Automobile display equipment data monitoring system and method based on intelligent control

Abstract

The invention discloses an automobile display equipment data monitoring system and method based on intelligent control, and relates to the technical field of automobile intelligent control and data monitoring. The system collects various types of data of an automobile body sensor, a camera and the like through a multi-source data acquisition module, and the data are processed through a data preprocessing module; the driver score calculation module and the vehicle environment score calculation module respectively obtain corresponding scores, and then initial vehicle speed and acceleration threshold values are generated in the basic threshold value calculation module and are adjusted in real time by means of the dynamic threshold value correction module. The holographic warning trigger module carries out graded warning according to comparison between real-time data and a threshold value, and the reinforcement learning optimization module optimizes a threshold value adjustment strategy by using a deep Q network. According to the invention, the running state of the vehicle can be comprehensively monitored, intelligent dynamic threshold setting and correction are realized, an alarm is given in time, and the driving safety and the intelligent level are effectively improved.

Description

Technical Field

[0002] The present invention relates to the technical field of automotive intelligent control and data monitoring, and specifically to a data monitoring system and method for automotive display devices based on intelligent control. Background Art

[0004] With the rapid development of the automotive industry and the continuous improvement of people's requirements for driving safety and comfort, the importance of the data monitoring system for automotive display devices has become increasingly prominent. Traditional automotive data monitoring systems have many limitations. In terms of data collection, the collection sources are single, and it is difficult to comprehensively obtain various types of information on vehicle operation. For example, relying only on some body sensors and not fully integrating camera, navigation, and CAN bus data results in serious deficiencies in perceiving the vehicle state and external environment. During the data processing process, there is a lack of effective time alignment and outlier filtering mechanisms. The time bases of different sensor data are chaotic, and there are abnormal data interfering with subsequent analysis.

[0005] For driver state assessment, only simply referring to driving mileage cannot comprehensively consider key factors such as driving stability and compliance with rules, and it is difficult to accurately reflect the true proficiency of the driver. Vehicle state monitoring only focuses on some engine parameters, ignoring the impact of brake system performance and external environmental factors on vehicle operation, and it is impossible to obtain a comprehensive and accurate comprehensive score of the vehicle environment. In terms of threshold setting, the thresholds are fixed or the adjustment methods are simple, and it is impossible to dynamically optimize according to driver individual differences, real-time road conditions, and vehicle states, making it difficult to ensure driving safety. When the vehicle speed or acceleration is abnormal, the warning method is single and not timely, and it is impossible to classify warnings according to the degree of overlimit and link with the safety system, making it difficult to effectively avoid danger; these problems seriously restrict the safety and intelligent level of automotive driving.

[0006] To solve the above problems, the present invention provides a data monitoring system and method for automotive display devices based on intelligent control. Summary of the Invention

[0008] The purpose of the present invention is to provide a data monitoring system and method for automotive display devices based on intelligent control to solve the problems raised in the prior art.

[0009] To achieve the above purpose, the present invention provides the following technical solutions:

[0010] A data monitoring system for automotive display devices based on intelligent control includes a multi-source data collection module, a data preprocessing module, a driver score calculation module, a vehicle score calculation module, a basic threshold calculation module, a dynamic threshold correction module, a holographic warning trigger module, and a reinforcement learning optimization module;

[0011] The multi-source data acquisition module is responsible for real-time acquisition of vehicle body sensor data, camera data, navigation data, CAN bus data, and user operation records; the data preprocessing module is responsible for time alignment, outlier filtering, and feature extraction of the collected raw data; the driver score calculation module calculates the driver proficiency score based on cumulative mileage, speed fluctuations, and violation records; the vehicle environment score calculation module calculates the comprehensive vehicle environment score by combining engine status, meteorology, and road type; the basic threshold calculation module generates initial vehicle speed and acceleration thresholds based on navigation speed limits, driver scores, and vehicle environment scores; the dynamic threshold correction module dynamically adjusts the thresholds through real-time feedback, scene recognition, and user habit learning; the holographic warning trigger module triggers a three-level holographic projection warning based on the comparison of real-time data with the thresholds; the reinforcement learning optimization module optimizes the threshold adjustment strategy through a deep Q network based on driving operations, vehicle status, threshold settings, and warning records.

[0012] The multi-source data acquisition module includes a sensor data acquisition unit, a camera data acquisition unit, a navigation data acquisition unit, a CAN bus data acquisition unit, and a user operation record acquisition unit;

[0013] The sensor data acquisition unit is responsible for collecting data from vehicle body sensors, engine operation parameter sensors, and brake system performance parameter sensors; for the vehicle speed sensor, acceleration sensor, and steering angle sensor on the vehicle body, it continuously collects real-time vehicle motion data, converts the vehicle's physical motion state into an electrical signal, and transmits it to the data acquisition module through the in-vehicle network; for the engine operation parameter sensors, including but not limited to the rotational speed sensor, water temperature sensor, and fuel consumption sensor, it real-time collects engine-related data and also transmits it to the data acquisition module through the in-vehicle network for evaluating the engine status; the brake system performance parameter sensors, including the brake pressure sensor and the brake pad wear sensor for obtaining qualitative information on whether the brake pad wear is within the preset normal threshold, collect brake system data and transmit it to the data acquisition module to provide a basis for vehicle braking performance evaluation;

[0014] The camera data acquisition unit uses cameras installed in the front, rear, and on both sides of the vehicle to capture images in real-time using image recognition algorithms; the cameras transmit the image data to the data acquisition module for identifying traffic signs, lane lines, and surrounding vehicles and pedestrians;

[0015] The navigation data acquisition unit obtains the vehicle position, driving direction, and planned route information provided by the navigation system in real-time through an interface with the vehicle navigation system, providing a data basis for subsequent scene recognition and threshold calculation;

[0016] The CAN bus data acquisition unit collects the engine operation parameters and brake system performance parameter data on the CAN bus connected to each electronic control unit of the vehicle through the CAN bus interface, providing a data basis for subsequent vehicle status scoring and threshold calculation.

[0017] The user operation record acquisition unit records the operations of the user manually adjusting the vehicle speed threshold and acceleration threshold during driving, and simultaneously records the driving scenario information at the time of operation, including but not limited to vehicle speed, road conditions, and weather, for subsequent user habit learning.

[0018] The data preprocessing module includes a time alignment unit, an outlier filtering unit, and a feature extraction unit;

[0019] The time alignment unit is responsible for handling the differences in data acquisition frequencies and time bases among different sensors and data sources; by constructing a timestamp mapping table, various types of data are unified into a time series based on the vehicle startup time, ensuring the accurate association of different data at the same time point and providing a unified time reference for subsequent analysis;

[0020] The outlier filtering unit calculates the mean and standard deviation for each data type according to the 3σ principle; when the deviation of a data point from the mean exceeds 3 times the standard deviation, it is determined as an outlier and removed, thereby ensuring the reliability and accuracy of the data;

[0021] The feature extraction unit is responsible for extracting features related to subsequent scoring calculation and threshold determination from the original data, including driving stability, steering angle change rate, and engine state; for driving stability, using the calculus method, by taking the derivative of the function of speed changing with mileage, the change trend of speed per unit mileage is obtained, and the speed change trend is quantified by collecting discrete speed and mileage data and calculating the absolute value of the change rate of speed per unit mileage using the central difference method; for the steering angle change rate, using the collected discrete steering angle and mileage data, the absolute value of the change rate of the steering angle per unit mileage is calculated using the central difference method; for the engine state, the engine load rate is calculated by measuring the engine torque and speed to obtain the actual output power, and then dividing it by the rated power; the air-fuel ratio is obtained by measuring the air and fuel flow rates, combining their respective densities to calculate the air quality and fuel quality, and then dividing the two; different engines have different ideal ranges and will actually fluctuate with the working conditions; the extraction of these features provides a data basis for the subsequent driver scoring calculation module, vehicle environment scoring calculation module, and basic threshold calculation module, helping the system evaluate the vehicle operation state and set thresholds.

[0022] The driver scoring calculation module includes a driving experience scoring unit, a driving stability scoring unit, and a rule compliance scoring unit;

[0023] The driving experience scoring unit is responsible for calculating the driving experience score based on the cumulative driving mileage; by using the logarithmic function relationship, the cumulative driving mileage data of the driver is converted into the corresponding score; the formula is as follows:

[0024] ;

[0025] The driving stability scoring unit uses the speed and steering angle data per unit mileage provided by the data preprocessing module to calculate the standard deviation of speed and the standard deviation of steering angle, and quantifies the fluctuation degree of speed and steering angle as the stability score; the calculation formula is as follows:

[0026] ;

[0027] The rule compliance scoring unit calculates the rule compliance score according to the ratio of the number of traffic rule violations to the annual driving mileage; its calculation formula is as follows:

[0028] ,

[0029] By correlating the number of violations with the annual driving mileage, it reflects the degree to which the driver complies with traffic rules during the historical driving process;

[0030] Finally, the above three scoring factors are weighted and averaged to obtain the driver proficiency score.

[0031] The vehicle environment scoring calculation module includes a vehicle status scoring unit and an environment status scoring unit;

[0032] The vehicle status scoring unit receives the engine operation parameters and brake system performance parameter data collected by the CAN bus interface, quantifies and evaluates these data, and comprehensively reflects the mechanical performance and operation status of the vehicle;

[0033] The specific method is: score the operation parameters of the engine and the brake system respectively, assign weights according to the importance of each part through the data transmitted by the CAN bus, and calculate the comprehensive score of the vehicle status by using the weighted average method; the formula for the engine status score E is as follows:

[0034]

[0035] Where a is the weight of the engine speed, b is the weight of the engine water temperature, and c is the weight of the engine fuel consumption. The values of a, b, and c range from 0 to 1, and a + b + c = 1; R is the score of the engine speed, which is determined according to the deviation degree of the actual engine speed from the standard speed range. The scoring standard is within the interval [1 - 10]. When the actual speed is within the standard range, R can be rated 8 - 10 points. If it is lower than the lower limit of the standard range, 1 point will be deducted for every 100 revolutions lower. If it is higher than the upper limit of the standard range, 1 point will be deducted for every 100 revolutions higher; T is the score of the engine water temperature, which is evaluated according to the difference between the actual engine water temperature and the normal water temperature range. The scoring standard is within the interval [1 - 10]. When the water temperature is within this interval, T can be rated 8 - 10 points. If the water temperature is lower than 80°C, 1 point will be deducted for every 2°C lower. If it is higher than 95°C, 1 point will be deducted for every 2°C higher; F is the score of the engine fuel consumption, which is obtained based on the actual fuel consumption and the standard fuel consumption. The standard fuel consumption is obtained from the theoretical fuel consumption data in the vehicle operation manual. The scoring standard is within the interval [1 - 10]; when the actual fuel consumption is within the range of 90% - 110% of the standard fuel consumption, F can be rated 8 - 10 points; if the actual fuel consumption is higher than 110% of the standard fuel consumption, 1 point will be deducted for every 10% higher; if it is lower than 90% of the standard fuel consumption, 1 point will be added for every 10% lower, and the maximum score is 10 points.

[0036] The formula for the braking system performance score B is as follows:

[0037]

[0038] Where d is the weight of the braking pressure and e is the weight of the brake pad wear, and d + e = 1; P is the score of the braking pressure, which is obtained by comparing the actual braking pressure acquired by the braking pressure sensor with the standard braking pressure range. The scoring standard is within the interval [1 - 10]. When the actual braking pressure is within the standard range, P can be rated 8 - 10 points. If it is lower than the standard lower limit, 1 point will be deducted for every 1 MPa lower. If it is higher than the standard upper limit, 1 point will be deducted for every 1 MPa higher; W is the score of the brake pad wear, which is obtained according to the wear degree feedback by the brake pad wear sensor. The scoring standard is within the interval [1 - 10]. The wear range of the brake pad is 20% - 80% of its total thickness; when the wear degree is within this range, W can be rated 8 - 10 points; if the wear degree is lower than 20%, 1 point will be deducted for every 5% lower. If it is higher than 80%, 1 point will be deducted for every 5% higher;

[0039] The formula for the vehicle comprehensive status score V is as follows:

[0040]

[0041] Where f is the weight of the engine status score and g is the weight of the braking system performance score;

[0042] The environmental status scoring unit quantitatively evaluates the external environmental conditions of the vehicle through the weather, road and traffic data provided by the vehicle computer system; providing data reference for dynamically adjusting the vehicle operation threshold and ensuring driving safety;

[0043] The specific method is as follows: the weather, road and traffic flow are scored respectively, weights are assigned according to the degree of influence of each factor on driving safety and efficiency, and the comprehensive score of the environmental status is calculated by the weighted average method; the weather condition score Ws determines the current weather type according to meteorological data, and each weather type corresponds to a fixed score, with sunny days scoring 10 points, cloudy days scoring 8 points, light rainy days scoring 6 points, heavy rain and rainstorms scoring 4 points, and snowy days and icy days scoring 2 points; the road type score Rs obtains road type information from the navigation system, and different road types are scored according to the specific display of the navigation system, with a score range of [1-10]; the traffic flow score Ts obtains the traffic volume level through the navigation system, and each level corresponds to a fixed score, with a score range of [1-10]; the level is divided according to the number of vehicles passing through a certain section of road in a unit time, and low traffic is less than 500 vehicles per hour, scoring 10 points, medium traffic is 500-1500 vehicles / hour, scoring 8 points, and high traffic is 1500-3000 vehicles / hour, scoring 6 points. points; congestion flow is greater than 3000 vehicles / hour and is scored 4 points; based on the above data, the formula for the comprehensive environmental status score En is as follows:

[0044]

[0045] Among them, h is the weight of weather conditions, j is the weight of traffic flow, i is the weight of road type, and h+i+j=1.

[0046] The basic threshold calculation module includes a vehicle speed threshold calculation unit and an acceleration threshold calculation unit;

[0047] The vehicle speed threshold calculation unit determines the vehicle speed threshold according to the navigation speed limit, the driver's proficiency and the vehicle's environment. The calculation formula is as follows:

[0048] ;

[0049] The navigation speed limit is provided by the vehicle's navigation system based on the legal speed limit information of the road section, and it is the basic reference value for calculating the vehicle speed threshold.

[0050] The acceleration threshold calculation unit determines an acceleration threshold suitable for the current driver level by combining the vehicle acceleration reference value and the driver proficiency score, and dynamically adjusts it according to the driver's proficiency.

[0051] The calculation formula is as follows:

[0052] ;

[0053] The vehicle acceleration reference value is a basic acceleration reference value set by the vehicle manufacturer according to the design performance and safety standards of the vehicle. The calculation method of the driver proficiency score is the same as that in the vehicle speed threshold calculation unit.

[0054] The dynamic threshold correction module includes a real-time feedback correction unit, a scene recognition correction unit, and a user habit learning correction unit;

[0055] The real-time feedback correction unit dynamically adjusts the initial threshold output by the basic threshold calculation module according to the real-time changes of the vehicle speed and acceleration. The specific method is as follows: The system continuously collects vehicle speed and acceleration data at a set frequency; calculates the ratio of the difference in speed between two adjacent sampling times to the sampling time interval; calculates the ratio of the difference in acceleration between two adjacent sampling times to the sampling time interval; when the speed change rate exceeds the preset range, adjusts the vehicle speed threshold and acceleration threshold according to the preset adjustment rules;

[0056] The scene recognition correction unit identifies the specific scene where the vehicle is located by fusing the data of the navigation system and the camera, and adjusts the vehicle speed and acceleration thresholds. The specific method is as follows: The navigation system continuously outputs the vehicle's position, driving direction, and planned route information. The camera uses image recognition algorithms to process the captured images of the vehicle's surroundings to identify traffic signs, lane lines, and surrounding objects; then fuses and analyzes the navigation information and the camera image recognition results, and adjusts the vehicle speed threshold and acceleration threshold according to the preset scene threshold adjustment rules;

[0057] The user habit learning correction unit records the operations of the user manually adjusting the threshold and the corresponding driving scene and behavior data, and automatically learns the user's habits by analyzing the impact of the user's multiple similar operations in the same scene on driving, so as to automatically adjust the threshold in the subsequent same scene;

[0058] The specific method is as follows: When the system detects each operation of the user manually adjusting the threshold, it records the operation type, adjustment amplitude, current driving scene data, and the user's driving behavior data; when the system detects that in the same scene, the same scene refers to the judgment of the same scene based on the exact matching algorithm of the driving scene data; the system automatically adjusts the threshold according to the user habit pattern analyzed previously.

[0059] The holographic warning trigger module includes a first-level warning trigger unit, a second-level warning trigger unit, and a third-level warning trigger unit;

[0060] The first-level warning trigger unit triggers the flashing of a light blue icon at the first level and a voice prompt when the vehicle speed or acceleration reaches 90% of the threshold; it is implemented through the in-vehicle display device and the voice broadcast system to remind the driver to pay attention to controlling the speed or acceleration.

[0061] The second-level warning trigger unit triggers the display of the difference of a yellow flashing triangle icon at the second level and a voice prompt when the vehicle speed or acceleration exceeds the threshold; it shows the difference between the vehicle speed or acceleration and the threshold on the in-vehicle display device and reminds the driver through voice broadcast that the threshold has been exceeded.

[0062] The third-level warning trigger unit triggers a third-level full-screen red flashing alarm and links to the safety system when the vehicle speed exceeds the threshold by 20% or the acceleration exceeds the threshold by 30%.

[0063] The reinforcement learning optimization module includes a state definition unit, an action definition unit, and a model training and updating unit.

[0064] The state definition unit collects the driver's acceleration, deceleration, and steering operation data in real time, obtains vehicle and environmental state information such as vehicle speed, acceleration, and vehicle environment score, records the current vehicle speed and acceleration threshold settings, tracks the most recently triggered warning level, and combines this information into a state vector.

[0065] The action definition unit assigns operation options for the system to adjust the threshold in different driving scenarios according to the predefined actions of increasing, decreasing, or maintaining the vehicle speed or acceleration threshold by 5%, providing an action space for model optimization; in the predefined settings, there are three actions for the vehicle speed threshold: increasing by 5%, decreasing by 5%, and remaining unchanged, and the same actions of increasing by 5%, decreasing by 5%, and remaining unchanged are set for the acceleration threshold for the system to select and execute based on the current state.

[0066] The specific method of the model training and updating unit is as follows: First, build a deep Q network. The number of neurons in its input layer is the same as the dimension of the state vector for inputting the state vector. The number of neurons in the output layer is the same as the size of the action space, and each neuron corresponds to the Q value of an action. The middle layer uses the ReLU non-linear activation function to extract the features of the state vector to enhance the expression ability; during the operation of the system, continuously collect data on states, actions, and the next state and store them in the experience replay buffer; randomly select samples from the buffer at preset time intervals. For each sample, use the current deep Q network to predict the Q value of the selected action in the current state. According to the simplified rule, the Q value of the next state is predicted by the target network, and then calculate the difference between the predicted Q value and the target Q value through a loss function such as the mean square error to obtain the loss value; then use the backpropagation algorithm to calculate the gradient of the loss function with respect to the network parameters, and use the stochastic gradient descent optimization algorithm to update the parameters according to the gradient to minimize the loss, and continuously repeat this process to improve the performance of the model.

[0067] A method for monitoring data of an automotive display device based on intelligent control, comprising the following steps:

[0068] S1. Using the vehicle body, camera, and navigation device, collect multi-dimensional data such as vehicle movement, environment, and driving operations in real time, and transmit it to the data acquisition module;

[0069] S2. Align the collected raw data in time, filter out outliers, and extract key features for subsequent analysis; calculate the driver proficiency score based on the cumulative mileage, driving stability, and compliance with rules; at the same time, calculate the comprehensive vehicle environment score in combination with the engine state, meteorology, road type, etc.

[0070] S3. Generate initial vehicle speed and acceleration thresholds based on the navigation speed limit, driver score, and vehicle environment score; dynamically adjust the basic thresholds by means of real-time feedback of vehicle speed and acceleration changes, identification of specific scenarios, and learning of user habits;

[0071] S4. Compare the real-time data with the thresholds, and trigger a three-level holographic projection warning according to the degree of over-limit to ensure driving safety;

[0072] S5. According to the real-time collected driver operations, vehicle and environmental states, current threshold settings, and the latest warning level information, combine them into a state vector, and pre-define actions to increase, decrease, or maintain the vehicle speed or acceleration threshold by 5%;

[0073] S6. Finally, build a model with a deep Q network, input the state vector to output the action Q value, continuously collect state and action data, randomly select samples at preset time intervals, calculate the loss function based on the predicted Q value, and adjust the neural network parameters through the backpropagation algorithm to optimize the model performance.

[0074] Compared with the prior art, the beneficial effects of the present invention are:

[0075] 1. Reasonable driver and vehicle environment scores: The driver score calculation module calculates the driver proficiency score by comprehensively considering the cumulative mileage, driving stability, and compliance with rules, scientifically and objectively reflecting the true level of the driver. The vehicle environment score calculation module calculates the comprehensive vehicle environment score by combining multiple factors such as the engine state, meteorology, and road type, comprehensively evaluating the vehicle operation state and external environment conditions, and providing a scientific basis for threshold setting and driving decisions.

[0076] 2. Intelligent and Dynamic Threshold Setting and Correction: The basic threshold calculation module generates initial vehicle speed and acceleration thresholds based on navigation speed limits, driver ratings, and vehicle environment ratings, fully considering various key factors to make the thresholds more in line with actual driving needs. The dynamic threshold correction module dynamically adjusts the thresholds by real-time feedback of vehicle speed and acceleration changes, identification of specific scenarios, and learning of user habits, enabling flexible adaptation to different driving scenarios and driver habits, and enhancing driving safety and comfort.

[0077] 3. Timely and Effective Warning and Safety Guarantee: The holographic warning trigger module compares real-time data with the thresholds and triggers three-level holographic projection warnings according to the degree of over-limit, from the first-level light blue icon flashing and voice prompt to the second-level yellow flashing triangle icon showing the difference and voice prompt, and then to the third-level red full-screen flashing alarm and linkage with the safety system, realizing hierarchical warnings, timely reminding the driver and taking safety measures, effectively avoiding danger, and enhancing driving safety guarantee. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 It is a schematic diagram of the system modules of a vehicle display device data monitoring system based on intelligent control according to the present invention;

[0080] Figure 2 It is a schematic diagram of the method flow of a vehicle display device data monitoring method based on intelligent control according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0082] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0083] Embodiment: As Figure 1 - Figure 2 shown, the present invention provides a technical solution, a vehicle display device data monitoring system based on intelligent control, which is characterized in that it includes a multi-source data acquisition module, a data preprocessing module, a driver rating calculation module, a vehicle rating calculation module, a basic threshold calculation module, a dynamic threshold correction module, a holographic warning trigger module, and a reinforcement learning optimization module;

[0084] The multi-source data acquisition module is responsible for real-time acquisition of vehicle body sensor data, camera data, navigation data, CAN bus data, and user operation records; the data preprocessing module is responsible for time alignment, outlier filtering, and feature extraction of the collected raw data; the driver score calculation module calculates the driver proficiency score based on cumulative mileage, speed fluctuations, and violation records; the vehicle environment score calculation module calculates the comprehensive vehicle environment score by combining engine status, meteorology, and road type; the basic threshold calculation module generates initial vehicle speed and acceleration thresholds based on navigation speed limits, driver scores, and vehicle environment scores; the dynamic threshold correction module dynamically adjusts the thresholds through real-time feedback, scenario recognition, and user habit learning; the holographic warning trigger module triggers a three-level holographic projection warning based on the comparison of real-time data with the thresholds; the reinforcement learning optimization module optimizes the threshold adjustment strategy through a deep Q network based on driving operations, vehicle status, threshold settings, and warning records.

[0085] The multi-source data acquisition module includes a sensor data acquisition unit, a camera data acquisition unit, a navigation data acquisition unit, a CAN bus data acquisition unit, and a user operation record acquisition unit;

[0086] The sensor data acquisition unit is responsible for collecting data from vehicle body sensors, engine operation parameter sensors, and brake system performance parameter sensors; for the vehicle speed sensor, acceleration sensor, and steering angle sensor on the vehicle body, it continuously collects real-time vehicle motion data, converts the vehicle's physical motion state into an electrical signal, and transmits it to the data acquisition module through the in-vehicle network; for the engine operation parameter sensors, including but not limited to the rotational speed sensor, water temperature sensor, and fuel consumption sensor, it real-time collects engine-related data and also transmits it to the data acquisition module through the in-vehicle network for evaluating the engine status; the brake system performance parameter sensors, including the brake pressure sensor and the brake pad wear sensor for obtaining qualitative information on whether the brake pad wear is within the preset normal threshold, collect brake system data and transmit it to the data acquisition module to provide a basis for vehicle braking performance evaluation;

[0087] The camera data acquisition unit uses cameras installed in the front, rear, and on both sides of the vehicle to capture images in real time using image recognition algorithms; the cameras transmit the image data to the data acquisition module for identifying traffic signs, lane lines, and surrounding vehicles and pedestrians;

[0088] The navigation data acquisition unit obtains the vehicle position, driving direction, and planned route information provided by the navigation system in real time through an interface with the vehicle navigation system, providing a data basis for subsequent scenario recognition and threshold calculation;

[0089] The CAN bus data acquisition unit collects the engine operation parameters and brake system performance parameter data on the CAN bus connected to each electronic control unit of the vehicle through the CAN bus interface, providing a data basis for subsequent vehicle state scoring and threshold calculation.

[0090] The user operation record acquisition unit records the operations of the user manually adjusting the vehicle speed threshold and acceleration threshold during driving, and simultaneously records the driving scenario information at the time of operation, including but not limited to vehicle speed, road conditions, and weather, for subsequent user habit learning.

[0091] The data preprocessing module includes a time alignment unit, an outlier filtering unit, and a feature extraction unit;

[0092] The time alignment unit is responsible for handling the differences in data acquisition frequencies and time bases among different sensors and data sources; by constructing a timestamp mapping table, various types of data are unified into a time series based on the vehicle startup time, ensuring the accurate association of different data at the same time point and providing a unified time reference for subsequent analysis;

[0093] The outlier filtering unit calculates the mean and standard deviation for each data type according to the 3σ principle; when the deviation of a data point from the mean exceeds 3 times the standard deviation, it is determined as an outlier and removed, thereby ensuring the reliability and accuracy of the data;

[0094] The feature extraction unit is responsible for extracting features related to subsequent scoring calculation and threshold determination from the original data, including driving stability, steering angle change rate, and engine state; for driving stability, using the calculus method, by taking the derivative of the function of speed changing with mileage, the change trend of speed per unit mileage is obtained, and the speed change trend is quantified by collecting discrete data of speed and mileage and calculating the absolute value of the change rate of speed per unit mileage using the central difference method; for the steering angle change rate, using the collected discrete data of steering angle and mileage, the absolute value of the change rate of steering angle per unit mileage is calculated using the central difference method; for the engine state, the engine load rate is calculated by measuring the engine torque and speed to calculate the actual output power, and then dividing it by the rated power; the air-fuel ratio is obtained by measuring the air and fuel flow rates, combining their respective densities to calculate the air quality and fuel quality, and then dividing the two; different engines have different ideal ranges and will actually fluctuate with the working conditions; the extraction of these features provides a data basis for the subsequent driver scoring calculation module, vehicle environment scoring calculation module, and basic threshold calculation module, helping the system evaluate the vehicle operation state and set thresholds.

[0095] The driver scoring calculation module includes a driving experience scoring unit, a driving stability scoring unit, and a rule compliance scoring unit;

[0096] The driving experience scoring unit is responsible for calculating the driving experience score based on the cumulative driving mileage; by using a logarithmic function relationship, the formula is as follows:

[0097] ,

[0098] Convert the cumulative driving mileage data of the driver into the corresponding score; this function form makes the growth rate of the score gradually slow down as the driving mileage increases, conforming to the actual experience growth law and reflecting the influence of driving experience on the overall score;

[0099] The driving stability scoring unit uses the speed and steering angle data per unit mileage provided by the data preprocessing module to calculate the standard deviation of speed and the standard deviation of steering angle; then according to the formula: ,

[0100] Quantify the fluctuation degree of speed and steering angle into the stability score; the smoother the changes in speed and steering angle, the smaller the corresponding standard deviation, and the higher the final stability score obtained, so as to evaluate the operation stability of the driver during driving;

[0101] The rule compliance scoring unit calculates the rule compliance score according to the ratio of the number of traffic rule violations to the annual driving mileage; its calculation formula is as follows:

[0102] ,

[0103] By correlating the number of violations with the annual driving mileage, it reflects the degree to which the driver complies with traffic rules during long-term driving; the lower the proportion of the number of violations, the higher the rule compliance score, reflecting the driver's compliance with traffic rules.

[0104] The vehicle environment scoring calculation module includes a vehicle status scoring unit and an environment status scoring unit;

[0105] The vehicle status scoring unit receives the engine operation parameters and brake system performance parameter data collected by the CAN bus interface, quantifies and evaluates these data, comprehensively reflects the mechanical performance and operation status of the vehicle, and provides basic data for subsequent determination of the overall vehicle operation parameters;

[0106] The specific method is: score the operation parameters of the engine and the brake system respectively, assign weights according to the importance of each part through the data transmitted by the CAN bus, and calculate the comprehensive score of the vehicle status by using the weighted average method; the formula for the engine status score E is as follows:

[0107]

[0108] Where a is the weight of the engine speed, b is the weight of the engine water temperature, and c is the weight of the engine fuel consumption. The values of a, b, and c range from 0 to 1, and a + b + c = 1; R is the score of the engine speed, which is determined according to the deviation degree of the actual engine speed from the standard speed range. The scoring standard is within the interval [1 - 10]. When the actual speed is within the standard range, R can be rated 8 - 10 points. If it is lower than the lower limit of the standard range, 1 point will be deducted for every 100 revolutions lower; if it is higher than the upper limit of the standard range, 1 point will be deducted for every 100 revolutions higher; T is the score of the engine water temperature, which is evaluated according to the difference between the actual engine water temperature and the normal water temperature range. The scoring standard is within the interval [1 - 10]. When the water temperature is within this interval, T can be rated 8 - 10 points. If the water temperature is lower than 80°C, 1 point will be deducted for every 2°C lower; if it is higher than 95°C, 1 point will be deducted for every 2°C higher; F is the score of the engine fuel consumption, which is obtained based on the actual fuel consumption and the standard fuel consumption. The standard fuel consumption is obtained according to the theoretical fuel consumption data in the vehicle operation manual. The scoring standard is within the interval [1 - 10]; when the actual fuel consumption is within the range of 90% - 110% of the standard fuel consumption, F can be rated 8 - 10 points; if the actual fuel consumption is higher than 110% of the standard fuel consumption, 1 point will be deducted for every 10% higher; if it is lower than 90% of the standard fuel consumption, 1 point will be added for every 10% lower, and the maximum score is 10 points;

[0109] The formula for the braking system performance score B is as follows:

[0110]

[0111] Where d is the weight of the braking pressure and e is the weight of the brake pad wear, and d + e = 1; P is the score of the braking pressure, which is obtained by comparing the actual braking pressure obtained by the braking pressure sensor with the standard braking pressure range. The scoring standard is within the interval [1 - 10]. When the actual braking pressure is within the standard range, P can be rated 8 - 10 points. If it is lower than the standard lower limit, 1 point will be deducted for every 1 MPa lower; if it is higher than the standard upper limit, 1 point will be deducted for every 1 MPa higher; W is the score of the brake pad wear, which is obtained according to the wear degree feedback by the brake pad wear sensor. The scoring standard is within the interval [1 - 10]. The wear range of the brake pad is 20% - 80% of its total thickness; when the wear degree is within this range, W can be rated 8 - 10 points; if the wear degree is lower than 20%, 1 point will be deducted for every 5% lower; if it is higher than 80%, 1 point will be deducted for every 5% higher;

[0112] The formula for the vehicle comprehensive status score V is as follows:

[0113]

[0114] Where f is the weight of the engine status score and g is the weight of the braking system performance score;

[0115] The environmental status scoring unit quantitatively evaluates the external environmental conditions of the vehicle's form through the weather, road, and traffic data provided by the in-vehicle system, providing data reference for dynamically adjusting the vehicle operation threshold and ensuring driving safety.

[0116] The specific method is as follows: Score the weather, road, and traffic flow respectively, assign weights according to the influence degree of each factor on driving safety and efficiency, and calculate the comprehensive score of the environmental status through the weighted average method. The weather condition score Ws determines the current weather type according to meteorological data, and each weather type corresponds to a fixed score. Sunny days can be scored 10 points, cloudy days 8 points, light rain days 6 points, heavy rain and rainstorm days 4 points, snowy days and icy bad weather 2 points. The road type score Rs obtains the road type information from the navigation system, and different road types are scored according to the specific display of the navigation system, with the scoring range being [1-10]. The traffic flow score Ts obtains the traffic volume level from the navigation system, and each level corresponds to a fixed score, with the scoring range being [1-10]. The traffic volume level is divided according to the number of vehicles passing through a certain section per unit time. Low traffic volume is scored 10 points when the number of vehicles passing through per hour is less than 500, medium traffic volume is 8 points for 500-1500 vehicles / hour; high traffic volume is 6 points for 1500-3000 vehicles / hour; congested traffic volume is scored 4 points when it is greater than 3000 vehicles / hour. According to the above data, the formula for the comprehensive score En of the environmental status is as follows:

[0117]

[0118] Among them, h is the weight of the weather condition, j is the weight of the traffic flow, i is the weight of the road type, and h + i + j = 1;

[0119] The basic threshold calculation module includes a vehicle speed threshold calculation unit and an acceleration threshold calculation unit. The vehicle speed threshold calculation unit determines the vehicle speed threshold according to the navigation speed limit, driver proficiency, and the environmental conditions of the vehicle. This threshold is used as the initial standard for judging whether the vehicle driving speed is safe and compliant. The calculation formula is as follows:

[0120] ;

[0121] Among them, the navigation speed limit is provided by the vehicle's navigation system according to the legal speed limit information of the section where the vehicle is located, and it is the basic reference value for calculating the vehicle speed threshold. In the formula, the part [1 - 0.1×(100 - driver proficiency score) / 100] reflects the influence of driver proficiency on the vehicle speed threshold. The higher the driver proficiency score, the smaller the value of (100 - driver proficiency score) / 100, and the closer the value after subtracting this value from 1 is to 1, which means that the vehicle speed threshold is more affected by driver proficiency and approaches the navigation speed limit.

[0122] The acceleration threshold calculation unit combines the vehicle acceleration reference value and the driver proficiency score to determine an acceleration threshold suitable for the current driver's level; at the same time, it makes dynamic adjustments according to the driver's proficiency level.

[0123] The calculation formula is as follows:

[0124] ;

[0125] The vehicle acceleration reference value is a basic acceleration reference value set by the vehicle manufacturer according to the design performance and safety standards of the vehicle, which reflects the reasonable acceleration range of the vehicle under normal conditions; the calculation method of the driver proficiency score is the same as that in the vehicle speed threshold calculation unit.

[0126] The dynamic threshold correction module includes a real-time feedback correction unit, a scene recognition correction unit, and a user habit learning correction unit;

[0127] The real-time feedback correction unit dynamically adjusts the initial threshold output by the basic threshold calculation module according to the real-time changes in vehicle speed and acceleration; the specific method is as follows: the system continuously collects vehicle speed and acceleration data at a set frequency; calculates the ratio of the difference in speed between two adjacent sampling moments to the sampling time interval; calculates the ratio of the difference in acceleration between two adjacent sampling moments to the sampling time interval; when the speed change rate exceeds 15% / s, adjust the vehicle speed threshold according to the following rules: every 0.5 seconds, multiply the current vehicle speed threshold by (1 - 0.5%) to get a new vehicle speed threshold; when the acceleration change rate exceeds 0.5m / s² / s, according to a pre-set attenuation rule related to acceleration, every 0.5 seconds, multiply the acceleration threshold by a decay coefficient less than 1, which is determined according to the vehicle dynamics characteristics and safety standards, to adjust the acceleration threshold;

[0128] The scene recognition correction unit identifies the specific scene where the vehicle is located by fusing the data of the navigation system and the camera, and adjusts the vehicle speed and acceleration thresholds to meet the safety driving requirements in different scenarios; the specific method is as follows: the navigation system continuously outputs the vehicle's position, driving direction, and planned route information, and the camera uses image recognition algorithms to process the captured images of the vehicle's surroundings to identify traffic signs, lane lines, and surrounding objects; then, fuse and analyze the navigation information and the camera image recognition results, and adjust the vehicle speed threshold and acceleration threshold according to the pre-set scene threshold adjustment rules;

[0129] The user habit learning correction unit records the operations of the user manually adjusting the threshold and the corresponding driving scene and behavior data, and automatically learns the user's habits by analyzing the impact of the user's multiple similar operations in the same scene on driving, so as to automatically adjust the threshold in the subsequent same scene;

[0130] The specific method is as follows: When the system detects each manual threshold adjustment operation by the user, it records the operation type (increase or decrease), the adjustment range, the driving scenario data at that time (including vehicle speed, road condition type, weather condition, characteristics of the section where the vehicle is located), and the user's driving behavior data (operation frequency and amplitude of acceleration, deceleration, and steering); When the system detects that the user performs 3 or more similar threshold adjustment operations in the same scenario (judging that the scenarios are the same based on an accurate matching algorithm for driving scenario data), it starts the analysis process; By analyzing the changes in the driving stability and safety indicators of the vehicle before and after these operations, it determines the direction and degree of the impact of the user's threshold adjustment on driving; Subsequently, when the vehicle is detected to be in the same scenario again, the system automatically adjusts the threshold according to the user's habitual pattern obtained from the previous analysis.

[0131] The holographic warning trigger module includes a first-level warning trigger unit, a second-level warning trigger unit, and a third-level warning trigger unit;

[0132] When the vehicle speed or acceleration reaches 90% of the threshold, the first-level warning trigger unit triggers the flashing of a first-level light blue icon and a voice prompt; It is realized through the in-vehicle display device and the voice broadcast system to remind the driver to pay attention to controlling the speed or acceleration;

[0133] When the vehicle speed or acceleration exceeds the threshold, the second-level warning trigger unit triggers the display of the difference value of a second-level yellow flashing triangle icon and a voice prompt; The difference between the vehicle speed or acceleration and the threshold is displayed on the in-vehicle display device, and the driver is reminded by voice broadcast that the threshold has been exceeded;

[0134] When the vehicle speed seriously exceeds the threshold (initially set as the vehicle speed exceeding the threshold by 20%) or the acceleration seriously exceeds the threshold (initially set as the acceleration exceeding the threshold by 30%), the third-level warning trigger unit triggers a third-level full-screen red flashing alarm and links to the safety system.

[0135] The reinforcement learning optimization module includes a state definition unit, an action definition unit, and a model training and updating unit;

[0136] The state definition unit real-time collects the driver's acceleration, deceleration, and steering operation data, obtains vehicle and environmental state information such as vehicle speed, acceleration, and vehicle environment score, records the current vehicle speed and acceleration threshold settings, tracks the most recently triggered warning level, and combines this information into a state vector;

[0137] The action definition unit endows the system with operation options for adjusting the threshold in different driving scenarios according to actions of predefined 5% increase, decrease, or maintenance of the vehicle speed or acceleration threshold, providing an action space for model optimization; in the predefined settings, there are three actions for the vehicle speed threshold: increasing by 5%, decreasing by 5%, and remaining unchanged, and the same actions of increasing by 5%, decreasing by 5%, and remaining unchanged are set for the acceleration threshold for the system to select and execute based on the current state;

[0138] The specific method of the model training and updating unit is as follows: First, build a deep Q-network. The number of neurons in its input layer is the same as the dimension of the state vector to input the state vector, the number of neurons in the output layer is the same as the size of the action space, and each neuron corresponds to the Q-value of an action. The middle layer uses the ReLU non-linear activation function to extract the features of the state vector to enhance the expression ability; during the operation of the system, continuously collect data on states, actions, and the next state and store them in the experience replay buffer; randomly select samples from the buffer at a preset time interval. For each sample, use the current deep Q-network to predict the Q-value of the selected action in the current state. According to the simplified rule, the Q-value of the next state is predicted by the target network, and then calculate the difference between the predicted Q-value and the target Q-value through a loss function such as the mean square error to obtain the loss value; then use the backpropagation algorithm to calculate the gradient of the loss function with respect to the network parameters, and use the stochastic gradient descent optimization algorithm to update the parameters according to the gradient to minimize the loss, and continuously repeat this process to improve the model performance.

[0139] A method for monitoring data of an automotive display device based on intelligent control, comprising the following steps:

[0140] S1. Use the vehicle body, camera, and navigation device to collect multi-dimensional data on vehicle movement, environment, driving operations, etc. in real time and transmit them to the data acquisition module;

[0141] S2. Align the collected raw data in time, filter out outliers, and extract key features for subsequent analysis; calculate the driver proficiency score according to the cumulative mileage, driving stability, and rule compliance; at the same time, calculate the comprehensive vehicle environment score in combination with the engine state, meteorology, road type, etc.;

[0142] S3. Generate initial vehicle speed and acceleration thresholds based on the navigation speed limit, driver score, and vehicle environment score; dynamically adjust the basic threshold by means of real-time feedback on vehicle speed and acceleration changes, identification of specific scenarios, and learning of user habits;

[0143] S4. Compare the real-time data with the threshold, and trigger a three-level holographic projection warning according to the degree of overrun to ensure driving safety;

[0144] S5. Combine the driver operations, vehicle and environmental states, current threshold settings, and the most recent warning level information collected in real time into a state vector, and pre-define actions to increase, decrease, or maintain the vehicle speed or acceleration threshold by 5%.

[0145] S6. Finally, build a model with a deep Q-network, input the state vector to output the action Q-value, continuously collect state and action data, randomly select samples at a preset time interval, calculate the loss function based on the predicted Q-value, and adjust the neural network parameters through the backpropagation algorithm to optimize the model performance.

[0146] Example:

[0147] Suppose there is a car equipped with a data monitoring system for automotive display devices based on intelligent control, and it is driving on an urban road.

[0148] After the car starts, various data collection devices start to work. The speed sensor, acceleration sensor, and steering angle sensor on the vehicle body collect the real-time motion data of the vehicle at a frequency of 100 times per second. The engine operation parameter sensors (such as the rotational speed sensor, water temperature sensor, and fuel consumption sensor) collect the engine-related data in real time. The braking system performance parameter sensors (braking pressure sensor and brake pad wear sensor) collect the braking system data and transmit these data to the data collection module. At the same time, the cameras in the front, rear, and on both sides of the vehicle use image recognition algorithms to capture images in real time for identifying traffic signs, lane lines, and surrounding vehicles and pedestrians. The camera identifies a speed limit sign of 60 km / h ahead. Through the interface with the navigation system, the vehicle position, driving direction, and planned route information are obtained. The navigation system shows that the vehicle is driving along the main urban road and the road ahead is a straight section.

[0149] The collected data enters the data preprocessing stage. The system constructs a timestamp mapping table to unify the data from different sensors and data sources into a time series based on the vehicle start time. Filter the outliers in the data according to the 3σ principle to ensure the reliability and accuracy of the data. Extract key features from the original data. For example, use the central difference method to calculate the absolute value of the change rate of speed per unit mileage. Suppose within a certain mileage, the speed changes from 50 km / h to 55 km / h within a driving distance of 1 km. Through the central difference method, the absolute value of the speed change rate is calculated to be approximately 5 km / h / km to quantify the speed fluctuation degree. Similarly, use the central difference method to calculate the absolute value of the change rate of the steering angle per unit mileage to measure the smoothness of the steering operation. Extract the engine load rate from the engine operation parameters. Suppose the measured engine torque is 200 N The engine speed is 2000 revolutions per minute. Through formula calculation, the actual output power is approximately 41.9 kW. Given that the rated power of this engine is 60 kW, the engine load rate is approximately 0.7. By measuring the air and fuel flow rates and combining their respective densities to calculate the air mass and fuel mass, the air-fuel ratio is then obtained. Suppose the calculated air-fuel ratio is 14.5 (fluctuating around the ideal air-fuel ratio of 14.7 for gasoline engines).

[0150] Next, calculate the driver proficiency score. Assume the cumulative driving mileage of the driver is 80,000 kilometers. Through formula calculation, the driving experience score is 25.5 points. Using the speed and steering angle data per unit mileage provided by the data preprocessing module, calculate the standard deviation of speed and the standard deviation of steering angle. Suppose the standard deviation of speed is 2 km / h and the standard deviation of steering angle is 3 degrees. According to formula calculation, the stability score is 8 points. Assume the driving mileage of this driver in the past year is 25,000 kilometers and the number of traffic rule violations is 2 times. According to formula calculation, the rule compliance score is 29.976 points. Weight the average of these three scoring factors to obtain that the driver proficiency score is approximately (25.5 + 8 + 29.976)÷3≈21.16 points.

[0151] Calculate the comprehensive vehicle environment score. In terms of vehicle state scoring, the engine operation parameters collected by the CAN bus interface show that the engine speed is 2000 revolutions per minute (the standard speed range is assumed to be 1500 - 3000 revolutions per minute), the water temperature is 85℃ (the normal water temperature range is 80 - 95℃), and the fuel consumption is calculated to be 105% of the standard fuel consumption (the standard fuel consumption is a fixed value according to the vehicle user manual). According to the scoring standard, the engine speed score R can be rated 9 points, the engine water temperature score T can be rated 9 points, and the engine fuel consumption score F can be rated 9 points. Assume the engine speed weight a = 0.4, the engine water temperature weight b = 0.3, and the engine fuel consumption weight c = 0.3. Through formula calculation, the engine state score E is 9 points. The brake system performance parameters show that the brake pressure is within the standard range (the standard brake pressure range is assumed to be 10 - 15 MPa and the actual brake pressure is 12 MPa), and the wear degree of the brake pads is 50% of the total thickness (within the range of 20% - 80%). Assume the brake pressure weight d = 0.6 and the brake pad wear weight e = 0.4. According to the scoring standard, the brake pressure score P can be rated 9 points and the brake pad wear score W can be rated 9 points. Through formula calculation, the brake system performance score B is 9 points. Assume the engine state score weight f = 0.6 and the brake system performance score weight g = 0.4. Through formula calculation, the comprehensive vehicle state score V is 9 points.

[0152] In terms of the environmental status score, the weather data provided by the in-vehicle system shows sunny weather, and according to the scoring standard, the weather condition score Ws is rated 10 points; the navigation system shows that the road type is the urban arterial road, and the road type score Rs is rated 7 points; the traffic flow level obtained by the navigation system is medium flow (1000 vehicles passing through per hour), and the traffic flow score Ts is rated 8 points. Assuming the weather condition weight h = 0.4, the road type weight i = 0.3, and the traffic flow weight j = 0.3, the comprehensive environmental status score En is calculated to be 8.5 points through the formula. Adding the vehicle status score and the environmental status score together, the comprehensive vehicle-environment score is approximately 17.5 points.

[0153] Calculate the basic threshold based on the above data. The legal speed limit provided by the navigation system for the section where the vehicle is located is 60 km / h. According to the vehicle speed threshold calculation formula, the vehicle speed threshold is calculated to be 53.5 km / h. The vehicle acceleration reference value is assumed to be 1.0 m / s², and according to the acceleration threshold calculation formula, the acceleration threshold is calculated to be 0.89 m / s².

[0154] During the vehicle driving process, the system continuously collects vehicle speed and acceleration data. Assume that at a certain moment, the vehicle speed change rate exceeds 15% / s. According to the dynamic threshold correction rule, every 0.5 seconds, the current vehicle speed threshold is multiplied by (1 - 0.5%) for adjustment. At the same time, through the fusion analysis of the navigation system and camera data, it is identified that the vehicle enters the school area scene. According to the pre-set scene threshold adjustment rule, the current vehicle speed threshold is adjusted from 53.5 km / h multiplied by (1 - 20%) to 42.8 km / h, and the current acceleration threshold is adjusted from 0.89 m / s² multiplied by (1 - 15%) to 0.76 m / s².

[0155] During driving, the system compares the vehicle speed and acceleration data with the thresholds in real time to trigger warnings. When the vehicle speed reaches 42.8×90% = 38.52 km / h, the in-vehicle display device triggers the flashing of a first-level light blue icon and voice prompt to remind the driver to pay attention to controlling the speed. If the vehicle speed exceeds 42.8 km / h, for example, reaches 45 km / h, the in-vehicle display device triggers a second-level yellow flashing triangle icon, displays the difference between the vehicle speed and the threshold (45 - 42.8 = 2.2 km / h), and reminds the driver that the threshold has been exceeded through voice broadcast. If the vehicle speed seriously exceeds the threshold and reaches 42.8×(1 + 20%) = 51.36 km / h, a third-level red full-screen flashing alarm is triggered and the safety system is linked.

[0156] During the entire driving process, the system also uses reinforcement learning to optimize the threshold adjustment strategy. It collects in real time the driver's acceleration, deceleration, and steering operation data, obtains vehicle and environmental state information such as vehicle speed, acceleration, and vehicle environment score, records the current vehicle speed and acceleration threshold settings, tracks the most recently triggered warning level, and combines this information into a state vector. According to the predefined rules, the system can choose actions to increase, decrease, or maintain the vehicle speed or acceleration threshold by 5% in different states. A deep Q-network is used to build the model, which takes the state vector as input and outputs the action Q-value. The system continuously collects state and action data, randomly selects samples from the data every 10 minutes, calculates the loss function based on the predicted Q-value, and adjusts the neural network parameters through the backpropagation algorithm to minimize the loss, continuously optimizing the model performance, so as to make the threshold adjustment strategy more reasonable.

[0157] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be construed as limiting the claimed invention.

Claims

1. An automobile display device data monitoring system based on intelligent control, characterized in that: It includes multi-source data acquisition module, data preprocessing module, driver score calculation module, vehicle score calculation module, basic threshold calculation module, dynamic threshold correction module, holographic warning trigger module and reinforcement learning optimization module; The multi-source data acquisition module is responsible for real-time acquisition of vehicle body sensor data, camera data, navigation data, CAN bus data and user operation records; the data preprocessing module is responsible for time alignment, outlier filtering and feature extraction of the collected raw data; The driver score calculation module calculates the driver proficiency score based on the accumulated mileage, speed fluctuation, and violation record; the vehicle environment score calculation module calculates the vehicle environment comprehensive score based on the engine status, weather, and road type; The basic threshold calculation module generates initial vehicle speed and acceleration thresholds according to the navigation speed limit, driver score, and vehicle environment score; the dynamic threshold correction module dynamically adjusts the threshold through real-time feedback, scene recognition, and user habit learning; the holographic warning trigger module triggers the three-level holographic projection warning based on the comparison between real-time data and the threshold; the reinforcement learning optimization module optimizes the threshold adjustment strategy through a deep Q network based on driving operations, vehicle status, threshold settings, and warning records.

2. According to claim 1, a data monitoring system for automobile display equipment based on intelligent control is characterized in that: The multi-source data acquisition module includes a sensor data acquisition unit, a camera data acquisition unit, a navigation data acquisition unit, a CAN bus data acquisition unit and a user operation record acquisition unit; The sensor data acquisition unit is responsible for collecting data from vehicle body sensors, engine operating parameter sensors, and brake system performance parameter sensors; for the vehicle body speed sensor, acceleration sensor, and steering angle sensor, it continuously collects real-time vehicle motion data, and converts the vehicle's physical motion state into electrical signals, which are transmitted to the data acquisition module via the vehicle network; for the engine operating parameter sensors, including but not limited to speed sensors, water temperature sensors, and fuel consumption sensors, it collects engine-related data in real time, which are also transmitted to the data acquisition module via the vehicle network for evaluating the engine state; Braking system performance parameter sensors, including brake pressure sensors and brake pad wear sensors used to obtain qualitative information on whether the brake pad wear is within a preset normal threshold, collect brake system data and transmit them to the data acquisition module to provide a basis for vehicle braking performance evaluation; The camera data acquisition unit uses cameras installed at the front, rear and both sides of the vehicle to capture images in real time using an image recognition algorithm; the camera transmits image data to the data acquisition module for identifying traffic signs, lane lines, and surrounding vehicles and pedestrians; The navigation data acquisition unit obtains the vehicle position, driving direction and planned route information provided by the navigation system in real time through the interface with the vehicle navigation system, providing a data basis for subsequent scene recognition and threshold calculation; The CAN bus data acquisition unit collects engine operating parameters and brake system performance parameter data on the CAN bus connected to each electronic control unit of the vehicle through the CAN bus interface, providing a data basis for subsequent vehicle status scoring and threshold calculation; The user operation record collection unit records the user's manual adjustment of the vehicle speed threshold and the acceleration threshold during driving, and records the driving scene information during the operation, including but not limited to the vehicle speed, road conditions, and weather, for subsequent user habit learning.

3. The automotive display device data monitoring system based on intelligent control according to claim 1 is characterized in that: The data preprocessing module includes a time alignment unit, an outlier filtering unit and a feature extraction unit; The time alignment unit is responsible for processing the differences in data acquisition frequency and time base between different sensors and data sources; By building a timestamp mapping table, all types of data are unified into a time series based on the vehicle start time, ensuring accurate association of different data at the same time point, providing a unified time reference for subsequent analysis; The outlier filtering unit calculates the mean and standard deviation of each data type according to the 3σ principle; when the deviation of a data point from the mean exceeds 3 times the standard deviation, it is determined to be an outlier and removed, thereby ensuring the reliability and accuracy of the data; The feature extraction unit is responsible for extracting features from the raw data for subsequent scoring calculation and threshold determination, including driving stability, steering angle change rate and engine status; for driving stability, the speed change trend per unit mileage is obtained by using the calculus method to derive the function of speed change with mileage, and the speed change trend is quantified by collecting discrete data of speed and mileage and using the central difference method to calculate the absolute value of the speed change rate per unit mileage; for the steering angle change rate, the absolute value of the steering angle change rate per unit mileage is calculated by using the collected discrete data of steering angle and mileage and using the central difference method; for the engine status, the engine load rate is obtained by measuring the engine torque and speed to calculate the actual output power, and then dividing it by the rated power; The air-fuel ratio is obtained by measuring the air and fuel flow rates, calculating the air mass and fuel mass in combination with their respective densities, and dividing the two. Different engines have different ideal ranges and the actual range will fluctuate with operating conditions; The extraction of these features provides a data basis for the subsequent driver score calculation module, vehicle environment score calculation module and basic threshold calculation module, helping the system to evaluate the vehicle's operating status and set thresholds.

4. The intelligent control-based automobile display device data monitoring system according to claim 3 is characterized in that: The driver scoring calculation module includes a driving experience scoring unit, a driving stability scoring unit and a rule compliance scoring unit; The driving experience scoring unit is responsible for calculating the driving experience score based on the accumulated driving mileage; the accumulated driving mileage data of the driver is converted into a corresponding score by adopting a logarithmic function relationship; the formula is as follows: ； The driving stability scoring unit uses the speed and steering angle data per unit mileage provided by the data preprocessing module to calculate the speed standard deviation and the steering angle standard deviation, and quantifies the fluctuation degree of the speed and the steering angle into a stability score; The calculation formula is as follows: ； The rule compliance scoring unit calculates the rule compliance score according to the ratio of the number of traffic rule violations to the annual mileage; the calculation formula is as follows: ， By correlating the number of violations with the annual mileage, it reflects the degree to which the driver complies with traffic rules in his historical driving process; Finally, the above three scoring factors are weighted averaged to obtain the driver proficiency score.

5. The automotive display device data monitoring system based on intelligent control according to claim 2 is characterized in that: The vehicle environment score calculation module includes a vehicle state scoring unit and an environment state scoring unit; The vehicle status scoring unit receives engine operating parameters and brake system performance parameter data collected by the CAN bus interface, performs quantitative evaluation on these data, and comprehensively reflects the mechanical performance and operating status of the vehicle; The specific method is: score the operating parameters of the engine and brake system respectively, assign weights to the data transmitted through the CAN bus according to the importance of each part, and calculate the comprehensive score of the vehicle status by weighted average method; the formula of the engine status score E is as follows: ； Where a is the weight of the engine speed, b is the weight of the engine water temperature, and c is the weight of the engine fuel consumption. The value range of a, b, and c is between 0 and 1, and a+b+c=1. R is the score of the engine speed, which is determined according to the degree of deviation between the actual engine speed and the standard speed range. The scoring standard is in the interval [1-10]. When the actual speed is within the standard range, R can be rated as 8-10 points. If it is lower than the lower limit of the standard range, 1 point will be deducted for every 100 revolutions lower. If it is higher than the upper limit of the standard range, 1 point will be deducted for every 100 revolutions higher. T is the score of the engine water temperature, which is determined according to the actual engine water temperature. The difference between the actual fuel consumption and the normal water temperature range is evaluated. The scoring standard is within the range [1-10]. If the water temperature is within this range, T can be rated as 8-10 points. If the water temperature is lower than 80℃, 1 point will be deducted for every 2℃ lower. If the water temperature is higher than 95℃, 1 point will be deducted for every 2℃ higher. F is the score of engine fuel consumption, which is based on the actual fuel consumption and the standard fuel consumption. The standard fuel consumption is based on the theoretical fuel consumption data in the vehicle manual. The scoring standard is within the range [1-10]. If the actual fuel consumption is within the range of 90%-110% of the standard fuel consumption, F can be rated as 8-10 points. If the actual fuel consumption is higher than the standard fuel consumption by 110%, 1 point will be deducted for every 10% higher. If the actual fuel consumption is lower than the standard fuel consumption by 90%, 1 point will be added for every 10% lower. The maximum score is 10 points. The formula for the brake system performance score B is as follows: ； Wherein, d is the weight of brake pressure, e is the weight of brake pad wear, and d+e=1; P is the score of brake pressure, which is obtained by comparing the actual brake pressure obtained by the brake pressure sensor with the standard brake pressure range. The scoring standard is in the interval [1-10]. If the actual brake pressure is within the standard range, P can be rated as 8-10 points. If it is lower than the lower limit of the standard, 1 point will be deducted for every 1MPa lower. If it is higher than the upper limit of the standard, 1 point will be deducted for every 1MPa higher. W is the score of brake pad wear, which is obtained according to the wear degree feedback from the brake pad wear sensor. The scoring standard is in the interval [1-10]. The wear range of the brake pad is 20%-80% of its total thickness. When the wear degree is within this range, W can be rated as 8-10 points. If the wear degree is lower than 20%, 1 point will be deducted for every 5% lower. If it is higher than 80%, 1 point will be deducted for every 5% higher. The formula for the vehicle's comprehensive status score V is as follows: ； Where f is the weight of the engine status score, and g is the weight of the brake system performance score; The environmental status scoring unit quantitatively evaluates the external environmental conditions of the vehicle through the weather, road and traffic data provided by the vehicle computer system; and provides data reference for dynamically adjusting the vehicle operation threshold and ensuring driving safety; The specific method is as follows: the weather, road and traffic flow are scored respectively, weights are assigned according to the degree of influence of each factor on driving safety and efficiency, and the comprehensive score of the environmental status is calculated by the weighted average method; the weather condition score Ws determines the current weather type according to meteorological data, and each weather type corresponds to a fixed score, with sunny days scoring 10 points, cloudy weather scoring 8 points, light rain weather scoring 6 points, heavy rain and rainstorm weather scoring 4 points, and snowy days and icy weather scoring 2 points; the road type score Rs obtains road type information from the navigation system, and different road types are scored according to the specific display of the navigation system, with a score range of [1-10]; the traffic flow score Ts obtains the traffic volume level through the navigation system, and each level corresponds to a fixed score, with a score range of [1-10]; the level is divided according to the number of vehicles passing through a certain section of road per unit time, and low traffic is less than 500 vehicles per hour, scoring 10 points, medium traffic is 500-1500 vehicles / hour, scoring 8 points, and high traffic is 1500-3000 vehicles / hour. If the traffic volume is more than 3,000 vehicles / hour, the score is 6 points; if the traffic volume is more than 3,000 vehicles / hour, the score is 4 points; Based on the above data, the formula for the comprehensive score En of the environmental status is as follows: ； Among them, h is the weight of weather conditions, j is the weight of traffic flow, i is the weight of road type, and h+i+j=1.

6. The automotive display device data monitoring system based on intelligent control according to claim 1 is characterized in that: The basic threshold calculation module includes a vehicle speed threshold calculation unit and an acceleration threshold calculation unit; The vehicle speed threshold calculation unit determines the vehicle speed threshold according to the navigation speed limit, the driver's proficiency and the vehicle's environment. The calculation formula is as follows: ； The navigation speed limit is provided by the vehicle's navigation system based on the legal speed limit information of the road section, and it is the basic reference value for calculating the vehicle speed threshold. The acceleration threshold calculation unit determines an acceleration threshold suitable for the current driver level by combining the vehicle acceleration reference value and the driver proficiency score, and dynamically adjusts it according to the driver's proficiency. The calculation formula is as follows: ； The vehicle acceleration baseline value is a basic acceleration reference value set by the vehicle manufacturer based on the vehicle's design performance and safety standards. The driver proficiency score is calculated in the same way as in the vehicle speed threshold calculation unit.

7. The automotive display device data monitoring system based on intelligent control according to claim 1 is characterized by: The dynamic threshold correction module includes a real-time feedback correction unit, a scene recognition correction unit and a user habit learning correction unit; The real-time feedback correction unit dynamically adjusts the initial threshold value output by the basic threshold value calculation module according to the real-time changes of the vehicle speed and acceleration; the specific method is: the system continuously collects the vehicle speed and acceleration data at a set frequency; calculates the ratio of the speed difference between two adjacent sampling moments to the sampling time interval; calculates the ratio of the acceleration difference between two adjacent sampling moments to the sampling time interval; when the speed change rate exceeds a preset range, adjusts the vehicle speed threshold value and the acceleration threshold value according to a preset adjustment rule; The scene recognition correction unit identifies the specific scene in which the vehicle is located by fusing the data of the navigation system and the camera, and adjusts the vehicle speed and acceleration thresholds. The specific method is as follows: the navigation system continuously outputs the vehicle's position, driving direction and planned route information, and the camera uses an image recognition algorithm to process the captured images around the vehicle to identify traffic signs, lane lines and surrounding objects; then the navigation information is fused and analyzed with the camera image recognition results, and the vehicle speed threshold and acceleration threshold are adjusted according to the pre-set scene threshold adjustment rules; The user habit learning and correction unit records the user's manual threshold adjustment operation and the corresponding driving scene and behavior data, and automatically learns the user's habits by analyzing the impact of multiple similar operations of the user on driving in the same scene, so as to automatically adjust the threshold in the subsequent same scene; The specific method is: each time the user manually adjusts the threshold, the system records the operation type, adjustment range, driving scene data at the time, and the user's driving behavior data; when the system detects the same scene, the same scene refers to the scene determined to be the same based on the precise matching algorithm of the driving scene data; the system automatically adjusts the threshold according to the user habit pattern obtained from the previous analysis.

8. The automotive display device data monitoring system based on intelligent control according to claim 1 is characterized by: The holographic warning trigger module includes a first-level warning trigger unit, a second-level warning trigger unit and a third-level warning trigger unit; The first-level warning trigger unit triggers a flashing light blue icon and voice prompt when the vehicle speed or acceleration reaches 90% of the threshold; this is achieved through the in-vehicle display device and voice broadcast system to remind the driver to pay attention to controlling the speed or acceleration; The secondary warning trigger unit triggers the secondary yellow flashing triangle icon to display the difference and voice prompt when the vehicle speed or acceleration exceeds the threshold; the difference between the vehicle speed or acceleration and the threshold is displayed on the in-vehicle display device, and the driver is reminded by voice that the threshold has been exceeded; The three-level warning trigger unit triggers a three-level red full-screen flashing alarm and links the safety system when the vehicle speed exceeds a threshold of 20%, or the acceleration exceeds a threshold of 30%.

9. The automobile display device data monitoring system based on intelligent control according to claim 1, characterized in that: The reinforcement learning optimization module includes a state definition unit, an action definition unit and a model training and updating unit; The state definition unit collects the driver's acceleration, deceleration, and steering operation data in real time, obtains vehicle and environmental state information such as vehicle speed, acceleration, and vehicle environmental score, records the current vehicle speed and acceleration threshold settings, tracks the most recently triggered warning level, and combines this information into a state vector; The action definition unit gives the system the operation option of adjusting the threshold value in different driving scenarios according to the predefined actions of increasing, decreasing, or maintaining the vehicle speed or acceleration threshold value by 5%, thereby providing an action space for model optimization; the predefined actions of increasing the vehicle speed threshold value by 5%, decreasing by 5%, and maintaining the same are set, and the actions of increasing the acceleration threshold value by 5%, decreasing by 5%, and maintaining the same are also set, so that the system can select and execute based on the current state; The specific method of the model training and updating unit is as follows: first, a deep Q network is built, the number of neurons in the input layer is consistent with the dimension of the state vector, so as to input the state vector, the number of neurons in the output layer is the same as the size of the action space, each neuron corresponds to the Q value of an action, and the middle layer uses the ReLU nonlinear activation function to extract the state vector features to enhance the expression ability; when the system is running, the data of the state, action and the next state are continuously collected and stored in the experience playback buffer; samples are randomly selected from the buffer at a preset time interval, and for each sample, the current deep Q network is used to predict the Q value of the selected action in the current state, and according to the simplified rules, the Q value of the next state is predicted by the target network, and then the difference between the predicted Q value and the target Q value is calculated by the loss function such as the mean square error to obtain the loss value; then the back propagation algorithm is used to calculate the gradient of the loss function to the network parameters, and the parameters are updated according to the gradient with the help of the stochastic gradient descent optimization algorithm to minimize the loss, and this process is repeated to improve the model performance.

10. A method for monitoring data of an automobile display device based on intelligent control, applied to a method for monitoring data of an automobile display device based on intelligent control as claimed in any one of claims 1 to 9, characterized in that: The following steps are involved: S1. Use the vehicle body, camera and navigation equipment to collect multi-dimensional data such as vehicle movement, environment, driving operation, etc. in real time and transmit it to the data acquisition module; S2, time-align the collected raw data, filter outliers, and extract key features for subsequent analysis; The driver proficiency score is calculated based on the accumulated mileage, driving stability, and compliance with the rules. At the same time, the comprehensive vehicle environment score is calculated based on the engine status, weather, road type, etc. S3: Generate initial vehicle speed and acceleration thresholds based on navigation speed limit, driver score, and vehicle environment score; dynamically adjust basic thresholds by taking advantage of real-time feedback of vehicle speed and acceleration changes, identifying specific scenarios, and learning user habits; S4: Compare the real-time data with the threshold value, and trigger the three-level holographic projection warning according to the degree of over-limit to ensure driving safety; S5. Combine the real-time collected driver operation, vehicle and environment status, current threshold setting and the latest warning level information into a state vector, and predefine the action of increasing, decreasing or maintaining the vehicle speed or acceleration threshold by 5%; S6. Finally, build a model with a deep Q network, input the state vector and output the action Q value, continuously collect state and action data, randomly select samples at preset time intervals, calculate the loss function based on the predicted Q value, adjust the neural network parameters through the back propagation algorithm, and optimize the model performance.

Images