Vehicle carbon footprint monitoring and driver behavior analysis system

Through the vehicle carbon footprint monitoring and driver behavior analysis system, the real-time monitoring of vehicle carbon emissions and driver behavior analysis are achieved using multi-sensors and 3D-CNN models, providing personalized emission reduction suggestions, and solving the carbon emission monitoring in the existing technology The problems of inaccuracy and lack of intelligent analysis have improved drivers' emission reduction capabilities.

CN119049016APending Publication Date: 2024-11-29PEKING UNIV SHENZHEN GRADUATE SCHOOL
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410926199.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-11
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

Existing carbon emission monitoring technology cannot achieve accurate and real-time monitoring of vehicle carbon emission data, and lacks an intelligent analysis system, making it difficult to automatically identify and analyze drivers' complex driving behaviors, and cannot give targeted and effective emission reduction suggestions. .

Method used

A vehicle carbon footprint monitoring and driver behavior analysis system was designed, including vehicle carbon emission monitoring module, driver behavior analysis module and improved carbon emission prediction and recommendation module, and the driver's driving behavior is used to utilize multi-sensors and 3D-CNN models to deal with drivers' driving behavior. Carry out comprehensive analysis, measure and record carbon emission data in real time, and provide personalized emission reduction suggestions.

Benefits of technology

Real-time monitoring and intelligent analysis of vehicle carbon emission data is realized, providing drivers with personalized and accurate emission reduction suggestions, helping drivers reduce carbon emissions more scientifically and efficiently.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049016A_ABST
    Figure CN119049016A_ABST
Patent Text Reader

Abstract

The invention discloses a vehicle carbon footprint monitoring and driver behavior analysis system, and belongs to the technical field of traffic transportation carbon emission monitoring and improvement. A vehicle carbon emission monitoring module tracks the carbon footprint in the vehicle driving process in real time, and Mixup is combined in a driver behavior analysis module to enhance the robustness of training data; a 3D-CNN model is used for conducting three-dimensional video accurate analysis on the driving habits of a driver, the conversion relation between the carbon emission footprint and the driver behaviors is obtained, the carbon footprint is fed back to the driver, the driver can clearly know the influence of the driving behaviors of the driver on carbon emission, and according to the driving habits of the driver, the driving behavior of the driver can be accurately analyzed. And personalized carbon emission reduction suggestions are provided. Compared with an existing vehicle carbon emission improvement mode, the method has the advantages that real-time monitoring and intelligent and accurate analysis of carbon emission data are achieved, personalized and accurate emission reduction suggestions are provided for a driver, and the driver is helped to reduce carbon emission more scientifically and more efficiently.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of transportation carbon emission monitoring and improvement technology, and in particular to a vehicle carbon footprint monitoring and driver behavior analysis system. Background Art

[0002] Today, the problem of carbon emissions is becoming increasingly serious, causing tremendous pressure on the global climate and environment. Vehicle carbon emissions are the main source of carbon emissions. With the rapid growth of the number of cars, the emission of greenhouse gases such as carbon dioxide generated by vehicles during operation continues to rise, exacerbating the speed of global warming. Therefore, solving the problem of vehicle carbon emissions has become a top priority.

[0003] Existing vehicle emission reduction technologies include promoting new energy vehicles and improving fuel efficiency. New energy vehicles, with their zero-emission or low-emission characteristics, effectively reduce the pollution caused by traditional fuel vehicles and promote the transformation of the transportation sector to clean and low-carbon. At the same time, improving fuel efficiency means that vehicles can travel longer distances per unit of fuel, which not only reduces energy consumption but also reduces greenhouse gas emissions. In addition, the application and promotion of many vehicle emission reduction technologies are of great significance to reducing vehicle carbon emissions, protecting the environment and achieving sustainable development.

[0004] However, existing carbon emission monitoring technology is still imperfect, and it is impossible to monitor vehicle carbon emission data accurately and in real time, making it difficult to fully reflect the vehicle's carbon emission status; at the same time, drivers lack intelligent analysis systems and have no means to receive real-time feedback on carbon emissions, making it difficult to automatically and accurately identify and analyze complex driving behaviors, and unable to provide targeted and effective emission reduction suggestions, making it difficult for drivers themselves to play a role in vehicle emission reduction.

[0005] Therefore, there is an urgent need for an intelligent system that can achieve real-time tracking of vehicle carbon emissions and provide drivers with driving behavior suggestions to solve the above problems. Summary of the invention

[0006] In order to solve the defects in the prior art, the present invention proposes a vehicle carbon footprint monitoring and driver behavior analysis system.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] A vehicle carbon footprint monitoring and driver behavior analysis system includes a vehicle carbon emission monitoring module, a driver behavior analysis module, and an improved carbon emission prediction and suggestion module:

[0009] The vehicle carbon emission monitoring module integrates multiple sensors and a data processing system to measure and record the vehicle's carbon emission data in real time;

[0010] The driver behavior analysis module obtains basic feature data and uses 3D-CNN to conduct a comprehensive analysis of the driver's driving behavior to identify bad driving habits that may lead to high carbon emissions. At the same time, it also analyzes the driver's activity mileage, travel time, and driving habit information to provide basic data for subsequent carbon emission predictions and recommendations.

[0011] The improved carbon emission prediction and suggestion module is used to predict the carbon emission of the vehicle under different driving behaviors and provide improvement suggestions to the driver.

[0012] Furthermore, the vehicle carbon emission monitoring module includes an emission detection sensor, a GPS positioning module, a data processing and analysis unit, and a data transmission and communication module:

[0013] The emission detection sensor uses non-dispersive infrared and chemiluminescence technology to measure the content of carbon dioxide and other greenhouse gases in vehicle exhaust in real time;

[0014] Wherein, the other greenhouse gases include methane, nitrous oxide, hydrofluorocarbons, perfluorocarbons and sulfur hexafluoride;

[0015] The GPS positioning module is used to record the vehicle's location information and generate trajectory information, and analyze the carbon emissions under different driving routes and driving conditions by combining it with carbon emission data;

[0016] The data processing and analysis unit is used to collect data from the emission detection sensor, and then clean, calibrate and convert it to ensure the accuracy and reliability of the measurement results; in particular, the other greenhouse gases are converted into carbon equivalents through their corresponding conversion coefficients, so that different types of greenhouse gases are measured in the same reference object:

[0017] Ceq=EX CO2 +EX CH4 ×GWP CH4 +EX N2O ×GWP N2O

[0018] Where Ceq represents carbon equivalent, EX CO2 、EX CH4 、EX N2O Represents the emissions of carbon dioxide, methane and nitrous oxide, respectively, GWP CH4 and GWP N2O represents the global warming potential of methane and nitrous oxide, respectively, which are multiples of their warming effect relative to carbon dioxide;

[0019] The data transmission and communication module is used to transmit the processed carbon emission data to the on-board intelligent device through wired communication technology, and at the same time transmit it to the remote data center in real time through Wi-Fi wireless communication technology.

[0020] Furthermore, the driver behavior analysis module includes a data acquisition module and a model classification module:

[0021] Wherein, the data acquisition module includes an in-vehicle video acquisition unit, a vehicle operation sensing unit, and a vehicle dynamic information GPS measurement unit;

[0022] Furthermore, the model classification module comprises the following steps:

[0023] a. The in-vehicle video data, vehicle operation information and vehicle dynamic parameters are obtained through the data acquisition module, the in-vehicle video data is processed by Kalman filtering to filter out noise, the vehicle operation information and vehicle dynamic parameters are processed by STFT, and the information about the change of the signal frequency component over time is provided from the time domain to the time-frequency domain, so as to improve the in-vehicle video data to capture the consistent dynamic characteristics of driving behavior:

[0024]

[0025] Among them, x(t) is the input vehicle manipulation information and vehicle dynamic parameter signal, w(t-τ) is the window function used to divide the signal into different segments, and e -jωt is a complex exponential function used to calculate the Fourier transform. STFT{x(t)}(τ,ω) is the result of the short-time Fourier transform, which represents the signal amplitude and phase at the time point τ and frequency ω;

[0026] b. Use Mixup to enhance training data, fuse samples and labels to improve the generalization ability of the model and the robustness of adversarial samples:

[0027]

[0028] Among them, L mix_up represents the loss function, D C represents cosine similarity, τ represents temperature parameter, z (1) ,z (2) Represents the feature vector of positive and negative samples, represents the mixed sample, λ represents the mixing amount, i and k are the horizontal and vertical coordinates of the image;

[0029] c. Design a 3D-CNN architecture, including at least 6 3D convolutional layers, pooling layers, and fully connected layers, and determine the convolution kernel size, step size, padding parameters, optimizer and learning rate training parameters in the network, and use data enhancement technology to improve the generalization ability of the model;

[0030] d. Input the preprocessed video data into the 3D-CNN model and train the model through the back-propagation algorithm and gradient descent optimizer;

[0031] e. The trained 3D-CNN model extracts features related to driver behavior from the in-vehicle video data and creates a feature set by combining vehicle handling information and vehicle dynamic parameters;

[0032] f. Use the extracted features to train SVM and RF classifiers to identify different behaviors of drivers;

[0033] g. Use an independent test set to evaluate the performance of the model, calculate the accuracy, recall, and F1 value, and tune the model based on the evaluation results:

[0034]

[0035] Among them, P is precision, R is recall, TP is true positive, FP is false positive, FN is false negative, and F1 is the F1 value;

[0036] h. Deploy the trained model to the vehicle, capture feature sample data in real time through the data acquisition module, use the deployed model to process and analyze real-time video data, identify the driver's current behavior, and provide feedback to the improved carbon emission prediction and suggestion module.

[0037] Furthermore, the improved carbon emission prediction and suggestion module includes a carbon emission real-time feedback module, a driving behavior improvement suggestion module and a report management module:

[0038] The carbon emission real-time feedback module is connected to the data transmission and communication module through wired communication technology, calculates carbon emissions and carbon equivalents in real time, and compares them with preset emission standards, quickly generates emission data reports, and helps drivers understand the impact of their driving behavior on carbon emissions;

[0039] The driving behavior improvement suggestion module, in combination with the vehicle carbon emission monitoring module and the driver behavior analysis module, provides personalized driving suggestions or warnings to drivers based on the analysis results to help them improve their driving habits and reduce safety hazards;

[0040] The report management module is used to provide report management functions. Users can view and download reports through the system interface to facilitate data analysis and decision-making.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention proposes a vehicle carbon footprint monitoring and driver behavior analysis system, which tracks the carbon footprint of the vehicle in real time during driving through the vehicle carbon emission monitoring module, combines Mixup in the driver behavior analysis module to enhance the robustness of training data, uses a 3D-CNN model to perform three-dimensional video precision analysis of the driver's driving habits, obtains the conversion relationship between carbon emission footprint and driver behavior, and feeds back the carbon footprint to the driver, so that they can clearly understand the impact of their driving behavior on carbon emissions, and provide personalized carbon emission reduction suggestions based on the driver's driving habits. Compared with the existing vehicle carbon emission improvement methods, the present invention realizes real-time monitoring and intelligent and accurate analysis of carbon emission data, provides personalized and accurate emission reduction suggestions for drivers, and helps drivers reduce carbon emissions more scientifically and efficiently. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0044] Figure 1 This is a schematic diagram of functional modules of a vehicle carbon footprint monitoring and driver behavior analysis system in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is 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.

[0046] In one embodiment, Figure 1 As shown, a vehicle carbon footprint monitoring and driver behavior analysis system is provided, including a vehicle carbon emission monitoring module, a driver behavior analysis module, and an improved carbon emission prediction and suggestion module:

[0047] The vehicle carbon emission monitoring module integrates multiple sensors and a data processing system to measure and record the vehicle's carbon emission data in real time;

[0048] The driver behavior analysis module obtains basic feature data and uses 3D-CNN to conduct a comprehensive analysis of the driver's driving behavior to identify bad driving habits that may lead to high carbon emissions. At the same time, it also analyzes the driver's activity mileage, travel time, and driving habit information to provide basic data for subsequent carbon emission predictions and recommendations.

[0049] The improved carbon emission prediction and suggestion module is used to predict the carbon emission of the vehicle under different driving behaviors and provide improvement suggestions to the driver.

[0050] Furthermore, the vehicle carbon emission monitoring module includes an emission detection sensor, a GPS positioning module, a data processing and analysis unit, and a data transmission and communication module:

[0051] The emission detection sensor uses non-dispersive infrared and chemiluminescence technology to measure the content of carbon dioxide and other greenhouse gases in vehicle exhaust in real time;

[0052] Wherein, the other greenhouse gases include methane, nitrous oxide, hydrofluorocarbons, perfluorocarbons and sulfur hexafluoride;

[0053] The GPS positioning module is used to record the vehicle's location information and generate trajectory information, and analyze the carbon emissions under different driving routes and driving conditions by combining it with carbon emission data;

[0054] The data processing and analysis unit is used to collect data from the emission detection sensor, and then clean, calibrate and convert it to ensure the accuracy and reliability of the measurement results; in particular, the other greenhouse gases are converted into carbon equivalents through their corresponding conversion coefficients, so that different types of greenhouse gases are measured in the same reference object:

[0055] Ceq=EX CO2 +EX CH4 ×GWP CH4 +EX N2O ×GWP N2O

[0056] Where Ceq represents carbon equivalent, EX CO2 、EX CH4 、EX N2O Represents the emissions of carbon dioxide, methane and nitrous oxide, respectively, GWP CH4 and GWP N2O represents the global warming potential of methane and nitrous oxide, respectively, which are multiples of their warming effect relative to carbon dioxide;

[0057] The data transmission and communication module is used to transmit the processed carbon emission data to the on-board intelligent device through wired communication technology, and at the same time transmit it to the remote data center in real time through Wi-Fi wireless communication technology.

[0058] Furthermore, the driver behavior analysis module includes a data acquisition module and a model classification module:

[0059] Wherein, the data acquisition module includes an in-vehicle video acquisition unit, a vehicle operation sensing unit, and a vehicle dynamic information GPS measurement unit;

[0060] Furthermore, the model classification module comprises the following steps:

[0061] a. The in-vehicle video data, vehicle operation information and vehicle dynamic parameters are obtained through the data acquisition module, the in-vehicle video data is processed by Kalman filtering to filter out noise, the vehicle operation information and vehicle dynamic parameters are processed by STFT, and the information about the change of the signal frequency component over time is provided from the time domain to the time-frequency domain, so as to improve the in-vehicle video data to capture the consistent dynamic characteristics of driving behavior:

[0062]

[0063] Among them, x(t) is the input vehicle manipulation information and vehicle dynamic parameter signal, w(t-τ) is the window function used to divide the signal into different segments, and e -jωt is a complex exponential function used to calculate the Fourier transform. STFT{x(t)}(τ,ω) is the result of the short-time Fourier transform, which represents the signal amplitude and phase at the time point τ and frequency ω;

[0064] b. Use Mixup to enhance training data, fuse samples and labels to improve the generalization ability of the model and the robustness of adversarial samples:

[0065]

[0066] Among them, L mix_up represents the loss function, D C represents cosine similarity, τ represents temperature parameter, z (1) ,z (2) Represents the feature vector of positive and negative samples, represents the mixed sample, λ represents the mixing amount, i and k are the horizontal and vertical coordinates of the image;

[0067] c. Design a 3D-CNN architecture, including at least 6 3D convolutional layers, pooling layers, and fully connected layers, and determine the convolution kernel size, step size, padding parameters, optimizer and learning rate training parameters in the network, and use data enhancement technology to improve the generalization ability of the model;

[0068] d. Input the preprocessed video data into the 3D-CNN model and train the model through the back-propagation algorithm and gradient descent optimizer;

[0069] e. The trained 3D-CNN model extracts features related to driver behavior from the in-vehicle video data and creates a feature set by combining vehicle handling information and vehicle dynamic parameters;

[0070] f. Use the extracted features to train SVM and RF classifiers to identify different behaviors of drivers;

[0071] g. Use an independent test set to evaluate the performance of the model, calculate the accuracy, recall, and F1 value, and tune the model based on the evaluation results:

[0072]

[0073] Among them, P is precision, R is recall, TP is true positive, FP is false positive, FN is false negative, and F1 is the F1 value;

[0074] h. Deploy the trained model to the vehicle, capture feature sample data in real time through the data acquisition module, use the deployed model to process and analyze real-time video data, identify the driver's current behavior, and provide feedback to the improved carbon emission prediction and suggestion module.

[0075] Furthermore, the improved carbon emission prediction and suggestion module includes a carbon emission real-time feedback module, a driving behavior improvement suggestion module and a report management module:

[0076] The carbon emission real-time feedback module is connected to the data transmission and communication module through wired communication technology, calculates carbon emissions and carbon equivalents in real time, and compares them with preset emission standards, quickly generates emission data reports, and helps drivers understand the impact of their driving behavior on carbon emissions;

[0077] The driving behavior improvement suggestion module, in combination with the vehicle carbon emission monitoring module and the driver behavior analysis module, provides personalized driving suggestions or warnings to drivers based on the analysis results to help them improve their driving habits and reduce safety hazards;

[0078] The report management module is used to provide report management functions. Users can view and download reports through the system interface to facilitate data analysis and decision-making.

Claims

1. A vehicle carbon footprint monitoring and driver behavior analysis system, characterized in that: Including vehicle carbon emission monitoring module, driver behavior analysis module and improved carbon emission prediction and suggestion module: The vehicle carbon emission monitoring module integrates multiple sensors and a data processing system to measure and record the vehicle's carbon emission data in real time; The driver behavior analysis module obtains basic feature data and uses 3D-CNN to conduct a comprehensive analysis of the driver's driving behavior to identify bad driving habits that may lead to high carbon emissions. At the same time, it also analyzes the driver's activity mileage, travel time, and driving habit information to provide basic data for subsequent carbon emission predictions and recommendations. The improved carbon emission prediction and suggestion module is used to predict the carbon emission of the vehicle under different driving behaviors and provide improvement suggestions to the driver.

2. A vehicle carbon footprint monitoring and driver behavior analysis system according to claim 1, characterized in that: The vehicle carbon emission monitoring module includes an emission detection sensor, a GPS positioning module, a data processing and analysis unit, and a data transmission and communication module: The emission detection sensor uses non-dispersive infrared and chemiluminescence technology to measure the content of carbon dioxide and other greenhouse gases in vehicle exhaust in real time; Wherein, the other greenhouse gases include methane, nitrous oxide, hydrofluorocarbons, perfluorocarbons and sulfur hexafluoride; The GPS positioning module is used to record the vehicle's location information and generate trajectory information, and analyze the carbon emissions under different driving routes and driving conditions by combining it with carbon emission data; The data processing and analysis unit is used to collect data from the emission detection sensor, and then clean, calibrate and convert it to ensure the accuracy and reliability of the measurement results; in particular, the other greenhouse gases are converted into carbon equivalents through their corresponding conversion coefficients, so that different types of greenhouse gases are measured in the same reference object: Ceq=EX CO2 +EX CH4 ×GWP CH4 +EX N2O ×GWP N2O Where Ceq represents carbon equivalent, EX CO2 、EX CH4 、EX N2O Represents the emissions of carbon dioxide, methane and nitrous oxide, respectively, GWP CH4 and GWP N2O represents the global warming potential of methane and nitrous oxide, respectively, which are multiples of their warming effect relative to carbon dioxide; The data transmission and communication module is used to transmit the processed carbon emission data to the on-board intelligent device through wired communication technology, and at the same time transmit it to the remote data center in real time through Wi-Fi wireless communication technology.

3. A vehicle carbon footprint monitoring and driver behavior analysis system according to claim 1, characterized in that: The driver behavior analysis module includes a data acquisition module and a model classification module: Wherein, the data acquisition module includes an in-vehicle video acquisition unit, a vehicle operation sensor unit, and a vehicle dynamic information GPS measurement unit.

4. The driver behavior analysis module according to claim 3, characterized in that: The model classification module comprises the following steps: a. The in-vehicle video data, vehicle operation information and vehicle dynamic parameters are obtained through the data acquisition module, the in-vehicle video data is processed by Kalman filtering to filter out noise, the vehicle operation information and vehicle dynamic parameters are processed by STFT, and the information about the change of the signal frequency component over time is provided from the time domain to the time-frequency domain, so as to improve the in-vehicle video data to capture the consistent dynamic characteristics of driving behavior: Among them, x(t) is the input vehicle manipulation information and vehicle dynamic parameter signal, w(t-τ) is the window function used to divide the signal into different segments, and e -jωt is a complex exponential function used to calculate the Fourier transform. STFT{x(t)}(τ,ω) is the result of the short-time Fourier transform, which represents the signal amplitude and phase at the time point τ and frequency ω; b. Use Mixup to enhance training data, fuse samples and labels to improve the generalization ability of the model and the robustness of adversarial samples: Among them, L mix_up represents the loss function, D C represents cosine similarity, τ represents temperature parameter, z (1) ,z (2) Represents the feature vector of positive and negative samples, represents the mixed sample, λ represents the mixing amount, i and k are the horizontal and vertical coordinates of the image; c. Design a 3D-CNN architecture, including at least 6 3D convolutional layers, pooling layers, and fully connected layers, and determine the convolution kernel size, step size, padding parameters, optimizer and learning rate training parameters in the network, and use data enhancement technology to improve the generalization ability of the model; d. Input the preprocessed video data into the 3D-CNN model and train the model through the back-propagation algorithm and gradient descent optimizer; e. The trained 3D-CNN model extracts features related to driver behavior from the in-vehicle video data and creates a feature set by combining vehicle handling information and vehicle dynamic parameters; f. Use the extracted features to train SVM and RF classifiers to identify different behaviors of drivers; g. Use an independent test set to evaluate the performance of the model, calculate the accuracy, recall, and F1 value, and tune the model based on the evaluation results: Among them, P is precision, R is recall, TP is true positive, FP is false positive, FN is false negative, and F1 is the F1 value; h. Deploy the trained model to the vehicle, capture feature sample data in real time through the data acquisition module, use the deployed model to process and analyze real-time video data, identify the driver's current behavior, and provide feedback to the improved carbon emission prediction and suggestion module.

5. The vehicle carbon footprint monitoring and driver behavior analysis system according to claim 1, characterized in that: The improved carbon emission prediction and suggestion module includes a carbon emission real-time feedback module, a driving behavior improvement suggestion module and a report management module: The carbon emission real-time feedback module is connected to the data transmission and communication module through wired communication technology, calculates carbon emissions and carbon equivalents in real time, and compares them with preset emission standards, quickly generates emission data reports, and helps drivers understand the impact of their driving behavior on carbon emissions; The driving behavior improvement suggestion module, in combination with the vehicle carbon emission monitoring module and the driver behavior analysis module, provides personalized driving suggestions or warnings to drivers based on the analysis results to help them improve their driving habits and reduce safety hazards; The report management module is used to provide report management functions. Users can view and download reports through the system interface to facilitate data analysis and decision-making.

Citation Information

Cited By

  • Automobile carbon footprint database construction method

    CN120873401A

  • Ship greenhouse gas carbon emission data processing method, device and equipment

    CN121051124A