Tire pressure sensor activation method, device, equipment and storage medium

Abstract

The present invention relates to the field of sensor activation, and in particular to a tire pressure sensor activation method, device, equipment and storage medium. The method comprises the following steps: obtaining vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array; performing vehicle vibration time-frequency analysis on the vehicle operation status monitoring parameters, and performing vehicle vibration situation mining to construct a vehicle vibration situation curve; performing road object recognition on the vehicle driving monitoring image based on the vehicle vibration situation curve, and performing real-time road condition scene structure rendering to construct a real-time road condition scene model; obtaining a vehicle historical behavior log; performing multi-time point vehicle position calculation on the vehicle historical behavior log, and reconstructing the dynamic vehicle movement trajectory, thereby constructing a historical vehicle dynamic trajectory; performing dynamic behavior feature analysis on the historical vehicle dynamic trajectory to construct a personalized driving portrait of the owner. The present invention realizes efficient and rapid tire pressure sensor activation.

Description

Technical Field

[0001] The present invention relates to the field of sensor activation, and in particular to an activation method, device, equipment and storage medium for a tire pressure sensor. Background Art

[0002] With the rapid development of the automobile industry and the continuous improvement of the level of intelligence, the tire pressure monitoring system (TPMS) has become an indispensable part of the modern automobile safety system. As the core component of the tire pressure monitoring system, the tire pressure sensor is mainly responsible for real-time monitoring of the tire pressure status and transmitting data to the vehicle control system to promptly remind the driver of abnormal tire pressure. Abnormal tire pressure not only affects the fuel efficiency and handling performance of the car, but also leads to serious safety accidents such as tread damage and tire blowout. Therefore, accurate and timely detection and adjustment of tire pressure has become the key to ensure driving safety.

[0003] In practical applications, the activation process of tire pressure sensors is crucial. Traditional tire pressure sensor activation methods usually rely on manual operation or configuration through vehicle diagnostic equipment, but these methods often have problems such as cumbersome operation, unstable activation process, and strong equipment dependence, resulting in untimely sensor activation or activation failure, thereby affecting the normal operation of the monitoring system. With the development of intelligent vehicle technology, traditional activation methods are gradually unable to meet the market's demand for high-efficiency and high-reliability activation processes. Therefore, there is an urgent need for an innovative and efficient tire pressure sensor activation method that can reduce human intervention while improving the accuracy and stability of the activation process, thereby ensuring that the tire pressure monitoring system can provide tire pressure information in real time and accurately, and improve the overall safety and driving experience of the vehicle. Therefore, the development of a more intelligent, efficient and universal tire pressure sensor activation method has become an important topic in current research and application. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a tire pressure sensor activation method, device, equipment and storage medium to solve at least one of the above technical problems.

[0005] To achieve the above object, the present invention provides a method for activating a tire pressure sensor, comprising the following steps:

[0006] Step S1: acquiring vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array; performing vehicle vibration time-frequency analysis on the vehicle operation status monitoring parameters, and performing vehicle vibration situation mining to construct a vehicle vibration situation curve;

[0007] Step S2: performing road object recognition on the vehicle driving monitoring image based on the vehicle vibration situation curve, and performing real-time road condition scene structure rendering to construct a real-time road condition scene model;

[0008] Step S3: Obtain the vehicle historical behavior log; calculate the vehicle position at multiple time points on the vehicle historical behavior log, and reconstruct the dynamic vehicle movement trajectory, thereby constructing the historical vehicle dynamic trajectory;

[0009] Step S4: Analyze the dynamic behavior characteristics of historical vehicle dynamic trajectories, and perform personalized driving behavior fitting of the owner to construct a personalized driving profile of the owner;

[0010] Step S5: predicting the personalized driving behavior of the owner in the real-time scene based on the personalized driving portrait of the owner, making intelligent sensor activation decisions, and building an intelligent activation strategy for the tire pressure sensor;

[0011] Step S6: Perform instant activation processing based on the tire pressure sensor intelligent activation strategy, then optimize the activation response delay and build a dynamic activation optimization engine.

[0012] The present invention also provides an activation device for a tire pressure sensor:

[0013] The vibration situation module acquires vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array; performs vehicle vibration time-frequency analysis on the vehicle operation status monitoring parameters, performs vehicle vibration situation mining, and constructs a vehicle vibration situation curve;

[0014] The road condition scene module recognizes road objects in the vehicle driving monitoring image based on the vehicle vibration situation curve, renders the real-time road condition scene structure, and builds a real-time road condition scene model;

[0015] The vehicle dynamic trajectory module obtains the vehicle historical behavior log; calculates the vehicle position at multiple time points on the vehicle historical behavior log, and reconstructs the dynamic vehicle movement trajectory, thereby constructing the historical vehicle dynamic trajectory;

[0016] The personalized portrait module analyzes the dynamic behavior characteristics of historical vehicle dynamic trajectories, fits the owner's personalized driving behavior, and constructs a personalized driving portrait of the owner;

[0017] The adaptive activation module predicts the personalized driving behavior of the owner in real-time scenarios based on the personalized driving portrait of the owner, makes intelligent sensor activation decisions, and builds an intelligent activation strategy for tire pressure sensors.

[0018] The activation optimization module performs instant activation processing based on the tire pressure sensor's intelligent activation strategy, and then optimizes the activation response delay to build a dynamic activation optimization engine.

[0019] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above-mentioned tire pressure sensor activation methods when executing the computer program.

[0020] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of any of the above-mentioned methods for activating a tire pressure sensor are implemented.

[0021] The tire pressure sensor activation method, device, equipment and storage medium provided by the present invention have the following beneficial effects: vehicle vibration time-frequency analysis and vibration situation mining help to understand the vehicle vibration situation and provide vibration feature information for subsequent processing; construction of a vehicle vibration situation curve can help track vehicle vibration changes and provide a dynamic visual display of the vehicle's driving status; road object recognition and road condition scene rendering help to identify obstacles on the road and real-time road conditions and improve driving safety; construction of a real-time road condition scene model can provide the driver with real-time road condition information and help make more informed driving decisions; multi-time point vehicle position calculation and dynamic vehicle movement trajectory reconstruction help to understand the vehicle's driving history and movement trajectory; construction of a historical vehicle dynamic trajectory can provide a vehicle A comprehensive understanding of the vehicle's driving habits and trajectories provides data support for subsequent analysis. Dynamic behavior feature analysis and driving behavior fitting help understand the owner's driving habits and behavioral characteristics. Building a personalized driving portrait of the owner can provide a basis for personalized driving behavior prediction and help improve driving experience and safety. Real-time scenario driving behavior prediction and intelligent activation decisions can help make intelligent driving decisions based on the owner's habits and real-time road conditions. Building a tire pressure sensor intelligent activation strategy can improve activation accuracy and efficiency and enhance vehicle safety and performance. The dynamic activation optimization engine helps optimize activation processing and response delay, improve sensor activation efficiency and response speed, and build an optimization engine to continuously improve the activation strategy, making the activation processing of the tire pressure sensor more flexible and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic flow chart of the steps of a method for activating a tire pressure sensor according to the present invention;

[0023] Figure 2 Detailed implementation flow chart of step S1;

[0024] Figure 3 Detailed implementation flow chart of step S2;

[0025] Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0026] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0027] The present application example provides a tire pressure sensor activation method, device, equipment and storage medium. The execution subjects of the tire pressure sensor activation method, device, equipment and storage medium include but are not limited to the following: mechanical equipment, data processing platform, cloud server node, network upload equipment, etc. that are equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio image management system, an information management system, and a cloud data management system.

[0028] See also Figures 1 to 4 The present invention provides a method for activating a tire pressure sensor, the method for activating a tire pressure sensor comprising the following steps:

[0029] Step S1: acquiring vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array; performing vehicle vibration time-frequency analysis on the vehicle operation status monitoring parameters, and performing vehicle vibration situation mining to construct a vehicle vibration situation curve;

[0030] Step S2: performing road object recognition on the vehicle driving monitoring image based on the vehicle vibration situation curve, and performing real-time road condition scene structure rendering to construct a real-time road condition scene model;

[0031] Step S3: Obtain the vehicle historical behavior log; calculate the vehicle position at multiple time points on the vehicle historical behavior log, and reconstruct the dynamic vehicle movement trajectory, thereby constructing the historical vehicle dynamic trajectory;

[0032] Step S4: Analyze the dynamic behavior characteristics of historical vehicle dynamic trajectories, and perform personalized driving behavior fitting of the owner to construct a personalized driving profile of the owner;

[0033] Step S5: predicting the personalized driving behavior of the owner in the real-time scene based on the personalized driving portrait of the owner, making intelligent sensor activation decisions, and building an intelligent activation strategy for the tire pressure sensor;

[0034] Step S6: Perform instant activation processing based on the tire pressure sensor intelligent activation strategy, then optimize the activation response delay and build a dynamic activation optimization engine.

[0035] The vehicle vibration time-frequency analysis and vibration situation mining can help understand the vehicle vibration situation and provide vibration feature information for subsequent processing. The construction of vehicle vibration situation curve can help track the changes in vehicle vibration and provide a dynamic visual display of the vehicle's driving status. Road object recognition and road scene rendering can help identify obstacles on the road and real-time road conditions, improve driving safety, and build a real-time road scene model to provide drivers with real-time road information to help make more informed driving decisions. Multi-time point vehicle position calculation and dynamic vehicle movement trajectory reconstruction can help understand the vehicle's driving history and movement trajectory. The construction of historical vehicle dynamic trajectory can provide a comprehensive understanding of the vehicle's driving habits and trajectory, which is helpful for subsequent analysis. Analysis provides data support, dynamic behavior feature analysis and driving behavior fitting help to understand the owner's driving habits and behavioral characteristics, building a personalized driving portrait of the owner can provide a basis for personalized driving behavior prediction, help improve driving experience and safety, real-time scenario driving behavior prediction and intelligent activation decision-making can help make intelligent driving decisions based on the owner's habits and real-time road conditions, building a tire pressure sensor intelligent activation strategy can improve activation accuracy and efficiency, enhance vehicle safety and performance, dynamic activation optimization engine helps to optimize activation processing and response delay, improve sensor activation efficiency and response speed, and build an optimization engine to continuously improve the activation strategy, making the activation processing of tire pressure sensors more flexible and efficient.

[0036] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a method for activating a tire pressure sensor of the present invention. In this example, the steps of the method for activating a tire pressure sensor include:

[0037] Step S1: acquiring vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array; performing vehicle vibration time-frequency analysis on the vehicle operation status monitoring parameters, and performing vehicle vibration situation mining to construct a vehicle vibration situation curve;

[0038] In this embodiment, a suitable sensor array is selected, including an accelerometer, a gyroscope, a GPS and a camera, etc. These sensors can monitor the running status and environmental factors of the vehicle in real time. The sensors are evenly arranged in key parts of the vehicle, such as wheels, chassis and body, to ensure that comprehensive vehicle dynamic information is obtained. The accelerometer is used to collect the vibration data of the vehicle, the gyroscope is used to measure the angular velocity of the vehicle, and the GPS Provide position and speed information, develop a data acquisition module, and use sensors to collect vehicle operating status monitoring parameters in real time, including the vehicle's acceleration, steering angle, speed, position information, and driving monitoring images. Ensure the time synchronization of each sensor data and use a unified timestamp to record the data of each sensor so that data from different sources can be accurately matched during subsequent analysis. Clean the collected raw data to remove noise and outliers. This process includes using filters (such as low-pass filters) to remove high-frequency noise to ensure data reliability. Format the cleaned data for subsequent analysis. Usually, it is necessary to convert the data into a unified format, such as a time series data format, for easy processing and storage. Select a suitable time-frequency analysis method, such as short-time Fourier transform (STFT) or wavelet transform. These methods can effectively analyze the temporal changes in the frequency characteristics of vehicle vibration signals. , perform time-frequency analysis on the pre-processed vehicle vibration data, extract the frequency components and time domain characteristics of the vibration signal, identify the vibration characteristics of the vehicle under different working conditions by analyzing the spectrum changes in different time periods, extract key features such as vibration amplitude, frequency change rate and main frequency components from the results of time-frequency analysis, select features related to the vehicle operating status for subsequent analysis and modeling, use statistical analysis methods (such as variance analysis) to analyze the extracted features, and evaluate their correlation with the vehicle operating status (such as driving mode, road conditions, etc.), and construct the vehicle vibration status curve based on the extracted vibration status features. This process involves visualizing the feature data to form a time series chart that shows the vibration changes of the vehicle in different time periods. The generated vibration status curve can be used for subsequent vehicle status monitoring, fault diagnosis and driving performance evaluation to help identify potential problems and optimize vehicle performance.

[0039] Step S2: performing road object recognition on the vehicle driving monitoring image based on the vehicle vibration situation curve, and performing real-time road condition scene structure rendering to construct a real-time road condition scene model;

[0040] In this embodiment, driving monitoring images are acquired in real time from the vehicle's camera. These images usually contain information such as roads, traffic signs, other vehicles and pedestrians, and are basic data for object recognition. The previously generated vehicle vibration situation curve is collected to extract features related to the vehicle's driving state. These features can help identify objects under specific driving conditions, such as dynamic changes during sharp turns or acceleration. The collected monitoring images are pre-processed, and image enhancement techniques (such as histogram equalization) and denoising algorithms (such as Gaussian filtering) are used to improve image quality. This process is intended to improve the accuracy of subsequent object recognition. According to actual needs, the images are cropped and scaled to ensure that the images input into the object recognition model have a uniform size to improve processing efficiency. A suitable deep learning object recognition model is selected, such as YOLO (You Only Look Once), Faster R-CNN or SSD (Single Shot MultiBox Detector), these models can effectively identify and locate multiple objects in real-time scenes. If the available data set is not sufficient to meet the needs, the transfer learning method can be used to fine-tune the pre-trained model on a specific road scene. By training on a specific data set, the model's recognition ability for specific objects (such as traffic signs, pedestrians and other vehicles) is improved. The pre-processed image is input into the object recognition model, and forward propagation is performed to perform object detection and classification. The model will output the category and location information (bounding box) of each recognized object. The recognition results are screened and the confidence threshold is set to exclude inaccurately recognized objects. The non-maximum suppression (NMS) algorithm is used to remove the bounding boxes of repeated detections to ensure that the final output object is accurately positioned. Based on the recognized road objects and their location information, a real-time road scene model is constructed. The recognition results are combined with the vehicle vibration situation curve to analyze the impact of objects in different dynamic states. Use 3D modeling tools (such as Blender or Unity) to convert the recognized objects into 3D models and adjust their positions in the scene to ensure accurate placement of the objects. Select a suitable rendering engine (such as Unreal Engine or Unity). 3D) for scene rendering. These engines support real-time rendering and can dynamically update the scene according to the driving status of the vehicle, realize the dynamic update mechanism of the scene, continuously update the position and status of objects in the scene according to the driving status of the vehicle and the real-time recognition results, and use LOD (level of detail) technology to dynamically adjust the rendering quality according to the distance between the object and the camera to optimize performance.

[0041] Step S3: Obtain the vehicle historical behavior log; calculate the vehicle position at multiple time points on the vehicle historical behavior log, and reconstruct the dynamic vehicle movement trajectory, thereby constructing the historical vehicle dynamic trajectory;

[0042] In this embodiment, the source of the vehicle historical behavior log is determined, which usually includes the on-board diagnostic system (OBD), GPS device, vehicle infotainment system and other sensor data. These devices can regularly record the vehicle's operating conditions, such as speed, position, acceleration and driving behavior (such as acceleration, braking, steering, etc.). A data collection plan is designed to ensure that the vehicle's historical behavior log is obtained in real time from various data sources. A cloud platform or a local server can be used to store these log data for subsequent analysis. Ensure that the collected log data uses a unified format (such as JSON or CSV) to facilitate subsequent data processing and analysis. Each log should contain key information such as timestamp, location coordinates, speed, and vehicle status. The historical behavior log is cleaned to remove redundant information and outliers (such as unreasonable speed or location data) to ensure data reliability and accuracy. This can be done by setting thresholds or using statistical analysis methods (such as Z-score) for anomaly detection. Select a suitable calculation method to convert the timestamp and location data in the historical behavior log into the dynamic position of the vehicle, usually using linear interpolation or Kalman filter (Kalman Filter) is used to smooth the position data and fill the gaps between the position data. The dynamic position of the vehicle at each time point is calculated by the integration method according to the timestamp and the speed information of the vehicle. This process requires converting the speed data into displacement and updating it according to the time change. The calculated multi-time point position data are integrated into a unified data structure to facilitate the subsequent trajectory reconstruction. A suitable dynamic trajectory reconstruction algorithm can be used. A path-based reconstruction method (such as the Dijkstra algorithm) or a time series-based reconstruction method (such as the dynamic time warping algorithm DTW) can be used to reconstruct a complete moving trajectory from the multi-time point position data. Based on the integrated multi-time point position data, the dynamic moving trajectory of the vehicle is generated using the selected reconstruction algorithm. The trajectory should contain the position information of each time point and be arranged in chronological order. The reconstructed trajectory is smoothed to eliminate the jitter caused by sensor noise and data calculation errors. This can be achieved through the moving average method or a more complex filtering algorithm (such as the Savitzky-Golay filter). Visualization tools (such as Matplotlib, Plotly or GIS) are used. The reconstructed dynamic trajectory is visualized by the software, and the vehicle's moving path is displayed graphically to facilitate the analysis and understanding of vehicle behavior. Further analysis can be performed through the visualized dynamic trajectory, such as identifying driving behaviors such as frequent lane changes and sudden braking, providing data support for subsequent driving behavior analysis and optimization.

[0043] Step S4: Analyze the dynamic behavior characteristics of historical vehicle dynamic trajectories, and perform personalized driving behavior fitting of the owner to construct a personalized driving profile of the owner;

[0044] In this embodiment, dynamic behavior features to be extracted are determined. These features usually include the vehicle's acceleration, speed change rate, steering angle, frequency of sudden braking and acceleration events, etc. Data analysis methods (such as principal component analysis PCA) can be used to reduce feature dimensions, thereby extracting key features, analyzing the reconstructed historical vehicle dynamic trajectory, and calculating the dynamic features in each time period. This can be done by writing a data processing script and using libraries in languages ​​such as Python or R (such as NumPy, Pandas) to perform feature calculations and calculate statistical features, such as the vehicle's average speed, maximum acceleration, and frequent steering angle range. These statistical features can reflect the owner's driving style and habits, and select a suitable driving behavior fitting model. Machine learning algorithms (such as support vector machines SVM, random forests, or neural networks) can be used to establish a personalized driving behavior model for the owner. These models can be classified and fitted based on the extracted features. The extracted dynamic behavior features are combined with corresponding driving behavior labels (such as normal driving, aggressive driving, cautious driving, etc.) to form a training data set. These labels can be obtained through historical data analysis or expert annotation. The data set is divided into a training set and a test set, usually 70%. The data is used for training, and 30% of the data is used for testing to verify the accuracy and generalization ability of the model. The selected machine learning model is trained using the training set. The model parameters are adjusted by optimizing the algorithm to improve the fitting accuracy. The cross-validation method can be used to evaluate the model performance to ensure its performance on unseen data. The trained model is evaluated on the test set, and indicators such as accuracy, recall, and F1-score are calculated to ensure that the model can effectively capture the personalized driving characteristics of the car owner. According to the output results of the fitting model, a personalized driving profile of the car owner is constructed. The driving profile usually includes the behavioral characteristics of the car owner under different driving conditions (such as acceleration habits, braking frequency, steering habits, etc.) and comparative analysis (such as comparison with similar drivers). Visualization tools (such as Tableau, Matplotlib) are used to visualize the driving characteristics of the car owner. or D3.js) to visualize the driving profile, intuitively display the owner's driving characteristics through charts, radar charts, etc., analyze the generated personalized driving profile of the owner, identify the advantages and disadvantages of driving habits. For example, if it is found that the owner has a high frequency of emergency braking, relevant safe driving suggestions or training courses can be provided, and a continuous update mechanism for the vehicle monitoring system can be established to regularly obtain new driving data. The driving profile is updated through periodic analysis and fitting to reflect the owner's latest driving behavior changes. The owner's feedback on the driving profile is collected to understand the difference between his self-perception of driving habits and the system analysis results. The feature extraction and fitting model are adjusted according to the feedback to improve the accuracy and practicality of the driving profile.

[0045] Step S5: predicting the personalized driving behavior of the owner in the real-time scene based on the personalized driving portrait of the owner, making intelligent sensor activation decisions, and building an intelligent activation strategy for the tire pressure sensor;

[0046] In this embodiment, a data interface is designed to obtain current traffic conditions, road conditions and vehicle status information from the real-time traffic scenario model. These data can be collected in real time through on-board sensors, GPS, cameras and other devices to ensure the timeliness and accuracy of the data, and select a suitable driving behavior prediction model, such as a time series-based LSTM (long short-term memory network) or a decision tree-based model. These models can predict the owner's driving behavior based on the owner's personalized driving portrait and real-time scenario data. The feature information of the real-time traffic data and the owner's personalized driving portrait is input into the prediction model. The model will analyze this information and output the driving behavior prediction results in the short future period, including acceleration, braking, steering, etc. The accuracy of the prediction results is evaluated to ensure that the model can effectively capture the owner's driving habits under specific road conditions. According to the prediction results, Based on the results, adjust the data acquisition and analysis strategies to improve the prediction accuracy, select appropriate decision algorithms, such as fuzzy logic-based decision systems or deep reinforcement learning algorithms. These algorithms can comprehensively consider the real-time driving behavior prediction results, vehicle status, and environmental factors to make reasonable sensor activation decisions. Define key decision parameters, including the vehicle's current speed, acceleration, steering angle, tire pressure status, and road condition information. These parameters will serve as inputs to the decision algorithm and affect the sensor activation strategy. Based on the real-time driving behavior prediction results and the set decision parameters, the system will execute intelligent activation decisions. Through the decision algorithm, it determines when to activate the tire pressure sensor to obtain The most accurate tire pressure data. For example, when sudden acceleration or sharp turns are predicted, the system will activate the tire pressure sensor in advance to monitor tire pressure changes and record the results of each activation decision, including the activation timing and sensor feedback data. These data will be used for subsequent decision optimization and model training to improve the intelligence level of the system. According to the results of dynamic driving behavior prediction and sensor activation decision, an intelligent activation strategy for tire pressure sensors is constructed. The strategy should clarify how to effectively activate sensors to obtain real-time tire pressure information in different driving scenarios. The constructed activation strategy is evaluated, and its effectiveness is tested using historical data and simulated environments. By comparing different strategies The accuracy and response time of tire pressure monitoring under certain conditions can be accurately identified, and the optimal strategy can be identified. According to real-time feedback and environmental changes, the activation strategy can be dynamically adjusted. For example, under different weather conditions or on different road types, the activation frequency and timing of the sensor can be optimized to ensure the real-time and accuracy of tire pressure monitoring. In the actual driving environment, the intelligent activation strategy can be implemented to monitor the activation effect and data feedback of the tire pressure sensor to ensure that the system can quickly respond to changes in driving behavior, obtain tire pressure data in real time, establish a user feedback mechanism, collect car owners' experience and suggestions on tire pressure monitoring, and further optimize the intelligent activation strategy based on user feedback to improve the user satisfaction and practicality of the system.

[0047] Step S6: Perform instant activation processing based on the tire pressure sensor intelligent activation strategy, then optimize the activation response delay and build a dynamic activation optimization engine.

[0048] In this embodiment, according to the tire pressure sensor intelligent activation strategy constructed in the previous step, the system calls the activation instruction at an appropriate time. The instruction will instruct the tire pressure sensor to activate under specific driving conditions (such as sudden braking, rapid acceleration or turning) to ensure real-time monitoring of tire pressure data. The activation signal is sent to the tire pressure sensor through the vehicle control unit. This process requires ensuring accurate signal transmission, which is usually achieved using a CAN (Controller Area Network) bus or other communication protocols. At the same time, the system status is monitored to ensure that the activation signal is successfully sent and received. The system will record the activation timestamp for subsequent response delay analysis. Once the tire pressure sensor is activated, the system starts to record the response data, which includes the timestamp of the tire pressure data returned by the sensor and the difference between the timestamp and the activation instruction. The activation response delay is calculated, and appropriate delay analysis tools and algorithms are selected, such as statistical analysis tools (such as Python SciPy in ) to analyze data and identify trends and characteristics of response delays. Timing analysis methods can be used to analyze changes in response delays under different driving conditions. Optimization strategies can be developed based on the results of delay data analysis. For example, if the response delay is high under certain conditions, the activation timing or signal transmission frequency needs to be adjusted. Algorithm optimization (such as dynamically adjusting the activation timing and increasing the signal transmission frequency) can be used to reduce activation response delays. Machine learning technology can be used to train models based on historical data to predict the best activation timing, thereby optimizing response delays. The system architecture of the dynamic activation optimization engine is designed, including data acquisition modules, data analysis modules, and activation decision modules. These modules need to be tightly integrated to achieve real-time data processing and response. The data flow path is designed to ensure that the data collected from the tire pressure sensor can be quickly transmitted to the analysis module. After delay optimization, The decision module then makes a real-time activation decision. The efficiency of data flow is the key to dynamic engine performance. Dynamic optimization algorithms are implemented to enable the engine to learn and adapt based on real-time and historical data. Reinforcement learning algorithms can be used to allow the system to continuously adjust activation strategies based on real-time feedback to adapt to different driving conditions and behaviors. The engine should have an adaptive mechanism that can automatically adjust activation strategies based on changes in the driving environment (such as road conditions, vehicle speed, etc.) to improve response speed and accuracy. The constructed dynamic activation optimization engine is fully tested to ensure its effectiveness and stability under different conditions. The engine's response delay and activation accuracy are verified through simulated driving scenarios. The engine's performance during the test is recorded to evaluate its effectiveness in real-time activation processing and delay optimization. Necessary adjustments and optimizations are made based on the test results to ensure the reliability of the engine in actual applications.

[0049] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0050] Step S11: acquiring vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array;

[0051] Step S12: performing parameter filtering optimization on the vehicle running state monitoring parameters to generate nonlinear optimized monitoring parameters;

[0052] Step S13: performing vehicle vibration time-frequency analysis on the nonlinear optimization monitoring parameters to generate vehicle vibration time-frequency characteristics;

[0053] Step S14: performing multi-scale time-frequency decomposition on the vehicle vibration time-frequency characteristics to obtain vibration frequency domain diagrams of multiple scales;

[0054] Step S15: performing frequency drift calculation on the vibration frequency domain graphs of multiple scales one by one to generate a frequency drift feature of each frequency domain graph;

[0055] Step S16: Calculate the peak frequency of periodic vibration of the vehicle based on the time-frequency characteristics of vehicle vibration to obtain the peak frequency of periodic vibration;

[0056] Step S17: mining the vehicle vibration situation of the frequency drift characteristics of each frequency domain graph based on the periodic vibration peak frequency, and constructing a vehicle vibration situation curve.

[0057] In this embodiment, a variety of suitable sensors are selected, such as acceleration sensors, GPS, vehicle speed sensors, tire pressure sensors and cameras. These sensors can comprehensively monitor the running status and driving images of the vehicle. The sensors are reasonably arranged at different parts of the vehicle to ensure that the vehicle's motion status data can be accurately collected. For example, the acceleration sensor can be arranged at the bottom of the vehicle body, and the camera should be installed at the front to obtain driving images. The sensors are used to collect real-time vehicle running status monitoring parameters, including acceleration, speed, steering angle, tire pressure, etc. At the same time, the real-time monitoring image of the vehicle driving is obtained through the camera, and the collected data is transmitted to the central processing unit in real time through a wireless or wired network to ensure the timeliness and accuracy of the data. Preprocess the collected vehicle operation status monitoring parameters, use filtering algorithms (such as Kalman filtering or mean filtering) to remove noise and outliers to ensure the smoothness and accuracy of the data, standardize the data of different sensors to eliminate the dimensional differences between sensors for subsequent analysis, select suitable nonlinear optimization algorithms, such as particle swarm optimization (PSO) or genetic algorithm (GA), optimize the filtered monitoring parameters, generate nonlinear optimized monitoring parameters through the optimization algorithm, ensure that these parameters can truly reflect the operation status of the vehicle, select suitable time-frequency analysis methods, such as short-time Fourier transform (STFT) or wavelet transform, these methods can effectively capture the time-frequency characteristics of vehicle vibration signals, and convert nonlinear The optimized monitoring parameters are input into the selected time-frequency analysis algorithm to perform time-frequency analysis of vehicle vibration and generate time-frequency characteristic diagrams of vehicle vibration. These characteristic diagrams contain the energy distribution information of vibration signals at different times and frequencies. Multi-scale time-frequency decomposition technology is selected, such as wavelet transform or empirical mode decomposition (EMD), to extract multi-scale characteristics of vibration signals. The time-frequency characteristics of vehicle vibration are decomposed at multiple scales to obtain vibration frequency domain diagrams at multiple scales. The frequency domain diagram of each scale reflects the vibration characteristics within different frequency ranges. The vibration frequency domain diagram of each scale is recorded for subsequent analysis and comparison. An appropriate frequency drift calculation method is selected, such as peak detection or centroid method, to identify the main frequency components in the frequency domain diagram. The vibration frequency domain diagram of each scale is analyzed. The dynamic frequency domain diagrams are analyzed one by one, and their frequency drift characteristics are calculated. This process involves identifying the significant frequency components in the frequency domain diagram and calculating their drift relative to the reference frequency. The frequency drift characteristics of each frequency domain diagram are recorded for subsequent mining and analysis. A suitable peak detection algorithm, such as the adaptive threshold method or the maximum value method, is selected to identify the peak frequency of periodic vibration in the vibration time-frequency characteristics. Based on the vehicle vibration time-frequency characteristics, the peak frequency is calculated. This process requires identifying the main frequency components of periodic vibrations and performing statistical analysis. The calculated peak frequency of periodic vibrations is recorded to provide basic data for subsequent dynamic potential mining. A suitable dynamic potential mining method is selected, such as based on dynamic time warping (DTW) or autoregressive model (AR).In order to analyze the relationship between the vehicle vibration mode and the frequency drift characteristics, the frequency drift characteristics of each frequency domain diagram are analyzed based on the peak frequency of the periodic vibration, and the vehicle vibration situation curve is constructed. This curve will reflect the vibration characteristics of the vehicle under different operating conditions. The final generated vehicle vibration situation curve is recorded and visualized for easy understanding and application.

[0058] In this embodiment, the parameter filtering optimization is specifically as follows:

[0059] Calculating the sampling frequency of each sensor in the multi-sensor array to generate a signal sampling frequency of each sensor;

[0060] Resampling the vehicle running status monitoring parameters according to the signal sampling frequency of each sensor to obtain time-synchronized running status monitoring parameters;

[0061] Identify abnormal parameter noise of time synchronization operation status monitoring parameters and extract monitoring parameter noise;

[0062] Perform high-frequency noise filtering on monitoring parameter noise to generate filtered noise-reduced monitoring parameters;

[0063] The nonlinear trend of the filter optimization monitoring parameters is removed to generate nonlinear optimization monitoring parameters.

[0064] In this embodiment, basic information of each sensor is collected from the multi-sensor array, including sensor type, working principle and output signal characteristics. Each sensor (such as acceleration sensor, temperature sensor, GPS, etc.) has different signal characteristics and sampling frequency requirements. According to the Nyquist theorem, the minimum sampling frequency of each sensor is determined. Generally speaking, the sampling frequency should be at least twice the highest frequency of the signal in order to accurately capture the signal characteristics. The output signal of each sensor is subjected to spectrum analysis. Commonly used tools include fast Fourier transform (FFT). The spectrum of the signal is obtained through FFT analysis, and the main frequency components and bandwidth of the signal are identified, so as to calculate the optimal sampling frequency. A suitable resampling method is selected, such as linear interpolation, spline interpolation or nearest neighbor interpolation. These methods can effectively convert data with different sampling frequencies into a unified sampling frequency. According to the calculated sampling frequency of each sensor, the corresponding vehicle operation status monitoring parameters are resampled. The data of all sensors are unified into a standard time series to achieve time synchronization. The resampled data are integrated to form a time series containing synchronized data of all sensors, which can ensure the consistency of data from different sensors in subsequent analysis. A suitable noise recognition algorithm is selected, such as threshold-based detection method, Z-score Detection method or anomaly detection method based on machine learning (such as isolation forest or support vector machine) is used to analyze the time-synchronized operating status monitoring parameters. The selected anomaly detection algorithm is used to identify abnormal parameters and noise. These anomalies are manifested as sudden signal changes or readings beyond the normal range. The identified noise and abnormal parameters are recorded for subsequent processing and analysis. Appropriate high-frequency noise filtering technology is selected, such as low-pass filter, Butterworth filter or wavelet transform. These methods can effectively remove high-frequency noise components in the signal. The monitoring parameter noise is input into the selected filter to filter the high-frequency noise. The cutoff frequency of the filter is set according to the requirements to ensure that only unnecessary high-frequency signals are removed. The filtered signal data is recorded to generate the monitoring parameters for filtering and noise reduction, laying the foundation for the next step of analysis. Appropriate detrending methods are selected, such as moving average method, difference method or polynomial fitting. These methods can effectively remove nonlinear trend components in the signal. The monitoring parameters after filtering optimization are input into the selected detrending algorithm to remove the nonlinear trend in the signal. This can be achieved by calculating the moving average of the signal or using polynomial fitting. The signal data after trend removal is recorded to form the final nonlinear optimized monitoring parameters, which will be used for subsequent analysis and decision support.

[0065] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0066] Step S21: performing road object recognition on the vehicle driving monitoring image, thereby marking a plurality of road objects;

[0067] Step S22: performing deep semantic segmentation on multiple road objects to generate semantic features of the road objects;

[0068] Step S23: performing real-time road structure element analysis on the semantic features of road objects to generate real-time road structure element data;

[0069] Step S24: Calculating the road surface roughness of the vehicle driving monitoring image based on the vehicle vibration situation curve to generate real-time road surface roughness;

[0070] Step S25: Based on the real-time road surface flatness and the real-time road surface smoothness, the real-time road condition scene structure is rendered to construct a real-time road condition scene model.

[0071] In this embodiment, the camera installed on the vehicle is used to obtain driving monitoring images in real time. These images should cover the road in front of the vehicle and the surrounding environment. A suitable target detection algorithm is selected, such as YOLO (You Only Look Once), Faster R-CNN or SSD (Single Shot MultiBox Detector). These algorithms can efficiently identify multiple targets in the image. The driving monitoring image is input into the trained target detection model for road object recognition. The model will output the category, bounding box position and confidence of each recognized object. According to the detection results, the road objects are marked on the original image, and the category and bounding box information of each object are recorded for subsequent processing. A suitable deep learning semantic segmentation model is selected, such as U-Net, DeepLab v3+ or Mask R-CNN, these models can perform pixel-level segmentation on images, preprocess the labeled images, such as normalization and data enhancement, to ensure the quality of model input, input the preprocessed images into the deep semantic segmentation model, generate category labels for each pixel, identify the precise boundaries of road objects, extract the semantic features of road objects based on the segmentation results, these features will be used for subsequent analysis and processing, select appropriate road structure element analysis methods, such as morphological analysis or analysis based on geometric features, these methods can extract the structural features of road objects, analyze the extracted semantic features of road objects, and generate real-time road structure element data, which may include information such as the size, shape, and distribution of objects, and record Record the analysis results to form road structure element data to support subsequent road condition analysis and modeling; integrate vehicle vibration situation curve data to ensure that it can reflect the vibration characteristics of the vehicle during driving; select a reasonable road surface roughness calculation method, such as the International Roughness Index (IRI) or fluctuation analysis method, which can derive road surface roughness information from vibration data; calculate road surface roughness based on the vehicle vibration situation curve and driving monitoring images; combine vibration characteristics with image data to evaluate road surface roughness; generate real-time road surface roughness data and associate it with image information for subsequent analysis and application; select appropriate real-time rendering technology, such as OpenGL in computer graphics Or DirectX, or use a game engine for scene rendering, build a road scene model based on the real-time road surface flatness and semantic features of road objects, adjust the visual performance of the scene according to the road characteristics, such as the color, texture and shape of the road surface, render the constructed scene model in real time, and generate a dynamic road scene display. This process should be able to reflect the changes in vehicle driving status and road characteristics in real time, display the rendering results on the user interface, provide a real-time road view, ensure that users can clearly see the current driving environment and road status, and record the data and results of the rendering process.Provide a basis for subsequent road condition analysis, optimization and decision support.

[0072] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:

[0073] Step S31: Obtain vehicle historical behavior log;

[0074] Step S32: Calculate the vehicle position at multiple time points on the vehicle historical behavior log, so as to obtain the vehicle position information at multiple time points;

[0075] Step S33: performing time series position change analysis on the vehicle position information at multiple time points to generate vehicle time series position change data;

[0076] Step S34: performing vehicle stop node identification on the vehicle position information at multiple time points, and extracting multiple vehicle stop nodes;

[0077] Step S35: reconstructing the dynamic vehicle movement trajectory of the vehicle time series position change data according to the multiple vehicle stop nodes, thereby constructing the historical vehicle dynamic trajectory.

[0078] In this embodiment, the source of the vehicle's historical behavior log is determined, which usually includes an on-board GPS device, an on-board computer, a vehicle management system or a mobile application. These data sources record the vehicle's operating status, location, speed, fuel consumption and other information. The historical behavior log is obtained through an API interface or a database query to ensure that the obtained data contains the vehicle's driving time, location coordinates, speed and other information for subsequent analysis. The obtained log data is formatted to ensure a unified data structure for subsequent processing. It is usually necessary to convert the data into a time series format, including fields such as timestamp, location, speed, etc., and select an appropriate location calculation method, such as linear interpolation, weighted average, etc. The average method or Kalman filter can deduce the vehicle position at multiple time points based on the historical log data. Based on the timestamps and position coordinates in the historical behavior log, the selected calculation method is used to deduce the vehicle position information at multiple time points. For discontinuous time data, the missing position information can be filled by the interpolation algorithm. The vehicle position information at each time point is recorded, including the accurate latitude and longitude coordinates and the corresponding timestamp to form a complete vehicle position data set. The appropriate time series analysis method is selected, such as time series analysis, moving average method or dynamic time warping (DTW), which is used to analyze the changing trend of vehicle position. By analyzing multiple time points The vehicle location information is used to calculate the speed, acceleration, and trajectory of the location change, and generate the vehicle time series location change data. These data will reflect the vehicle's movement pattern in different time periods. The analysis results are visualized and a chart of the location change is generated to facilitate subsequent analysis and interpretation. A suitable stop node identification algorithm is selected, such as a threshold-based stop detection method or a clustering algorithm (such as DBSCAN). These methods can effectively identify the vehicle's stop time at certain locations. The vehicle location information at multiple time points is analyzed to identify potential stop nodes. A stop time threshold is set to determine whether the vehicle's stop time at a certain location reaches the threshold to confirm the stop node and extract The identified multiple vehicle stop nodes, including the latitude and longitude coordinates of the stop location, the start and end time of the stop and other information, provide a basis for the subsequent dynamic trajectory reconstruction. Select an appropriate trajectory reconstruction method, such as linear interpolation, spline interpolation or shortest path algorithm, to ensure that the selected method can effectively connect the vehicle's moving trajectory and stop nodes. Reconstruct the vehicle's dynamic moving trajectory based on the extracted stop nodes and time-series position change data, connect the stop nodes with the vehicle's moving segments, create a complete driving trajectory, record the reconstructed historical vehicle dynamic trajectory, and visualize it to generate a trajectory map to ensure that the trajectory can clearly reflect the vehicle's driving path and stop conditions.

[0079] In this embodiment, step S4 includes the following steps:

[0080] Step S41: segmenting the historical vehicle dynamic trajectory into multiple time periods, thereby obtaining multiple dynamic trajectory segments;

[0081] Step S42: Calculating the direction change frequency of each dynamic trajectory segment, and generating the direction change frequency of each trajectory segment;

[0082] Step S43: extracting the lane change frequency according to the direction change frequency of each trajectory segment, thereby obtaining the lane change frequency of each trajectory segment;

[0083] Step S44: performing maximum transient acceleration identification on multiple dynamic trajectory segments to generate the maximum transient acceleration of each trajectory segment;

[0084] Step S45: performing emergency braking frequency analysis on multiple dynamic trajectory segments to generate emergency braking frequencies;

[0085] Step S46: performing dynamic behavior characteristic analysis on the emergency braking frequency, the maximum transient acceleration of each trajectory segment, and the lane change frequency of each trajectory segment to generate dynamic behavior characteristics of each trajectory segment;

[0086] Step S47: Fit the personalized driving behavior of the vehicle owner to the dynamic behavior characteristics of each trajectory segment to construct a personalized driving profile of the vehicle owner.

[0087] In this embodiment, historical vehicle dynamic trajectory data is collected, including timestamps, position coordinates, speed, acceleration and other information to ensure the integrity and accuracy of the data. A suitable trajectory segmentation algorithm is selected, such as segmentation based on time intervals or segmentation based on behavioral characteristics (for example, thresholds for changes in speed and acceleration). These methods can effectively identify different stages of the trajectory. According to the selected segmentation method, the historical trajectory data is segmented by time period or behavioral characteristics to generate multiple dynamic trajectory segments. Each trajectory segment should contain corresponding time, position and speed information. The start and end time, position coordinates and other information of each trajectory segment are recorded to provide basic data for subsequent analysis. A suitable direction change frequency calculation method is selected, such as a calculation method based on angle difference. The direction change of each trajectory segment is calculated using the vector method. For each trajectory segment, the frequency of its direction change is calculated. The specific steps include: identifying the number of direction changes by calculating the angle change between adjacent position points, and normalizing it to obtain the frequency, recording the direction change frequency of each trajectory segment, forming a mapping table containing trajectory segments and direction change frequencies, selecting a suitable lane change frequency calculation algorithm, such as based on the threshold judgment of the direction change, identifying the lane change by setting the angle threshold of the direction change, judging whether the lane change occurs in the trajectory segment according to the direction change frequency of each trajectory segment, calculating the number of lane changes in each trajectory segment, and normalizing it to frequency, recording the lane change frequency of each trajectory segment, for subsequent Dynamic behavior analysis provides data support and selects a suitable transient acceleration calculation method. Usually, the first-order difference method is used or data is directly obtained through an acceleration sensor. For each trajectory segment, its transient acceleration is calculated and the maximum transient acceleration in each segment is identified. This can be achieved by performing extreme value detection on the acceleration data and recording the maximum transient acceleration value of each trajectory segment to provide basic data for subsequent dynamic behavior feature analysis. A suitable emergency braking recognition algorithm is selected and a threshold value of the acceleration value (such as negative acceleration greater than a certain value) is usually set to judge emergency braking events. According to the acceleration data of each trajectory segment, the number of occurrences of emergency braking events is identified. By traversing the acceleration data of each trajectory segment, the number of times the negative acceleration exceeds the threshold is counted and recorded. The frequency of emergency braking for each trajectory segment provides data support for subsequent analysis. Select a suitable dynamic behavior feature analysis method, such as cluster analysis or principal component analysis (PCA), to comprehensively consider the frequency of emergency braking, maximum transient acceleration, and lane change frequency. The frequency of emergency braking, maximum transient acceleration, and lane change frequency are integrated into a feature vector for dynamic behavior feature analysis. Through statistical analysis, different driving behavior patterns are identified. The dynamic behavior feature data of each trajectory segment is recorded to provide a basis for personalized driving behavior fitting. Select a suitable personalized driving behavior fitting method, such as a machine learning model (such as a decision tree, random forest, or support vector machine) or a rule-based fitting method. Use the collected dynamic behavior feature data to train the machine learning model.In order to identify and fit the personalized driving behavior of the car owner, the model should be able to output the car owner's driving style (such as aggressive, gentle, stable, etc.), generate a personalized driving portrait of the car owner, record the driving behavior characteristics and style, and provide a basis for subsequent driving improvement suggestions and personalized services.

[0088] In this embodiment, step S5 includes the following steps:

[0089] Step S51: predicting the personalized driving behavior of the vehicle owner in a real-time scenario on the real-time road condition scenario model according to the personalized driving portrait of the vehicle owner, so as to obtain the vehicle operation prediction data in the real-time scenario;

[0090] Step S52: predicting the vehicle tire pressure fluctuation based on the real-time scenario vehicle operation prediction data, thereby generating the real-time scenario vehicle tire pressure fluctuation prediction data;

[0091] Step S53: calculating the pre-activation time point of the tire pressure sensor based on the real-time scenario vehicle tire pressure fluctuation prediction data, thereby obtaining the pre-activation time point of the tire pressure sensor;

[0092] Step S54: Calculate the vehicle speed based on the real-time scenario vehicle operation prediction data to obtain the real-time scenario vehicle prediction speed;

[0093] Step S55: Adaptively adjust the activation frequency of the predicted vehicle speed in the real-time scenario, thereby obtaining an adaptive activation frequency parameter of the tire pressure sensor;

[0094] Step S56: Make a sensor intelligent activation decision based on the tire pressure sensor pre-activation time point and the tire pressure sensor adaptive activation frequency parameters, and build a tire pressure sensor intelligent activation strategy.

[0095] In this embodiment, the personalized driving portrait of the car owner, including information such as driving habits, style and preferences, and the data of the real-time traffic scene model, which usually includes current traffic conditions, road conditions and weather information, selects a suitable prediction algorithm, such as long short-term memory network (LSTM), random forest or support vector machine (SVM). These models can effectively capture time series data and nonlinear relationships. The selected model is trained to identify features related to the car owner's personalized driving behavior and predict the real-time road conditions. The model will output the vehicle's operating status prediction data in the real-time scene, including the vehicle's acceleration, steering, braking and other behaviors, and record the real-time scene vehicle operation prediction data for subsequent tire pressure fluctuation prediction and transmission. Provide support for sensor activation decisions, select appropriate tire pressure fluctuation prediction models, such as time series analysis, regression analysis or machine learning models (such as random forests), to analyze the impact of vehicle operating status on tire pressure, analyze vehicle speed, acceleration and environmental factors (such as temperature, load, etc.) according to real-time scenario vehicle operation prediction data, predict tire pressure fluctuations, establish a model to associate vehicle operating parameters with tire pressure changes, output tire pressure fluctuation prediction data, record the generated real-time scenario vehicle tire pressure fluctuation prediction data, ensure data accuracy and real-time, select a suitable time point calculation method, usually a threshold-based algorithm is used to determine when to activate the tire pressure sensor, for example, a threshold is set according to the rate of change of tire pressure fluctuations, When the fluctuation exceeds the threshold, the activation is triggered. According to the tire pressure fluctuation prediction data, the time point when the tire pressure sensor needs to be pre-activated is identified. This involves detecting the critical points of fluctuation changes and recording these time points to trigger the sensor. The time point of tire pressure sensor pre-activation is recorded to prepare for subsequent real-time monitoring and data collection. The appropriate vehicle speed calculation method is selected, which is usually based on the rate of change of position data. For example, the speed is obtained by calculating the displacement and time interval of adjacent time points using the differential method. The real-time speed of the vehicle is calculated based on the real-time scenario vehicle operation prediction data. This process requires extracting the position information of continuous time points and applying the selected speed calculation method. The calculated real-time scenario vehicle prediction speed is recorded to provide the tire pressure sensor with a real-time prediction data. Provide a basis for adjusting the activation frequency of the tire pressure sensor, select a suitable adaptive activation frequency adjustment algorithm, such as PID control algorithm or fuzzy logic control. These methods can dynamically adjust the activation frequency of the tire pressure sensor according to the vehicle speed. According to the real-time scene vehicle predicted speed, calculate the adaptive activation frequency parameters of the tire pressure sensor, and dynamically adjust the activation frequency of the sensor according to the change of vehicle speed to ensure that the tire pressure data can be obtained in time when the driving conditions change. Record the adaptive activation frequency parameters to provide a basis for the subsequent intelligent activation decision of the tire pressure sensor, select a suitable intelligent activation decision algorithm, such as decision tree, rule engine or model-based prediction algorithm, to ensure that the algorithm can comprehensively consider the pre-activation time point and adaptive frequency parameters,By integrating the pre-activation time point and adaptive activation frequency parameters of the tire pressure sensor, and selecting the decision algorithm, an intelligent activation strategy for the tire pressure sensor is formulated. This strategy will ensure that the sensor is activated at the most appropriate time to obtain real-time tire pressure data, record the constructed tire pressure sensor intelligent activation strategy, provide a basis for practical application, and ensure its effectiveness and reliability in vehicle operation.

[0096] In this embodiment, step S6 includes the following steps:

[0097] Step S61: performing instant activation processing based on the tire pressure sensor intelligent activation strategy, and collecting tire pressure sensor activation response data;

[0098] Step S62: Calculating activation response delay of the tire pressure sensor activation response data to obtain activation response delay parameters;

[0099] Step S63: performing optimal activation time analysis according to the activation response delay parameter, thereby generating an optimal activation time for the sensor;

[0100] Step S64: optimizing the activation response delay of the tire pressure sensor intelligent activation strategy based on the optimal activation time of the sensor to generate a response delay optimized activation strategy;

[0101] Step S65: Perform dynamic transfer learning on the response delay optimization activation strategy to build a dynamic activation optimization engine.

[0102] In this embodiment, according to the previously constructed intelligent activation strategy for tire pressure sensors, the system will instantly activate the tire pressure sensor according to real-time data and vehicle status. This process needs to ensure that the sensor is activated at an appropriate time to obtain accurate tire pressure data. Once the sensor is activated, the system will start the data acquisition module to record the response data of the tire pressure sensor, which includes the tire pressure reading of the sensor after activation, the environmental conditions during activation (such as temperature and humidity), and the vehicle status data during activation (such as speed and acceleration). The collected activation response data is stored in a database to ensure the integrity of the data and the availability of subsequent analysis. A suitable delay calculation method is selected, usually using a timestamp difference method. By comparing the timestamp of the sensor activation instruction with the timestamp of the actually obtained tire pressure data, the response delay is calculated. The activation response data is analyzed to extract the activation instruction time and the response time, and the activation response delay parameter is calculated. The specific steps include: for each activation event, the difference between the instruction time and the response time is calculated, and the delay of each activation is recorded. The calculated activation response delay parameter is stored in the system for subsequent analysis. A suitable optimal activation time analysis algorithm is selected, such as a statistical analysis-based method or a machine learning algorithm (such as cluster analysis) to determine the delay. The minimum activation timing, using the activation response delay data, analyzes the response effects of different activation timings, identifies the optimal activation time by comparing the delay data under different conditions, that is, the time point with the minimum delay, records the generated optimal activation time of the sensor, provides a basis for subsequent optimization strategies, selects appropriate optimization methods, such as optimization algorithms based on feedback control or strategy optimization using genetic algorithms, to ensure that the activation response delay is minimized, and adjusts and optimizes the intelligent activation strategy of the tire pressure sensor according to the identified optimal activation time, including: resetting the activation timing, adjusting the sensor activation strategy according to different operating conditions and driving habits, recording the generated response delay optimization activation strategy, and ensuring its feasibility and effectiveness in practical applications, selecting appropriate transfer learning algorithms, such as domain adaptation or model fine-tuning, to ensure that the optimization strategy can adapt to different driving environments and conditions, based on previous activation strategies and delay optimization data, dynamic transfer learning is performed, and a dynamic activation optimization engine is constructed. By integrating real-time data, the activation strategy is continuously updated and optimized to meet the needs of vehicles in different scenarios, the performance of the transfer learning model is regularly verified, and adjustments are made based on actual application feedback to ensure that the dynamic activation optimization engine can continuously improve the response speed and accuracy of sensor activation.

[0103] In this embodiment, the present invention further provides an activation device for a tire pressure sensor, comprising:

[0104] The vibration situation module acquires vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array; performs vehicle vibration time-frequency analysis on the vehicle operation status monitoring parameters, performs vehicle vibration situation mining, and constructs a vehicle vibration situation curve;

[0105] The road condition scene module recognizes road objects in the vehicle driving monitoring image based on the vehicle vibration situation curve, renders the real-time road condition scene structure, and builds a real-time road condition scene model;

[0106] The vehicle dynamic trajectory module obtains the vehicle historical behavior log; calculates the vehicle position at multiple time points on the vehicle historical behavior log, and reconstructs the dynamic vehicle movement trajectory, thereby constructing the historical vehicle dynamic trajectory;

[0107] The personalized portrait module analyzes the dynamic behavior characteristics of historical vehicle dynamic trajectories, fits the personalized driving behavior of the owner, and constructs a personalized driving portrait of the owner;

[0108] The adaptive activation module predicts the personalized driving behavior of the owner in real-time scenarios based on the personalized driving portrait of the owner, makes intelligent sensor activation decisions, and builds an intelligent activation strategy for tire pressure sensors.

[0109] The activation optimization module performs instant activation processing based on the tire pressure sensor's intelligent activation strategy, and then optimizes the activation response delay to build a dynamic activation optimization engine.

[0110] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements any of the above-mentioned tire pressure sensor activation methods when executing the computer program.

[0111] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the tire pressure sensor activation method described in any one of the above items are implemented.

[0112] Those skilled in the art clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0113] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it is stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or partly or all or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media for storing program codes such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks or optical disks.

[0114] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0115] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for activating a tire pressure sensor, characterized in that: The following steps are involved: Step S1: acquiring vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array; performing vehicle vibration time-frequency analysis on the vehicle operation status monitoring parameters, and performing vehicle vibration situation mining to construct a vehicle vibration situation curve; The specific steps of step S2 are: Step S21: performing road object recognition on the vehicle driving monitoring image, thereby marking a plurality of road objects; Step S22: performing deep semantic segmentation on multiple road objects to generate semantic features of the road objects; Step S23: performing real-time road structural element analysis on the semantic features of road objects to generate real-time road structural element data; Step S24: Calculating the road surface roughness of the vehicle driving monitoring image based on the vehicle vibration situation curve to generate real-time road surface roughness; Step S25: Rendering a real-time road scene structure based on the real-time road surface flatness and real-time road structure element data to construct a real-time road scene model; Step S3: Obtain the vehicle historical behavior log; calculate the vehicle position at multiple time points on the vehicle historical behavior log, and reconstruct the dynamic vehicle movement trajectory, thereby constructing the historical vehicle dynamic trajectory; Step S4: Analyze the dynamic behavior characteristics of historical vehicle dynamic trajectories, and perform personalized driving behavior fitting of the owner to construct a personalized driving profile of the owner; Step S5: predicting the personalized driving behavior of the owner in the real-time scene based on the personalized driving portrait of the owner, making intelligent sensor activation decisions, and building an intelligent activation strategy for the tire pressure sensor; Step S6: Perform instant activation processing based on the tire pressure sensor intelligent activation strategy, then optimize the activation response delay and build a dynamic activation optimization engine.

2. The method for activating a tire pressure sensor according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: acquiring vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array; Step S12: performing parameter filtering optimization on the vehicle running state monitoring parameters to generate nonlinear optimized monitoring parameters; Step S13: performing vehicle vibration time-frequency analysis on the nonlinear optimization monitoring parameters to generate vehicle vibration time-frequency characteristics; Step S14: performing multi-scale time-frequency decomposition on the vehicle vibration time-frequency characteristics to obtain vibration frequency domain diagrams of multiple scales; Step S15: performing frequency drift calculation on the vibration frequency domain graphs of multiple scales one by one to generate a frequency drift feature of each frequency domain graph; Step S16: Calculate the peak frequency of periodic vibration of the vehicle based on the time-frequency characteristics of vehicle vibration to obtain the peak frequency of periodic vibration; Step S17: mining the vehicle vibration situation of each frequency domain graph based on the periodic vibration peak frequency, and constructing a vehicle vibration situation curve; The parameter filtering optimization is specifically as follows: Calculating the sampling frequency of each sensor in the multi-sensor array to generate a signal sampling frequency of each sensor; Resampling the vehicle running status monitoring parameters according to the signal sampling frequency of each sensor to obtain time-synchronized running status monitoring parameters; Identify abnormal parameter noise of time synchronization operation status monitoring parameters and extract monitoring parameter noise; Perform high-frequency noise filtering on monitoring parameter noise to generate filtered noise-reduced monitoring parameters; The nonlinear trend of the filter optimization monitoring parameters is removed to generate nonlinear optimization monitoring parameters.

3. The method for activating a tire pressure sensor according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: Obtain vehicle historical behavior log; Step S32: Calculate the vehicle position at multiple time points on the vehicle historical behavior log, so as to obtain the vehicle position information at multiple time points; Step S33: performing time series position change analysis on the vehicle position information at multiple time points to generate vehicle time series position change data; Step S34: performing vehicle stop node identification on the vehicle position information at multiple time points, and extracting multiple vehicle stop nodes; Step S35: reconstructing the dynamic vehicle movement trajectory of the vehicle time series position change data according to the multiple vehicle stop nodes, thereby constructing the historical vehicle dynamic trajectory.

4. The method for activating a tire pressure sensor according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: segmenting the historical vehicle dynamic trajectory into multiple time periods, thereby obtaining multiple dynamic trajectory segments; Step S42: Calculating the direction change frequency of each dynamic trajectory segment, and generating the direction change frequency of each trajectory segment; Step S43: extracting the lane change frequency according to the direction change frequency of each trajectory segment, thereby obtaining the lane change frequency of each trajectory segment; Step S44: performing maximum transient acceleration identification on multiple dynamic trajectory segments to generate the maximum transient acceleration of each trajectory segment; Step S45: performing emergency braking frequency analysis on multiple dynamic trajectory segments to generate emergency braking frequencies; Step S46: performing dynamic behavior characteristic analysis on the emergency braking frequency, the maximum transient acceleration of each trajectory segment, and the lane change frequency of each trajectory segment to generate dynamic behavior characteristics of each trajectory segment; Step S47: Fit the personalized driving behavior of the vehicle owner to the dynamic behavior characteristics of each trajectory segment to construct a personalized driving profile of the vehicle owner.

5. The method for activating a tire pressure sensor according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: predicting the personalized driving behavior of the vehicle owner in a real-time scenario on the real-time road condition scenario model according to the personalized driving portrait of the vehicle owner, so as to obtain the vehicle operation prediction data in the real-time scenario; Step S52: predicting the vehicle tire pressure fluctuation based on the real-time scenario vehicle operation prediction data, thereby generating the real-time scenario vehicle tire pressure fluctuation prediction data; Step S53: calculating the pre-activation time point of the tire pressure sensor based on the real-time scenario vehicle tire pressure fluctuation prediction data, thereby obtaining the pre-activation time point of the tire pressure sensor; Step S54: Calculate the vehicle speed based on the real-time scenario vehicle operation prediction data to obtain the real-time scenario vehicle prediction speed; Step S55: Adaptively adjust the activation frequency of the predicted vehicle speed in the real-time scenario, thereby obtaining an adaptive activation frequency parameter of the tire pressure sensor; Step S56: Make a sensor intelligent activation decision based on the tire pressure sensor pre-activation time point and the tire pressure sensor adaptive activation frequency parameters, and build a tire pressure sensor intelligent activation strategy.

6. The method for activating a tire pressure sensor according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: performing instant activation processing based on the tire pressure sensor intelligent activation strategy, and collecting tire pressure sensor activation response data; Step S62: Calculating activation response delay of the tire pressure sensor activation response data to obtain activation response delay parameters; Step S63: performing optimal activation time analysis according to the activation response delay parameter, thereby generating an optimal activation time for the sensor; Step S64: optimizing the activation response delay of the tire pressure sensor intelligent activation strategy based on the optimal activation time of the sensor to generate a response delay optimized activation strategy; Step S65: Perform dynamic transfer learning on the response delay optimization activation strategy to build a dynamic activation optimization engine.

7. A tire pressure sensor activation device, characterized in that: The method for activating the tire pressure sensor according to claim 1 comprises: The vibration situation module acquires vehicle operation status monitoring parameters and vehicle driving monitoring images in real time based on a multi-sensor array; performs vehicle vibration time-frequency analysis on the vehicle operation status monitoring parameters, performs vehicle vibration situation mining, and constructs a vehicle vibration situation curve; The road condition scene module recognizes road objects in vehicle driving monitoring images, calculates the road surface smoothness of vehicle driving monitoring images based on the vehicle vibration situation curve, and builds a real-time road condition scene model; The vehicle dynamic trajectory module obtains the vehicle historical behavior log; calculates the vehicle position at multiple time points on the vehicle historical behavior log, and reconstructs the dynamic vehicle movement trajectory, thereby constructing the historical vehicle dynamic trajectory; The personalized portrait module analyzes the dynamic behavior characteristics of historical vehicle dynamic trajectories, fits the owner's personalized driving behavior, and constructs a personalized driving portrait of the owner; The adaptive activation module predicts the personalized driving behavior of the owner in real-time scenarios based on the personalized driving portrait of the owner, makes intelligent sensor activation decisions, and builds an intelligent activation strategy for tire pressure sensors. The activation optimization module performs instant activation processing based on the tire pressure sensor's intelligent activation strategy, and then optimizes the activation response delay to build a dynamic activation optimization engine.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the tire pressure sensor activation method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the tire pressure sensor activation method according to any one of claims 1 to 6 are implemented.

Images