognition submodule;

[0049] The abnormal behavior verification module includes an image processing submodule, a target detection submodule, and a target tracking submodule;

[0050] The traffic prediction module includes a big data analysis submodule, a traffic flow prediction submodule, and a congestion prediction submodule;

[0051] The intelligent navigation module includes a route planning submodule, a traffic information updating submodule, and a coordinated driving submodule.

[0052] Compared with the prior art, the advantages and positive effects of the present invention are:

[0053] In the present invention, by comprehensively utilizing vehicle monitoring systems, machine learning algorithms, computer vision technology, cloud computing and big data analysis technology, geographic information systems, and wireless sensor networks, all-round monitoring of the real-time driving status of vehicles, identification and verification of abnormal driving behaviors, prediction and analysis of traffic flow, planning of intelligent navigation routes, and sharing of information between vehicles and cooperative driving functions are realized. By collecting vehicle data in real time and identifying abnormal driving behaviors, traffic risks can be discovered and warned in a timely manner, reducing the incidence of traffic accidents. Traffic flow prediction and intelligent route planning help drivers avoid congestion points and potential accident risk points, save driving time, and improve driving experience. The sharing of information between vehicles and cooperative driving improves driving efficiency and safety, and helps to achieve the goals of intelligent transportation systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0055] Figure 2 This is a detailed flow chart of S1 of the present invention;

[0056] Figure 3 This is a detailed flow chart of S2 of the present invention;

[0057] Figure 4 This is a detailed flow chart of S3 of the present invention;

[0058] Figure 5 This is a detailed flow chart of S4 of the present invention;

[0059] Figure 6 This is a detailed flow chart of S5 of the present invention;

[0060] Figure 7 This is a detailed flow chart of S6 of the present invention;

[0061] Figure 8 is a system flow chart of the present invention;

[0062] Fig. 9 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0064] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0065] Embodiment 1

[0066] See also Figure 1 The present invention provides a technical solution: a vehicle monitoring method, comprising the following steps:

[0067] S1: Use the vehicle monitoring system to collect real-time vehicle data including speed, acceleration, and steering angle, and transmit it to the server via GPRS for storage and processing to generate a basic vehicle data set;

[0068] S2: Based on the basic vehicle data set, a machine learning algorithm including a support vector machine or a decision tree is used to analyze the vehicle's driving status, identify abnormal driving behavior, and generate an abnormal driving report;

[0069] S3: Combines abnormal driving reports with computer vision technology, processes vehicle images captured by surveillance cameras in real time through target detection and target tracking algorithms, confirms and verifies abnormal driving behaviors, and generates vehicle behavior verification results;

[0070] S4: Use cloud computing and big data analysis technology to conduct in-depth analysis of basic vehicle data sets, combine vehicle behavior verification results, predict traffic flow and congestion points, and generate traffic flow prediction reports;

[0071] S5: Based on traffic flow forecast reports and geographic information systems, it provides drivers with intelligent route planning, avoids potential traffic jams and accident risk points, and generates intelligent navigation suggestions;

[0072] S6: Combine intelligent navigation suggestions and wireless sensor networks to achieve information sharing and collaborative driving between vehicles. Vehicles adjust themselves according to the driving status and road conditions of surrounding vehicles to generate collaborative driving strategies.

[0073] First, by collecting, storing and processing real-time vehicle data, a basic vehicle dataset can be constructed, providing a basis for further analysis and monitoring.

[0074] Secondly, by using machine learning algorithms such as support vector machines or decision trees to analyze the driving status of vehicles, abnormal driving behaviors can be identified and abnormal driving reports can be generated, which helps improve the safety and compliance of driving behaviors and reduce the occurrence of traffic accidents.

[0075] By combining abnormal driving reports with computer vision technology, real-time processing of vehicle images captured by surveillance cameras can confirm and verify abnormal driving behavior. This verification method can more accurately judge the driver's behavior and generate vehicle behavior verification results.

[0076] By using cloud computing and big data analysis technology, in-depth analysis of basic vehicle data sets and combining vehicle behavior verification results, it is possible to predict traffic flow and congestion points. The traffic flow forecast report generated in this way can provide a reference for traffic management and planning to optimize traffic flow and reduce congestion.

[0077] Based on traffic flow forecast reports and geographic information systems, intelligent route planning is provided to drivers, which can help them avoid potential traffic jams and accident risk points. Such intelligent navigation suggestions can improve driving efficiency and safety.

[0078] Finally, by combining intelligent navigation suggestions and wireless sensor networks, information sharing and cooperative driving between vehicles can be realized. Vehicles can adjust themselves according to the driving status and road conditions of surrounding vehicles and generate cooperative driving strategies. This kind of cooperative driving can improve traffic efficiency and reduce the occurrence of traffic accidents.

[0079] See also Figure 2 ,Use the vehicle monitoring system to collect real-time vehicle data including speed, acceleration, and steering angle, and transmit it to the server via GPRS for storage and processing. The steps to generate the basic vehicle data set are as follows:

[0080] S101: Communicate with the vehicle electronic control unit using the OBD-II interface, obtain speed, acceleration, and steering angle parameters through a serial communication protocol, and generate original vehicle data;

[0081] S102: preprocessing the original vehicle data using a data cleaning algorithm including missing value filling and outlier removal to generate preprocessed vehicle data;

[0082] S103: using the GPRS module and the transmission control protocol TCP, the pre-processed data is sent to the server to generate a transmission record;

[0083] S104: The server uses a relational database management system to store transmission records, ensures data integrity through a data quality detection algorithm, and generates a basic vehicle data set.

[0084] First, in terms of real-time vehicle data collection, it can provide accurate parameters such as speed, acceleration and steering angle, providing reliable basic data for subsequent analysis and decision-making. Second, in terms of data preprocessing and transmission, through cleaning algorithms and GPRS transmission, it can ensure the accuracy and timeliness of data, providing support for precise analysis and decision-making. Third, in terms of data storage and management, this method can ensure the integrity and reliability of data through relational database management systems and data quality detection algorithms, providing a solid foundation for subsequent data analysis and application.

[0085] See also Figure 3 Based on the basic vehicle data set, a machine learning algorithm including a support vector machine or a decision tree is used to analyze the vehicle's driving status, identify abnormal driving behavior, and generate an abnormal driving report. The specific steps are:

[0086] S201: extracting feature vectors from the basic vehicle data set using a principal component analysis method to generate a feature data set;

[0087] S202: using a support vector machine algorithm with a radial basis function kernel, classifying abnormal driving behaviors on the feature data set, and generating an abnormal driving candidate list;

[0088] S203: Using a decision tree algorithm with information gain as a feature selection method, screening the abnormal driving candidate list to generate a screened abnormal driving list;

[0089] S204: Based on the screened abnormal driving list, a template matching algorithm is applied to generate an abnormal driving report.

[0090] First, in step S201, the key features are extracted from the basic vehicle data set through principal component analysis, which can reduce the dimension of the data and retain the most representative information, providing efficient input for abnormal driving behavior analysis. Secondly, in steps S202 and S203, machine learning algorithms such as support vector machines and decision trees are used to classify and filter the feature data sets. These algorithms can accurately identify abnormal driving behaviors and extract the most relevant features through information gain and feature selection methods to generate a filtered abnormal driving list. Finally, in step S204, a template matching algorithm is applied to generate an abnormal driving report. The template matching algorithm can match the filtered abnormal driving behavior with the predefined abnormal driving template, further determine the specific circumstances of the abnormal driving behavior, and generate a corresponding report.

[0091] See also Figure 4, combined with abnormal driving reports and computer vision technology, through target detection and target tracking algorithms, real-time processing of vehicle images captured by surveillance cameras, confirming and verifying abnormal driving behaviors, and generating vehicle behavior verification results in the following steps:

[0092] S301: Analyze the image captured by the surveillance camera using a target detection algorithm to generate a target detection result;

[0093] S302: Based on the target detection result, a simple online real-time tracking algorithm is used to track the target, and a target tracking result is generated based on the time series data;

[0094] S303: using a data fusion algorithm, specifically a Kalman filter, to fuse the target tracking result and the abnormal driving report to generate fused data;

[0095] S304: Based on the fused data and the preset threshold, a Bayesian reasoning model is applied to confirm and verify the abnormal driving behavior and generate a vehicle behavior verification result.

[0096] First, in step S301, the target detection algorithm is used to analyze the vehicle image, which can accurately identify and locate the vehicle and provide key target detection results. This provides basic data for the subsequent identification and verification of abnormal driving behavior. Secondly, in step S302, the target tracking algorithm is used to track the vehicle in real time. By associating the vehicle targets between different frames, a target tracking result with a time series can be generated. Such target tracking can provide the vehicle's motion trajectory and state information, and a more comprehensive understanding of vehicle behavior. In step S303, the target tracking results and abnormal driving reports are fused through the data fusion algorithm. Data fusion can integrate information from different sources and provide more accurate and stable vehicle behavior data. This fusion result can provide a more reliable basis for the subsequent confirmation and verification of abnormal driving behavior. Finally, in step S304, the Bayesian reasoning model, fused data and preset thresholds can be used to confirm and verify abnormal driving behavior. Based on the comparative analysis of the fused data, it can be determined whether the vehicle has abnormal behavior and generate reliable vehicle behavior verification results.

[0097] See also Figure 5 , using cloud computing and big data analysis technology, in-depth analysis of the basic vehicle data set, combined with vehicle behavior verification results, predicting traffic flow and congestion points, and generating traffic flow prediction reports in the following steps:

[0098] S401: Upload the basic vehicle data set and the vehicle behavior verification result to the cloud server using the Hypertext Transfer Protocol secure protocol to generate a cloud data set;

[0099] S402: Based on the cloud data set, use the big data processing framework to perform data analysis and feature extraction to generate a traffic analysis report;

[0100] S403: using an autoregressive integrated moving average model to perform time series analysis on the traffic analysis report, predict traffic flow, and generate traffic flow prediction data;

[0101] S404: Integrate traffic flow prediction data, apply data visualization technology, and use heat maps to generate traffic flow prediction reports.

[0102] First, in step S401, the data is uploaded to the cloud server, and the powerful computing power and storage resources of cloud computing are used for efficient processing, which greatly accelerates the speed and accuracy of data analysis. Secondly, in step S402, the big data processing framework is applied for data analysis and feature extraction. Using the big data processing framework, large-scale data sets can be processed and useful feature information can be extracted from them, providing reliable basic data for subsequent traffic analysis. Such data analysis can deeply explore key indicators such as driving behavior and road conditions, providing important reference and decision support for traffic management. In step S403, time series analysis is performed through the autoregressive integrated moving average model to accurately predict traffic flow. Predicting traffic flow can detect congestion points and peak hours in advance, help transportation departments and drivers make corresponding adjustments and decisions, optimize traffic flow, and alleviate congestion. Finally, in step S404, data visualization technology such as heat maps are used to intuitively display the traffic flow prediction results. Through the visual report, users can clearly understand the traffic conditions and congestion, helping them make more informed travel decisions. This visual presentation makes traffic information easier to understand and use, and improves the efficiency and accuracy of traffic management.

[0103] See also Figure 6 Based on traffic flow forecast reports and geographic information systems, intelligent route planning is provided for drivers to avoid potential traffic congestion and accident risk points. The specific steps for generating intelligent navigation suggestions are as follows:

[0104] S501: Based on the traffic flow forecast report, the shortest path algorithm and the geographic information system are used to calculate the optimal route and generate optimal route data;

[0105] S502: Apply the risk assessment model, conduct risk assessment on the optimal route through the Markov decision process, and generate a risk assessment report;

[0106] S503: combining the risk assessment report and the optimal route data, using a genetic algorithm to optimize the route, and generating an optimized optimal route;

[0107] S504: Based on the generated optimized optimal route, the optimized optimal route is pushed to the vehicle information system in real time by using data wrapper technology and message queue telemetry transmission protocol to generate real-time navigation data and intelligent navigation suggestions.

[0108] First, in step S501, by calculating the optimal route, the driver can avoid congested areas, reduce traffic delays, save time and fuel consumption, and provide an efficient travel experience. Secondly, in step S502, the optimal route is evaluated through a risk assessment model, and the driver can obtain information about road safety. This helps to reduce potential accident risks and enhance the driver's safety awareness and decision-making ability, thereby improving road safety. In step S503, the route is optimized through a genetic algorithm, and the driver can choose a better path based on actual road conditions and traffic flow. This optimization can reduce driving time and driving distance, and can adapt to real-time changing traffic conditions during navigation, helping the driver to reach the destination more efficiently. Finally, in step S504, the optimized optimal route and intelligent navigation suggestions are pushed to the vehicle information system in real time, and the driver can receive navigation guidance and traffic information in real time. This provides real-time navigation functions, supports drivers to make more informed driving decisions, reduces driving errors, and provides a more holistic travel experience.

[0109] See also Figure 7 , combined with intelligent navigation suggestions and wireless sensor networks, information sharing and cooperative driving between vehicles are realized. Vehicles adjust themselves according to the driving status and road conditions of surrounding vehicles. The specific steps for generating cooperative driving strategies are:

[0110] S601: Use event-driven architecture and complex event processing algorithms to monitor vehicle behavior, traffic flow, and road conditions in real time and generate real-time monitoring data;

[0111] S602: Generate voice reminders and warnings based on real-time monitoring data using natural language processing technology and speech recognition technology;

[0112] S603: Use embedded systems and microservices architecture to connect real-time monitoring data with the emergency response center to generate collaborative driving strategies.

[0113] First, in step S601, accurate real-time monitoring data is generated by real-time monitoring of vehicle behavior, traffic flow and road conditions. This helps drivers understand the current traffic conditions, predict congestion and dangerous situations in advance, and then take corresponding actions to improve road safety and driving efficiency. Secondly, in step S602, natural language processing and speech recognition technology are used to convey important information to drivers through voice reminders and warnings. This allows drivers to focus more on road driving, while providing real-time road condition updates and warnings, reducing accident risks and enhancing driving safety. In step S603, through embedded systems and microservices architecture, real-time monitoring data is connected to the emergency response center to achieve collaborative driving strategies. This enables vehicles to make intelligent decisions based on real-time data, such as adjusting driving speed, avoiding conflicts, and improving road traffic efficiency and vehicle mobility.

[0114] See also Figure 8 A vehicle monitoring system is used to execute the above-mentioned vehicle monitoring method. The vehicle monitoring system is composed of a data collection module, a driving status analysis module, an abnormal behavior verification module, a traffic prediction module, and an intelligent navigation module.

[0115] The data collection module communicates with the vehicle's electronic control unit through the OBD-II interface, processes the collected speed, acceleration, and steering angle data using data cleaning and preprocessing algorithms, and generates preprocessed vehicle data;

[0116] The driving status analysis module uses support vector machine or decision tree algorithm to analyze the driving status of the vehicle based on the preprocessed vehicle data and generates an abnormal driving report;

[0117] The abnormal behavior verification module combines abnormal driving reports with computer vision target detection and target tracking algorithms to process vehicle images captured by surveillance cameras in real time and generate vehicle behavior verification results;

[0118] The traffic prediction module uses cloud computing and big data analysis technology to predict traffic flow and congestion points based on vehicle behavior verification results and pre-processed vehicle data, and generates a traffic flow prediction report;

[0119] The intelligent navigation module generates intelligent navigation suggestions and collaborative driving strategies based on traffic flow forecast reports, combined with geographic information systems and wireless sensor networks.

[0120] First, the data collection module collects vehicle data by communicating with the vehicle's electronic control unit, and processes data such as speed, acceleration, and steering angle through data cleaning and preprocessing algorithms. This data preprocessing helps to extract and accurately record information related to vehicle driving. Secondly, the driving state analysis module uses the preprocessed vehicle data to analyze the vehicle's driving state using a support vector machine or decision tree algorithm. This module can detect abnormal driving behavior and generate abnormal driving reports, thereby improving the driver's safety awareness and the standardization of driving behavior. The abnormal behavior verification module combines the abnormal driving report and uses computer vision target detection and target tracking algorithms to process vehicle images in real time. This can verify whether the driver's behavior is consistent with the abnormal report, increasing the reliability and accuracy of the monitoring system. The traffic prediction module uses cloud computing and big data analysis technology to predict traffic flow and congestion points based on the vehicle behavior verification results and preprocessed vehicle data. This module can provide traffic flow prediction reports to help drivers avoid congested areas, choose better driving routes, improve driving efficiency and reduce travel time. Finally, the intelligent navigation module generates intelligent navigation suggestions and collaborative driving strategies based on traffic flow prediction reports, geographic information systems and wireless sensor networks. This can help drivers plan the best route, avoid potential traffic jams and accident risks, provide personalized navigation guidance, and improve driving safety and efficiency.

[0121] See also Fig. 9 ,The data collection module includes a data interface submodule, a data cleaning submodule, and a data preprocessing submodule. These modules collect vehicle data by communicating with the vehicle's electronic control unit, and clean and preprocess the data. These processing steps help extract effective vehicle information and provide a high-quality data foundation for subsequent analysis and application.

[0122] The driving status analysis module includes a data analysis submodule, a machine learning submodule, and an anomaly identification submodule. This module uses the collected vehicle data for analysis and applies machine learning algorithms to identify the vehicle's driving status. Through these analyses and identifications, an accurate driving status report can be provided to help drivers evaluate and improve their driving behavior.

[0123] The abnormal behavior verification module includes an image processing submodule, a target detection submodule, and a target tracking submodule. This module processes vehicle images captured by surveillance cameras and performs image processing, target detection, and target tracking to verify the presence of abnormal driving behavior. This helps monitor the driver's behavior in real time and promptly issue an alarm or take further measures to improve driving safety.

[0124] The traffic prediction module includes a big data analysis submodule, a flow prediction submodule, and a congestion prediction submodule. This module uses big data analysis technology to analyze vehicle behavior verification results and pre-processed vehicle data to predict traffic flow and the occurrence of congestion points. With these prediction results, drivers can choose a smoother driving route, improve driving efficiency, and reduce delays caused by congestion.

[0125] The intelligent navigation module includes a route planning submodule, a traffic information update submodule, and a cooperative driving submodule. This module combines traffic forecast results, geographic information systems, and wireless sensor networks to generate intelligent navigation suggestions and cooperative driving strategies. This intelligent navigation system can help drivers plan the best route, update real-time traffic information, and cooperate with other vehicles to improve overall traffic efficiency and safety.

[0126] Working principle:

[0127] First, the data collection module collects vehicle data by communicating with the vehicle's electronic control unit, and processes data such as speed, acceleration and steering angle through data cleaning and preprocessing algorithms to generate preprocessed vehicle data. The driving state analysis module uses the preprocessed vehicle data to analyze the vehicle's driving state using a support vector machine or decision tree algorithm to generate an abnormal driving report. The abnormal behavior verification module combines the abnormal driving report and uses computer vision target detection and target tracking algorithms to process vehicle images in real time to generate vehicle behavior verification results. Based on the vehicle behavior verification results and preprocessed vehicle data, the traffic prediction module uses cloud computing and big data analysis technology to predict traffic flow and congestion points and generate a traffic flow prediction report. The intelligent navigation module generates intelligent navigation suggestions and collaborative driving strategies based on traffic flow prediction reports, geographic information systems and wireless sensor networks.

[0128] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A vehicle monitoring method, characterized in that: The following steps are involved: The vehicle monitoring system is used to collect real-time vehicle data including speed, acceleration, and steering angle, and transmitted to the server via GPRS for storage and processing to generate a basic vehicle data set; Based on the basic vehicle data set, a machine learning algorithm including a support vector machine or a decision tree is used to analyze the driving state of the vehicle, identify abnormal driving behavior, and generate an abnormal driving report; The steps of abnormal driving report are specifically as follows: The principal component analysis method is used to extract feature vectors from the basic vehicle data set to generate a feature data set; Using a support vector machine algorithm with a radial basis function kernel, classifying the abnormal driving behavior on the feature data set to generate an abnormal driving candidate list; Using a decision tree algorithm with information gain as a feature selection method, screening the abnormal driving candidate list to generate a screened abnormal driving list; Based on the screened abnormal driving list, the template matching algorithm is used to generate an abnormal driving report; Combining the abnormal driving report with computer vision technology, the vehicle images captured by the surveillance camera are processed in real time through target detection and target tracking algorithms to confirm and verify abnormal driving behaviors and generate vehicle behavior verification results; The steps of verifying the vehicle behavior result are specifically as follows: Use target detection algorithms to analyze images captured by surveillance cameras and generate target detection results; Based on the target detection result, a simple online real-time tracking algorithm is used to track the target, and a target tracking result is generated based on the time series data; Using a data fusion algorithm, specifically a Kalman filter, to fuse the target tracking result and the abnormal driving report to generate fused data; Based on the fused data and the preset threshold, applying the Bayesian reasoning model to confirm and verify the abnormal driving behavior and generate a vehicle behavior verification result; Using cloud computing and big data analysis technology, the basic vehicle data set is deeply analyzed, and combined with the vehicle behavior verification results, traffic flow and congestion points are predicted to generate a traffic flow prediction report; The steps of the traffic flow forecast report are specifically as follows: Using the Hypertext Transfer Protocol secure protocol, uploading the basic vehicle data set and the vehicle behavior verification result to the cloud server to generate a cloud data set; Based on the cloud data set, a big data processing framework is used to perform data analysis and feature extraction to generate a traffic analysis report; Use the autoregressive integrated moving average model to conduct time series analysis on traffic analysis reports, predict traffic flow, and generate traffic flow prediction data; Integrate the traffic flow forecast data, apply data visualization technology, and use heat maps to generate a traffic flow forecast report; Based on the traffic flow forecast report and geographic information system, provide intelligent route planning for drivers, avoid potential traffic congestion and accident risk points, and generate intelligent navigation suggestions; The steps of the intelligent navigation suggestion are specifically as follows: Based on the traffic flow forecast report, using the shortest path algorithm and geographic information system, calculate the optimal route and generate optimal route data; Applying the risk assessment model, through the Markov decision process, to conduct risk assessment on the optimal route, and generating a risk assessment report; Combining the risk assessment report and the optimal route data, using a genetic algorithm to optimize the route and generate an optimized optimal route; Based on the generated optimized optimal route, the optimized optimal route is pushed to the vehicle information system in real time by using data wrapper technology and message queue telemetry transmission protocol to generate real-time navigation data and intelligent navigation suggestions; By combining the intelligent navigation suggestions and wireless sensor networks, information sharing and cooperative driving between vehicles are realized. Vehicles adjust themselves according to the driving status and road condition information of surrounding vehicles to generate cooperative driving strategies.

2. The vehicle monitoring method according to claim 1, characterized in that: The vehicle monitoring system is used to collect real-time vehicle data including speed, acceleration, and steering angle, and transmitted to the server via GPRS for storage and processing. The steps to generate the basic vehicle data set are as follows: Use the OBD-II interface to communicate with the vehicle's electronic control unit, obtain speed, acceleration, and steering angle parameters through the serial communication protocol, and generate raw vehicle data; Preprocessing the original vehicle data using a data cleaning algorithm including missing value filling and outlier removal to generate preprocessed vehicle data; Using the GPRS module and the transmission control protocol TCP, the pre-processed data is sent to the server to generate a transmission record; The server uses a relational database management system to store the transmission records, ensures data integrity through a data quality detection algorithm, and generates a basic vehicle data set.

3. The vehicle monitoring method according to claim 1, characterized in that: Combined with the intelligent navigation suggestions and wireless sensor networks, information sharing and cooperative driving between vehicles are realized. Vehicles adjust themselves according to the driving status and road condition information of surrounding vehicles. The specific steps of generating a cooperative driving strategy are as follows: Use event-driven architecture and complex event processing algorithms to monitor vehicle behavior, traffic flow, and road conditions in real time and generate real-time monitoring data; Based on the real-time monitoring data, natural language processing technology and speech recognition technology are used to generate voice reminders and warnings; Using embedded systems and microservice architecture, the real-time monitoring data is connected to the emergency response center to generate a coordinated driving strategy.

4. A vehicle monitoring system, characterized in that: The vehicle monitoring system is used to execute the vehicle monitoring method described in any one of claims 1-3, and the vehicle monitoring system is composed of a data collection module, a driving status analysis module, an abnormal behavior verification module, a traffic prediction module, and an intelligent navigation module.

5. The vehicle monitoring system according to claim 4, characterized in that: The data collection module communicates with the electronic control unit of the vehicle through the OBD-II interface, processes the collected speed, acceleration, and steering angle data using a data cleaning and preprocessing algorithm, and generates preprocessed vehicle data; The driving state analysis module analyzes the driving state of the vehicle based on the preprocessed vehicle data using a support vector machine or a decision tree algorithm to generate an abnormal driving report; The abnormal behavior verification module combines the abnormal driving report with the target detection and target tracking algorithm of computer vision to process the vehicle images captured by the surveillance camera in real time and generate the vehicle behavior verification result; The traffic prediction module predicts traffic flow and congestion points based on the vehicle behavior verification results and pre-processed vehicle data using cloud computing and big data analysis technology to generate a traffic flow prediction report; The intelligent navigation module generates intelligent navigation suggestions and collaborative driving strategies based on traffic flow forecast reports in combination with geographic information systems and wireless sensor networks.

6. The vehicle monitoring system according to claim 4, characterized in that: The data collection module includes a data interface submodule, a data cleaning submodule, and a data preprocessing submodule; The driving state analysis module includes a data analysis submodule, a machine learning submodule, and an abnormality recognition submodule; The abnormal behavior verification module includes an image processing submodule, a target detection submodule, and a target tracking submodule; The traffic prediction module includes a big data analysis submodule, a traffic flow prediction submodule, and a congestion prediction submodule; The intelligent navigation module includes a route planning submodule, a traffic information updating submodule, and a coordinated driving submodule.

Images