A vehicle light control data processing method and system based on cloud computing

Abstract

The present invention provides a method and system for processing vehicle light control data based on cloud computing, in the field of intelligent transportation and the Internet of Things, the method comprising: obtaining vehicle light status data, including the on or off status, brightness, flickering frequency and energy consumption of the vehicle light; cleaning, classifying, aggregating and analyzing the status data to obtain processed status data; identifying the stability, energy consumption change and abnormal situation of the vehicle light through preset rules according to the processed status data to obtain a recognition result; reversely optimizing the control logic of the vehicle light according to the recognition result and the control parameters of the vehicle light to obtain a reverse optimization result, the control parameters including the initial brightness setting and the flickering mode; generating optimized vehicle light control parameters and instructions according to the reverse optimization result. The present invention can automatically adjust the control parameters of the vehicle light through cloud computing and reinforcement learning technology to adapt to different driving conditions and optimization goals, and realize intelligent and automated vehicle light control.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation and Internet of Things, and in particular to a vehicle light control data processing method and system based on cloud computing. Background Art

[0002] In today's automotive industry, with the continuous advancement of technology and the growing demand for intelligence, headlight control systems are undergoing a transformation from traditional manual adjustment to intelligent and automated control. However, this transformation process faces many technical challenges, which constitute the background technology for the development of cloud computing-based headlight control data processing methods.

[0003] First, traditional headlight control systems often rely on manual settings and adjustments, and cannot achieve real-time status monitoring and data analysis. This makes it difficult to ensure the stability and safety of headlights during driving, especially at night or in bad weather conditions, where unstable light output may increase the risk of traffic accidents.

[0004] Secondly, traditional headlight control systems have shortcomings in energy consumption management. Due to the lack of effective data analysis methods, it is impossible to accurately identify changes in the energy consumption of headlights and formulate effective energy-saving strategies accordingly. This not only leads to energy waste, but also may shorten the service life of headlights and increase maintenance and replacement costs.

[0005] In addition, traditional headlight control systems also have limitations in monitoring and handling abnormal conditions. Once the headlights have abnormal conditions such as overheating and unstable voltage, they are often unable to be discovered and handled in time, which may cause damage to the headlights and even affect driving safety. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a cloud computing-based vehicle light control data processing method and system, which can automatically adjust the control parameters of the vehicle lights through cloud computing and reinforcement learning technology to adapt to different driving conditions and optimization goals, thereby realizing intelligent and automated vehicle light control.

[0007] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0008] In a first aspect, a method for processing vehicle light control data based on cloud computing is provided, the method comprising:

[0009] Get the status data of the lights, including whether they are on or off, brightness, flashing frequency, and energy consumption;

[0010] Clean, classify, aggregate and analyze the status data to obtain processed status data;

[0011] According to the processed status data, the stability, energy consumption changes and abnormal conditions of the vehicle lights are identified through preset rules to obtain the identification results;

[0012] According to the recognition result and the control parameters of the vehicle light, the control logic of the vehicle light is reversely optimized to obtain a reverse optimization result, wherein the control parameters include an initial brightness setting and a flashing mode;

[0013] Generate optimized vehicle light control parameters and instructions based on the reverse optimization results;

[0014] The headlights are controlled according to the optimized headlight control parameters and instructions.

[0015] Further, the status data is cleaned, classified, aggregated and analyzed to obtain processed status data, including:

[0016] By checking whether there are duplicate records in the status data set, and cleaning the missing values ​​and abnormal values ​​according to the inspection results, the cleaned data is obtained;

[0017] Classify the cleaned data according to the type and status of the headlights to obtain classified data;

[0018] Aggregate the classified data according to the time window and state, and extract features from the aggregated data to obtain aggregated and feature-extracted data;

[0019] Perform data analysis on the aggregated and feature extracted data to obtain processed status data;

[0020] Furthermore, the classified data is aggregated according to the time window and the state, and features are extracted from the aggregated data to obtain aggregated and feature-extracted data, including:

[0021] According to the time window and state, the classified data is passed Aggregate to obtain aggregate data, where is in the time window t Three aggregate features within, including mean, sum, and count; Indicates that in the time window t The state inside is s The average value of Indicates that in the time window t The state inside is s The number of valid samples of Indicates that in the time window t The state inside is s No. i The value of the record, Indicates that in the time window t The state inside iss The sum of Indicates that in the time window t The state inside is s The number of records reflects the number of occurrences of the state in the time window; s is the "on" or "off" state, t is the timestamp;

[0022] Extract the average, sum, count, and state duration from the aggregated data to get the aggregated and feature extracted data.

[0023] Furthermore, data analysis is performed on the aggregated and feature-extracted data to obtain processed status data, including:

[0024] Arrange the aggregated data in time series, draw a time series graph, and analyze the changing trend of the state data in different time windows to obtain the changing trend analysis results;

[0025] Get the average, sum, and count features by Calculate the correlation analysis of the features to obtain the association analysis results, where: Indicates that in the time window t The state inside is s The average value of Indicates that in the time window t The state inside is s The overall average value, Indicates that in the time window t The state inside is s The sum of Indicates that the state in time window t is s The overall mean of the sum of is the correlation coefficient;

[0026] pass Identify outliers of feature data to obtain outlier detection results, where: Indicates that in the time window t The state inside is s The average value of is the feature data point to be detected. represents the total number of samples in the dataset, and is the index, Indicates the standard deviation unit difference between the current feature data point to be detected and the sample mean;

[0027] The analysis and processing results are obtained based on the change trend analysis results, correlation analysis results and outlier detection results.

[0028] Furthermore, based on the processed status data, the stability, energy consumption changes and abnormal conditions of the vehicle lights are identified by preset rules to obtain identification results, including:

[0029] According to the processed status data, the threshold of brightness change, the normal range of energy consumption and the energy consumption trend analysis are set to obtain the preset rules;

[0030] According to preset rules, the stability, energy consumption changes and abnormal conditions of the vehicle lights are identified to obtain the identification results.

[0031] Further, according to the recognition result and the control parameters of the headlight, the control logic of the headlight is reversely optimized to obtain a reverse optimization result, wherein the control parameters include an initial brightness setting and a flashing mode, including:

[0032] Based on the identification results, determine the optimization goals, including: improving stability, reducing energy consumption and handling abnormal situations;

[0033] According to the control parameters and optimization objectives of the headlights, a reverse optimization strategy is obtained;

[0034] According to the inverse optimization strategy, the control parameters are automatically adjusted through reinforcement learning to obtain the inverse optimization results.

[0035] Furthermore, according to the inverse optimization strategy, the control parameters are automatically adjusted through reinforcement learning to obtain the inverse optimization results, including:

[0036] According to the reverse optimization strategy, the state space and action space are determined. The state space includes the ambient light intensity, vehicle speed, current brightness, current flashing mode, current energy consumption and abnormal state, and the action space includes the parameters for setting the initial brightness, initial brightness and working frequency;

[0037] According to the reverse optimization strategy, Get the reward function, where is a reward function that evaluates the effect in a given state; , is the weight parameter; is the target brightness value, is the actual output brightness value, is the highest score for stability, with a value of 100; It represents the upper limit of energy consumption under ideal conditions, which is 200 watts; is the bonus score for successfully handling an exception, with a value of 50; is the current energy consumption value;

[0038] According to the state space, action space and reward function, the control parameters and control parameters of the headlights are automatically adjusted to obtain the inverse optimization results.

[0039] In a second aspect, a vehicle light control data processing system based on cloud computing includes:

[0040] The acquisition module is used to acquire the status data of the vehicle lights, including the on or off status, brightness, flashing frequency and energy consumption of the vehicle lights; clean, classify, aggregate and analyze the status data to obtain processed status data; and identify the stability, energy consumption changes and abnormal conditions of the vehicle lights through preset rules based on the processed status data to obtain identification results;

[0041] The processing module is used to reversely optimize the control logic of the headlights according to the recognition results and the control parameters of the headlights to obtain reverse optimization results, wherein the control parameters include initial brightness settings and flashing modes; generate optimized headlight control parameters and instructions according to the reverse optimization results; and control the headlights according to the optimized headlight control parameters and instructions.

[0042] According to a third aspect, a computing device includes:

[0043] one or more processors;

[0044] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0045] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0046] The above solution of the present invention includes at least the following beneficial effects:

[0047] By real-time monitoring and analysis of the status data of the lights, potential unstable factors such as brightness fluctuations and abnormal flashing frequencies can be discovered and handled in a timely manner, thereby improving the stability of the lights; stable light output helps improve driving safety at night or in bad weather conditions and reduce the risk of traffic accidents. Through data analysis and processing, the energy consumption changes of the lights can be accurately identified, and effective energy-saving strategies can be formulated, such as adjusting the brightness and flashing mode; to reduce energy consumption, the abnormal state of the lights can be monitored in real time, and measures can be taken quickly to prevent the abnormal situation from further developing and causing damage to the lights or affecting driving safety. Through cloud computing and reinforcement learning technology, the control parameters of the lights can be automatically adjusted to adapt to different driving conditions and optimization goals, realizing intelligent and automated light control. The optimized light control logic can automatically adjust the light output, which helps to improve the intelligence and automation level of the entire transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1It is a flow chart of a vehicle light control data processing method based on cloud computing provided by an embodiment of the present invention.

[0049] Figure 2 It is a schematic diagram of a vehicle light control data processing system based on cloud computing provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in a form and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0051] like Figure 1 As shown, an embodiment of the present invention provides a vehicle light control data processing method based on cloud computing, the method comprising the following steps:

[0052] Step 11, obtaining status data of the vehicle lights, including the on or off status, brightness, flashing frequency and energy consumption of the vehicle lights;

[0053] Step 12, cleaning, classifying, aggregating and analyzing the status data to obtain processed status data;

[0054] Step 13, based on the processed state data, identifying the stability, energy consumption change and abnormality of the vehicle lamp by using preset rules to obtain an identification result;

[0055] Step 14, according to the recognition result and the control parameters of the headlight, reversely optimize the control logic of the headlight to obtain a reverse optimization result, the control parameters including the initial brightness setting and the flashing mode;

[0056] Step 15, generating optimized vehicle light control parameters and instructions according to the reverse optimization result;

[0057] Step 16, controlling the headlights according to the optimized headlight control parameters and instructions.

[0058] In the embodiment of the present invention, by acquiring the status data of the headlights and performing detailed analysis, the headlight control logic is reversely optimized to ensure that the headlights consume the least energy while providing the required brightness and stability; by real-time monitoring and analysis of the stability of the headlights, unstable situations can be detected in time; by identifying and optimizing the control logic, the instability of the system can be effectively reduced to ensure that the headlights continue to work in the optimal state; by processing the status data and identifying abnormal situations, the headlights can respond quickly and take measures when a fault or abnormality occurs; by reverse optimizing the control parameters, the headlight settings can be automatically adjusted without human intervention to adapt to different environments and driving conditions; by accurately adjusting the brightness and flashing frequency, the headlights can better adapt to different road conditions and weather conditions and enhance driving safety; by optimizing the control logic of the headlights, the high energy consumption and excessive work of the headlights in unnecessary situations are reduced, and by effective abnormality detection and processing, faults can be identified and solved in real time; by finely adjusting the brightness and flashing frequency of the headlights, the owner can experience a more comfortable, safe and intelligent driving experience. Reverse optimization of the headlight control logic not only takes into account energy efficiency and stability, but also adjusts the working mode of the headlights in different environments.

[0059] In a preferred embodiment of the present invention, the above step 11 may include:

[0060] By connecting the sensor to the microcontroller, the sensor status data is received and processed, including the on or off status, brightness, flashing frequency and energy consumption of the car light.

[0061] In the embodiments of the present invention, by continuously monitoring the status of the lights, any abnormal situation can be discovered in time, and the brightness of the lights can be automatically adjusted according to the ambient light. By monitoring the energy consumption of the lights, high-energy-consuming lamps can be identified and optimized to reduce overall energy consumption, thereby saving fuel and electricity. The lights can be automatically turned off when they are not necessary to reduce ineffective energy consumption. By analyzing the data, the trend of lamp performance degradation can be identified, so that measures can be taken in advance to avoid failures. Faults can be automatically identified and alarms can be issued, reducing the frequency and complexity of manual inspections.

[0062] In a specific embodiment of the present invention, the sensor is connected to the microcontroller using a connecting wire and an interface, and a code is written to initialize the sensor and set configuration parameters. For the light sensor, it is configured to measure the ambient light intensity with an appropriate resolution and frequency, and for the current sensor, it is ensured that it can accurately measure the current when the headlight is working; the code is written to monitor the output of the current sensor or directly read the headlight control signal to turn the headlight on or off, and adjust the brightness level of the headlight according to the output of the light sensor. It is controlled by the timer function of the microcontroller and calculated by the current measured by the current sensor and the known voltage.

[0063] In a preferred embodiment of the present invention, the above step 12 may include:

[0064] Step 121, by checking whether there are duplicate records in the status data set, and cleaning the missing values ​​and abnormal values ​​according to the checking result, to obtain cleaned data;

[0065] Step 122, classifying the cleaned data according to the type and state of the vehicle lamp to obtain classified data;

[0066] Step 123, aggregating the classified data according to the time window and the state, and extracting features from the aggregated data to obtain aggregated and feature-extracted data;

[0067] Step 124, performing data analysis processing on the aggregated and feature extracted data to obtain processed status data;

[0068] In an embodiment of the present invention, by checking whether there are duplicate records in the status data set and cleaning missing values ​​and abnormal values, the accuracy and completeness of the data can be ensured; by classifying the cleaned data according to the type and status of the headlights, the data can be made more orderly and easier to manage; by aggregating the classified data according to the time window and status, and extracting features from the aggregated data, the value of the data can be refined. The aggregated data can reflect the changing trend of the data in the time dimension, and the feature extraction can highlight the key information and patterns in the data. By performing data analysis and processing on the aggregated and feature-extracted data, the potential rules and associations in the data can be deeply explored. By processing the status data, the status and performance of the headlights can be more accurately understood.

[0069] In a specific embodiment of the present invention, a state data set is loaded into a data processing environment, duplicate records in the data set are checked and deleted, the data set is traversed, all records containing missing values ​​are identified, and the missing values ​​are processed according to their nature and quantity. Statistical methods are used to identify outliers, correct or delete outliers, and complete data cleaning. Determine the classification criteria for headlight types and states. For example, headlight types may include low beams, high beams, and turn signals; states may include on, off, and flashing; add new columns or fields to the data set to store the headlight type and state classification labels for each record. Traverse each record in the data set and assign corresponding classification labels to it according to its headlight type and state information. Aggregate the classified data according to the time window and state, and extract useful features from the aggregated data. Select statistical analysis or time series analysis for analysis, and use the selected analysis method to process the features.

[0070] In a preferred embodiment of the present invention, the above step 123 may include:

[0071] Step 1231, according to the time window and state, the classified data is Aggregate to obtain aggregate data, where is in the time window t Three aggregate features within, including mean, sum, and count; Indicates that in the time window t The state inside is s The average value of Indicates that in the time window t The state inside is s The number of valid samples of Indicates that in the time window t The state inside is s No. i The value of the record, Indicates that in the time window t The state inside is s The sum of Indicates that in the time window t The state inside is s The number of records reflects the number of occurrences of the state in the time window; s is the "on" or "off" state, t is the timestamp;

[0072] Step 1232, extract the average, sum, count and state duration from the aggregated data to obtain aggregated and feature extracted data.

[0073] In the embodiment of the present invention, the classified data can be simplified by aggregating them according to time windows and states. The aggregated data can reveal the changing trends and patterns of the light state over time, which helps to identify potential periodic behaviors or abnormal changes. By calculating aggregate features such as averages, sums, and counts, the noise and fluctuations in the original data can be reduced. The averages, sums, counts, and state duration features can be extracted from the aggregated data to highlight the key information and patterns in the data. The extracted features can be used as a basis for decision making.

[0074] In a specific embodiment of the present invention, the time window size is selected, and the state to be analyzed, such as "on" or "off", is determined, and the data is grouped according to the time window and state. For each time window and state combination, the average, sum, and count are calculated, and the calculated aggregate features are stored in a new data frame, which should contain the time window, state, and corresponding average, sum, and count columns. The average column, sum column, and count column are directly obtained from the aggregate data frame, and the state is calculated in each time window. s duration.

[0075] In a preferred embodiment of the present invention, the above step 124 may include:

[0076] Step 1241, arranging the aggregated data in time series, drawing a time series graph, and analyzing the change trend of the state data in different time windows to obtain a change trend analysis result;

[0077] Step 1242, obtain the average, sum and count features, by Calculate the correlation analysis of the features to obtain the association analysis results, where: Indicates that in the time window t The state inside is s The average value of Indicates that in the time window t The state inside is s The overall average value, Indicates that in the time window t The state inside is s The sum of Indicates that the state in time window t is s The overall mean of the sum of is the correlation coefficient;

[0078] Step 1243, pass Identify outliers of feature data to obtain outlier detection results, where: Indicates that in the time window t The state inside is s The average value of is the feature data point to be detected. represents the total number of samples in the dataset, and is the index, Indicates the standard deviation unit difference between the current feature data point to be detected and the sample mean;

[0079] Step 1244, obtaining analysis and processing results according to the change trend analysis results, the correlation analysis results and the outlier detection results.

[0080] In the embodiment of the present invention, by drawing a time series graph, the changing trend of the status data in different time windows can be intuitively displayed. The time series graph helps to identify periodic behaviors in the data, such as whether the turning on and off of the lights shows a certain pattern. By observing the time series graph, it is easier to find abnormal points or mutations in the data. By calculating the correlation between the average, sum and count features, correlation analysis helps to optimize feature selection and remove redundant or weakly correlated features; by identifying and processing outliers, the overall quality and reliability of the data can be improved. Outliers may reflect problems in the data collection or processing process. By combining the results of the trend analysis, the correlation analysis and the outlier detection results, the status of the lights can be comprehensively evaluated and analyzed.

[0081] In a specific embodiment of the present invention, it is ensured that the aggregated data is arranged in the order of timestamps or time windows, and the timestamps or time windows are set as x Axis, status data (such as average, sum, or count) as y Axis, draw a time series graph to obtain the changing trend of state data in different time windows; extract the average, total and count features from the aggregated data, calculate the correlation coefficient between the features through the correlation calculation function, the correlation coefficient is between -1 and 1, indicating the strength and direction of the linear relationship between the features; analyze the value and direction of the correlation coefficient to determine the degree of association between the features; calculate the standard deviation unit difference between the feature data and the sample mean, for each time window and state combination of feature data points, calculate its standard deviation unit difference with the sample mean, according to the detection criteria, identify the outliers in the feature data; integrate the time series graph, correlation analysis results and outlier detection results to form a comprehensive analysis report.

[0082] In a preferred embodiment of the present invention, the above step 13 may include:

[0083] Step 131, according to the processed status data, setting the threshold of brightness change, setting the normal range of energy consumption and energy consumption trend analysis to obtain preset rules;

[0084] Step 132, identifying the stability, energy consumption change and abnormal conditions of the vehicle lights according to preset rules to obtain an identification result.

[0085] In the embodiment of the present invention, by setting the threshold of brightness change, the normal range of energy consumption and performing energy consumption trend analysis, a clear judgment standard is provided for the stability of the headlights, energy consumption changes and identification of abnormal situations. Setting the normal range of energy consumption and trend analysis helps to optimize the energy consumption management of the headlights, ensuring that the headlights can operate efficiently while reducing energy costs. Through energy consumption trend analysis, possible energy consumption changes in the future can be predicted. By identifying the stability of the headlights, potential problems can be discovered and solved in a timely manner to ensure that the headlights can operate stably in harsh environments. Identifying energy consumption changes helps to discover energy consumption anomalies in a timely manner, take corresponding energy-saving measures, and reduce energy consumption during the operation of the headlights. By identifying abnormal situations, faults or problems in the headlight system can be quickly responded to and handled to avoid adverse effects on vehicle driving.

[0086] In a specific embodiment of the present invention, the status data of the vehicle lights, including brightness value, energy consumption value, working time, etc., are collected, and the data is cleaned to remove abnormal values ​​or noise; historical data is analyzed to determine the normal brightness fluctuation range and set upper and lower thresholds; the energy consumption range is determined based on the rated power of the vehicle lights and historical energy consumption data; time series analysis and regression analysis methods are used to analyze the trend of energy consumption changes over time. Based on the brightness change threshold, normal range of energy consumption and energy consumption trend analysis results, preset rules are formulated. The rules cover normal situations, warning conditions and abnormal judgment criteria. Among them, the brightness data of the headlights is monitored in real time and compared with the preset brightness change threshold. If the brightness change is within the threshold range, the stability of the headlights is considered to be good, otherwise, a stability warning is issued; the energy consumption data of the headlights is recorded and analyzed in real time, and the energy consumption values ​​are compared with the preset normal range to identify whether the energy consumption is too high or too low; combined with the energy consumption trend analysis, it is determined whether the energy consumption change is reasonable, for example, whether there is a sudden increase or decrease; based on the analysis results of brightness stability and energy consumption changes, possible abnormal situations are identified, which may include: abnormal brightness fluctuations, energy consumption exceeding the normal range, and abnormal energy consumption trends; for identified abnormalities, the alarm mechanism is triggered in a timely manner, and the abnormal information is recorded.

[0087] In a preferred embodiment of the present invention, the above step 14 may include:

[0088] Step 141, determining optimization goals according to the recognition results, including: improving stability, reducing energy consumption and handling abnormal situations;

[0089] Step 142, obtaining a reverse optimization strategy according to the control parameters and optimization target of the vehicle light;

[0090] Step 143, according to the inverse optimization strategy, the control parameters are automatically adjusted through reinforcement learning to obtain the inverse optimization result.

[0091] In the embodiment of the present invention, by determining the optimization target, such as improving stability, reducing energy consumption and handling abnormal situations, the setting of the optimization target helps to comprehensively improve the performance of the headlight, including improving lighting effects, extending service life and reducing energy consumption; by optimizing the stability of the headlight and reducing abnormal situations, according to the control parameters and optimization targets of the headlight, a reverse optimization strategy is formulated, so that a more accurate optimization effect can be achieved to meet specific needs. The reverse optimization strategy helps to shorten the optimization cycle, reduce the number of trials and errors, and improve the optimization efficiency. By formulating the reverse optimization strategy, it is possible to better adapt to different use environments and conditions, and improve the adaptability and reliability of the headlight. By automatically adjusting the control parameters through reinforcement learning, the intelligent control of the headlight system can be realized, and the control accuracy and response speed can be improved. The control parameters of the automatic headlight can optimize the lighting effect according to the actual situation and improve driving safety and comfort. By automatically adjusting the control parameters, energy can be used more effectively, the energy consumption of the headlight can be reduced, and the requirements of energy saving and environmental protection can be met.

[0092] In a specific embodiment of the present invention, the identified problems are classified and prioritized, and based on the analysis, the optimization goals are clarified, such as improving the stability of the headlights to reduce brightness fluctuations, reducing energy consumption to save energy, and promptly handling abnormal situations to prevent; setting specific quantitative indicators, such as the percentage of stability improvement, the target value of energy consumption reduction, and the response time of abnormal situations. The control parameters of the headlights are obtained, including the initial brightness setting, flashing mode, operating voltage, and current. According to the optimization goal, the reverse optimization strategy is designed, that is, how to achieve the desired optimization effect by adjusting the control parameters, analyzing the interaction between the parameters and possible constraints, such as voltage and current limitations, and the trade-off between brightness and energy consumption; formulating a specific adjustment plan, including the range of parameter adjustment, step size, and expected adjustment effect. According to the complexity of the problem and the optimization goal, a reinforcement learning algorithm is selected, and the state, action, and reward are obtained through the formula, and the control parameters are automatically adjusted to achieve the reverse optimization goal, and the performance indicators during the adjustment process, such as stability, energy consumption, and abnormal situations, are monitored.

[0093] In a preferred embodiment of the present invention, the above step 143 may include:

[0094] Step 1431, according to the reverse optimization strategy, determine the state space and action space, the state space includes the ambient light intensity, vehicle speed, current brightness, current flashing mode, current energy consumption and abnormal state, and the action space includes parameters for setting initial brightness, initial brightness and operating frequency;

[0095] Step 1432, according to the reverse optimization strategy, by Get the reward function, where is a reward function that evaluates the effect in a given state; , is the weight parameter; is the target brightness value, is the actual output brightness value, is the highest score for stability, with a value of 100; It represents the upper limit of energy consumption under ideal conditions, which is 200 watts; is the bonus score for successfully handling an exception, with a value of 50; is the current energy consumption value.

[0096] Step 1433, automatically adjust the control parameters, the control parameters of the headlights, according to the state space, the action space and the reward function to obtain the reverse optimization result.

[0097] In the embodiment of the present invention, by determining the state space, including ambient light intensity, vehicle speed, current brightness, current flickering mode, current energy consumption and abnormal state, various factors affecting the performance of the headlight can be comprehensively considered; by clarifying the action space, including setting the parameters of initial brightness, initial brightness and working frequency, accurate adjustment of the headlight control can be achieved; by using the reward function, the optimization effect under a given state can be quantified, and the weight parameter in the reward function can balance multiple optimization goals, such as brightness, stability and energy consumption, to ensure the comprehensiveness of the optimization result. By setting a reward score for successfully handling abnormal situations, abnormal situations can be handled more actively, and the reliability and safety of the headlight can be improved. According to the state space, action space and reward function, the control parameters can be automatically adjusted to achieve intelligent optimization of the headlight and improve the optimization efficiency and accuracy. Automatic adjustment of control parameters can dynamically adapt to changes in the environment and use conditions to ensure that the headlight is always in the best state. By comprehensively optimizing multiple aspects such as brightness, flickering mode, energy consumption and stability, the comprehensive performance of the headlight can be comprehensively improved.

[0098] In a specific embodiment of the present invention, the brightness of the surrounding environment is measured in real time to obtain the current speed of the vehicle, record the current brightness setting of the headlights, record the current flashing mode of the headlights, monitor the current energy consumption of the headlights, and detect whether the headlights are abnormal. According to factors such as ambient light intensity and vehicle speed, the initial brightness of the headlights is set, and the initial brightness of the headlights and the flashing or changing working frequency are adjusted to adapt to different driving conditions and optimization goals. The initial control parameters are set, including initial brightness, flashing mode and working frequency, and the ambient light intensity, vehicle speed, current brightness, flashing mode, energy consumption and abnormal state are monitored in real time through sensors; according to the current state and action, the reward value is calculated using the reward function, and the control parameters are automatically adjusted according to the reward value, including increasing or decreasing the brightness, changing the flashing mode, and adjusting the working frequency; when an abnormal state is detected, timely processing is carried out according to the preset strategy, such as reducing the brightness and turning off the headlights, so as to achieve reverse optimization of the headlight control parameters.

[0099] In a preferred embodiment of the present invention, the above step 15 may include:

[0100] Step 151, collecting optimization data and comparing optimization targets according to the reverse optimization results to extract key parameters;

[0101] Step 152, setting the control logic of the vehicle light according to the key parameters, and generating a control instruction according to the control logic.

[0102] In an embodiment of the present invention, by collecting data generated during the reverse optimization process and performing comparative analysis, the optimization effect can be evaluated based on actual data, and by comparing the data before and after the optimization, the key parameters that have the greatest impact on the optimization target can be accurately located. According to the key parameters, the control logic of the headlights is set, and control instructions are generated according to the control logic. Based on the control logic set by the key parameters, intelligent control of the headlights can be achieved, so that the headlights can automatically adjust according to changes in the environment and usage conditions, thereby improving the lighting effect. Through the generated control instructions, the state of the headlights can be quickly adjusted, thereby improving the response speed and flexibility of the headlights. The setting of the control logic and instructions can comprehensively consider multiple optimization targets such as energy consumption and stability, ensuring that the headlights can maintain low energy consumption and high stability while operating efficiently.

[0103] In a specific embodiment of the present invention, during the reverse optimization process, the status data of the vehicle lights and the changes in control parameters are recorded in real time, and the data include brightness values, energy consumption, flashing modes, operating voltage and current before and after optimization; the optimized data are compared with the set optimization goals, such as the percentage of stability improvement and the target value of energy consumption reduction; based on the comparison results, the control parameters that have the greatest impact on the optimization goals are identified; the values ​​or ranges of key parameters are extracted, and the control logic of the vehicle lights is designed based on the extracted key parameters. The control logic generates control instructions based on the requirements under different usage scenarios and conditions, such as the brightness difference between day and night, the flashing modes under different road conditions, etc.

[0104] like Figure 2 As shown, an embodiment of the present invention further provides a vehicle light control data processing system 20 based on cloud computing, comprising:

[0105] The acquisition module 21 is used to acquire the status data of the vehicle lamp, including the on or off status, brightness, flashing frequency and energy consumption of the vehicle lamp; clean, classify, aggregate and analyze the status data to obtain processed status data; and identify the stability, energy consumption change and abnormal situation of the vehicle lamp through preset rules according to the processed status data to obtain the identification result;

[0106] The processing module 22 is used to reversely optimize the control logic of the headlights according to the recognition results and the control parameters of the headlights to obtain reverse optimization results, wherein the control parameters include initial brightness settings and flashing modes; generate optimized headlight control parameters and instructions according to the reverse optimization results; and control the headlights according to the optimized headlight control parameters and instructions. It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0107] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0108] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0109] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A vehicle light control data processing method based on cloud computing, characterized in that: The method comprises: Get the status data of the lights, including whether they are on or off, brightness, flashing frequency, and energy consumption; Clean, classify, aggregate and analyze the status data to obtain processed status data; According to the processed status data, the stability, energy consumption changes and abnormal conditions of the vehicle lights are identified through preset rules to obtain the identification results; According to the recognition result and the control parameters of the vehicle light, the control logic of the vehicle light is reversely optimized to obtain a reverse optimization result, wherein the control parameters include an initial brightness setting and a flashing mode; Generate optimized vehicle light control parameters and instructions based on the reverse optimization results; The headlights are controlled according to the optimized headlight control parameters and instructions; the headlight control logic is reversely optimized according to the recognition results and the headlight control parameters to obtain reverse optimization results, and the control parameters include initial brightness settings and flashing modes, including: Based on the identification results, determine the optimization goals, including: improving stability, reducing energy consumption and handling abnormal situations; According to the control parameters and optimization objectives of the headlights, a reverse optimization strategy is obtained; According to the inverse optimization strategy, the control parameters are automatically adjusted through reinforcement learning to obtain the inverse optimization results; According to the inverse optimization strategy, the control parameters are automatically adjusted through reinforcement learning to obtain the inverse optimization results, including: According to the reverse optimization strategy, the state space and action space are determined. The state space includes the ambient light intensity, vehicle speed, current brightness, current flashing mode, current energy consumption and abnormal state, and the action space includes the parameters for setting the initial brightness, initial brightness and working frequency; According to the reverse optimization strategy, Get the reward function, where is a reward function that evaluates the effect in a given state; , is the weight parameter; is the target brightness value, is the actual output brightness value, is the highest score for stability, with a value of 100; It represents the upper limit of energy consumption under ideal conditions, which is 200 watts; is the bonus score for successfully handling an exception, with a value of 50; is the current energy consumption value; According to the state space, action space and reward function, the control parameters and control parameters of the headlights are automatically adjusted to obtain the inverse optimization results.

2. The vehicle light control data processing method based on cloud computing according to claim 1, characterized in that: Clean, classify, aggregate and analyze the status data to obtain processed status data, including: By checking whether there are duplicate records in the status data set, and cleaning the missing values ​​and abnormal values ​​according to the inspection results, the cleaned data is obtained; Classify the cleaned data according to the type and status of the headlights to obtain classified data; Aggregate the classified data according to the time window and state, and extract features from the aggregated data to obtain aggregated and feature-extracted data; Data analysis is performed on the aggregated and feature extracted data to obtain processed status data.

3. The vehicle light control data processing method based on cloud computing according to claim 2, characterized in that: Aggregate the classified data according to the time window and state, and extract features from the aggregated data to obtain aggregated and feature-extracted data, including: According to the time window and state, the classified data is passed Aggregate to obtain aggregate data, where is in the time window t Three aggregate features within, including mean, sum, and count; Indicates that in the time window t The state inside is s The average value of Indicates that in the time window t The state inside is s The number of valid samples of Indicates that in the time window t The state inside is s No. i The value of the record, Indicates that in the time window t The state inside is s The sum of Indicates that in the time window t The state inside is s The number of records reflects the number of occurrences of the state in the time window; s is the "on" or "off" state, t is the timestamp; Extract the average, sum, count, and state duration from the aggregated data to get the aggregated and feature extracted data.

4. The vehicle light control data processing method based on cloud computing according to claim 3 is characterized in that: Perform data analysis on the aggregated and feature-extracted data to obtain processed status data, including: Arrange the aggregated data in time series, draw a time series graph, and analyze the changing trend of the state data in different time windows to obtain the changing trend analysis results; Get the average, sum, and count features by Calculate the correlation analysis of the features to obtain the association analysis results, where: Indicates that in the time window t The state inside is s The average value of Indicates that in the time window t The state inside is s The overall average value of Indicates that in the time window t The state inside is s The sum of Indicates that the state in time window t is s The overall mean of the sum of is the correlation coefficient; pass Identify outliers of feature data to obtain outlier detection results, where: Indicates that in the time window t The state inside is s The average value of is the feature data point to be detected. represents the total number of samples in the dataset, and is the index, Indicates the standard deviation unit difference between the current feature data point to be detected and the sample mean; The analysis and processing results are obtained based on the change trend analysis results, correlation analysis results and outlier detection results.

5. The vehicle light control data processing method based on cloud computing according to claim 4 is characterized in that: Based on the processed status data, the stability, energy consumption changes and abnormal conditions of the headlights are identified through preset rules to obtain identification results, including: According to the processed status data, the threshold of brightness change, the normal range of energy consumption and the energy consumption trend analysis are set to obtain the preset rules; According to preset rules, the stability, energy consumption changes and abnormal conditions of the vehicle lights are identified to obtain the identification results.

6. A vehicle light control data processing system based on cloud computing, characterized in that: include: An acquisition module is used to obtain the status data of the vehicle lights, including the on or off status, brightness, flashing frequency and energy consumption of the vehicle lights; Clean, classify, aggregate and analyze the status data to obtain processed status data; identify the stability, energy consumption changes and abnormal conditions of the headlights through preset rules based on the processed status data to obtain identification results; The processing module is used to reversely optimize the control logic of the headlight according to the recognition result and the control parameters of the headlight to obtain a reverse optimization result, wherein the control parameters include an initial brightness setting and a flashing mode; generate optimized headlight control parameters and instructions according to the reverse optimization result; control the headlight according to the optimized headlight control parameters and instructions; reversely optimize the control logic of the headlight according to the recognition result and the control parameters of the headlight to obtain a reverse optimization result, wherein the control parameters include an initial brightness setting and a flashing mode, including: Based on the identification results, determine the optimization goals, including: improving stability, reducing energy consumption and handling abnormal situations; According to the control parameters and optimization objectives of the headlights, a reverse optimization strategy is obtained; According to the inverse optimization strategy, the control parameters are automatically adjusted through reinforcement learning to obtain the inverse optimization results; According to the inverse optimization strategy, the control parameters are automatically adjusted through reinforcement learning to obtain the inverse optimization results, including: According to the reverse optimization strategy, the state space and action space are determined. The state space includes the ambient light intensity, vehicle speed, current brightness, current flashing mode, current energy consumption and abnormal state, and the action space includes the parameters for setting the initial brightness, initial brightness and working frequency; According to the reverse optimization strategy, Get the reward function, where is a reward function that evaluates the effect in a given state; , is the weight parameter; is the target brightness value, is the actual output brightness value, is the highest score for stability, with a value of 100; It represents the upper limit of energy consumption under ideal conditions, which is 200 watts; is the bonus score for successfully handling an exception, with a value of 50; is the current energy consumption value; According to the state space, action space and reward function, the control parameters and control parameters of the headlights are automatically adjusted to obtain the inverse optimization results.

7. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which implements the method according to any one of claims 1 to 5 when executed by a processor.

Images