Intelligent vehicle auditing method and system based on big data edge computing device

Abstract

The invention discloses an intelligent vehicle auditing method and system based on a big data edge computing device, and relates to the technical field of intelligent vehicles, and the method comprises the steps: collecting environment information and vehicle operation state data through a sensor on an intelligent vehicle, and carrying out the preprocessing of the data; based on the identified situation features, the fog node cooperates with a cloud database to generate an auditing strategy for a specific situation; the audit strategy instruction set is sent back to the intelligent vehicle, and the vehicle carries out strategy adjustment and feeds back effect data; and the fog node receives the feedback data to evaluate the strategy performance, optimizes the existing auditing strategy based on the evaluation result, and sends the optimized strategy to the intelligent vehicle. According to the method, the edge calculation and the cloud database are fused, the personalized strategy of the intelligent vehicle is optimized, the environmental adaptability, the driving safety and the decision-making efficiency are greatly improved, the risk is reduced, the innovation promotes the road safety and optimizes the traffic, the far-reaching positive influence on the economy and the environmental protection is achieved, and a new era of intelligent travel is led.

Description

Technical Field

[0001] The present invention relates to the field of intelligent vehicle technology, and in particular to an intelligent vehicle audit method and system based on big data edge computing equipment. Background Art

[0002] As an important part of future transportation, smart vehicles have been rapidly promoted in recent years with the rapid development of artificial intelligence, Internet of Things and big data technologies, especially in the fields of vehicle environment perception, path planning and decision control. With the help of advanced technologies such as sensor fusion, machine learning and edge computing, smart vehicles have been able to achieve autonomous driving in complex traffic environments. Edge computing, as an emerging technology, plays a vital role in the application of smart vehicles. It reduces latency and improves the real-time and security of data processing by processing data near the data source. However, despite significant progress in edge computing and smart vehicle technology, there are still some challenges and limitations.

[0003] The environmental perception and decision-making process of intelligent vehicles mainly relies on data collected by on-board sensors and auxiliary analysis in the cloud. However, traditional data processing methods often face problems such as data transmission delay, privacy protection and bandwidth limitations, which limit the response speed and safety of intelligent vehicles. In addition, due to the lack of effective data preprocessing and situational recognition algorithms, intelligent vehicles often find it difficult to accurately understand driving situations when facing complex and changeable traffic environments, thus affecting the accuracy and timeliness of decision-making. Especially in the context of big data, how to efficiently process and analyze massive amounts of vehicle operation status data has become one of the bottlenecks restricting the development of intelligent vehicles. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an intelligent vehicle audit method and system based on big data edge computing devices to solve the problems of slow response speed and inaccurate situational understanding of intelligent vehicles in a big data environment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides an intelligent vehicle audit method based on a big data edge computing device, which includes collecting environmental information and vehicle operating status data through sensors on the intelligent vehicle, and preprocessing the data; after the fog node receives the preprocessed data, it performs a preliminary analysis; based on the preliminary analysis results, the fog node performs a situational awareness analysis, combines external data sources to build a driving situation view, and uses a machine learning model to identify situational features; based on the identified situational features, the fog node collaborates with the cloud database to generate an audit strategy for a specific situation; the audit strategy instruction set is sent back to the intelligent vehicle, the vehicle implements the strategy adjustment and feeds back the effect data; the fog node receives the feedback data to evaluate the strategy performance, optimizes the existing audit strategy based on the evaluation results, and sends the optimized strategy to the intelligent vehicle.

[0008] As a preferred solution of the intelligent vehicle audit method based on big data edge computing equipment described in the present invention, wherein: the environmental information and vehicle operation status data are collected by sensors on the intelligent vehicle, and the data is pre-processed, and the specific steps are as follows:

[0009] The high-precision sensors equipped in smart vehicles collect environmental information and vehicle operating status data, and use Z-score to identify and remove measurements that are significantly deviated from the normal range;

[0010] For missing data caused by signal interference and hardware failure, the K-nearest neighbor algorithm is used to predict missing values;

[0011] Use UTC time to align all raw data collected by sensors to the same time base, so that the time of data across sensors remains consistent;

[0012] Convert the encoding format of the original data to UTF-8 to improve data compatibility, and apply the maximum and minimum scaling method to scale the data to the specified range.

[0013] As a preferred solution of the intelligent vehicle audit method based on big data edge computing equipment described in the present invention, the fog node performs preliminary analysis after receiving the pre-processed data. The specific steps are as follows:

[0014] Adaptive weighted fusion algorithm is used to dynamically adjust the weight of pre-processed data for data fusion. The expression is:

[0015]

[0016] Among them, W i represents the weight of the i-th sensor data, (σi) 2 is the variance of the sensor measurement error, d i is the distance between the sensor and the target;

[0017] Define the environment complexity C, and use the adaptive weighted fusion data to evaluate the complexity of the current environment to guide the depth of subsequent data analysis and resource allocation. The expression is:

[0018]

[0019] Where n is the total number of sensors, Ni is the number of targets detected by the i-th sensor, and N ag is the average number of targets detected by all sensors;

[0020] Define the risk index R, based on the complexity of the environment and the vehicle status, and the expression is:

[0021]

[0022] Where v is the vehicle speed, σ 2 is the standard deviation of velocity, α is the absolute value of acceleration, and A is the average value of acceleration;

[0023] According to the risk index R, the fog node generates decision recommendations and feeds back the analysis results to the vehicle control system and cloud server to optimize global traffic management and vehicle autonomous decision-making.

[0024] As a preferred solution of the intelligent vehicle audit method based on big data edge computing equipment described in the present invention, wherein: the fog node performs situational awareness analysis based on the preliminary analysis results, builds a driving situation view in combination with external data sources, and uses a machine learning model to identify situational features. The specific steps are as follows:

[0025] Based on the preprocessed sensor data and preliminary analysis results, including the environmental complexity C and the risk index R, the fog node further integrates external data sources to form a comprehensive driving scenario score J, which is expressed as:

[0026] J=w W W t +w F F t +w S S t +C+R;

[0027] Among them, w w is the weight coefficient of weather conditions, w F is the traffic flow weight coefficient, w s is the mass coefficient of road construction information, W t , which is a quantitative index of weather conditions, F t Represents the quantitative index of traffic flow, S t Refers to the impact degree of road construction information;

[0028] Convert the scenario score J into a specific scenario view to help the autonomous driving system better understand the current driving environment;

[0029] The multi-dimensional feature fusion network MFFN is used to extract the comprehensive feature vector. MFFN contains multiple branches. Each branch extracts features for a data type through convolutional neural network and long short-term memory network, and then merges them into a unified feature representation through the gating mechanism. The expression is:

[0030] F=MFFN(V 1 ,V 2 ,…,V m );

[0031] Among them, V i represents the input vector of the i-th data type, and F is the final scenario feature vector.

[0032] As a preferred solution of the intelligent vehicle audit method based on big data edge computing equipment described in the present invention, the fog node cooperates with the cloud database based on the identified situational features to generate an audit strategy for a specific situation. The specific steps are as follows:

[0033] The feature enhancement function E is introduced to enhance the scene feature vector F, and the expression is:

[0034] F'=E(F,J,t,l,w);

[0035] Among them, F' is the enhanced scene feature vector, t is the current timestamp, and l is the vehicle position coordinate;

[0036] By designing the situational relationship matrix A, the mutual influence between different situational characteristics is evaluated, and the expression is:

[0037]

[0038] Among them, a ik is the correlation coefficient of the i-th situation feature in the k-th sensor data, A ij represents the correlation strength between the i-th and j-th situational features, m is the dimension of the vector, and α ik is the vector α i The kth element of jk is the vector α j The kth element of ;

[0039] Based on the situational association matrix A and the enhanced situational feature vector F', the audit strategy generation function G is constructed, and the expression is:

[0040]

[0041] Among them, wi′ is the weight of the i-th situational feature, G is the score of the audit strategy, which reflects the pertinence and effectiveness of the strategy, and n represents the number of features;

[0042] Design a strategy adjustment function Φ to dynamically adjust the strategy based on the characteristics of the current situation. The expression is:

[0043]

[0044] Among them, Φ is the adjusted audit strategy score, β controls the influence of G on the output of the entire function, γ controls the influence of R on the output of the entire function, and δ controls the influence of C on the output of the entire function.

[0045] As a preferred solution of the intelligent vehicle audit method based on big data edge computing equipment described in the present invention, wherein: the audit strategy instruction set is sent back to the intelligent vehicle, the vehicle implements the strategy adjustment and feeds back the effect data, the specific steps are as follows:

[0046] Based on the audit strategy score Φ, the audit strategy is converted into an executable instruction set I, which is expressed as:

[0047] I=T(Φ,V s ,E);

[0048] Among them, T is the instruction conversion function, V s is the current state information of the vehicle, and E is the environmental information;

[0049] The wireless communication protocol Comp is used to send the instruction set I to the intelligent vehicle. The vehicle control system adjusts the vehicle behavior according to the received instruction set I and collects the effect data D after executing the strategy in real time through the vehicle's built-in monitoring system. e , the expression is:

[0050] D e =Q(I,V' s );

[0051] Among them, V s ' is the vehicle status information after the strategy is executed, Q is the effect data collection function;

[0052] The collected effect data D e , uploaded to the fog nodes and cloud database through the wireless communication protocol Comp.

[0053] As a preferred solution of the intelligent vehicle audit method based on big data edge computing equipment described in the present invention, wherein: the fog node receives feedback effect data to evaluate the strategy performance, optimizes the existing audit strategy based on the evaluation result, and sends the audit optimized strategy to the intelligent vehicle. The specific steps are as follows:

[0054] The fog node evaluates and analyzes the feedback effect data, and the expression is:

[0055]

[0056] Among them, Ev is the policy evaluation function, C s is the current policy configuration, ρ i is the evaluation weight of the i-th effect data, M is the total number of items of effect data, D ei represents the specific value of the i-th effect data;

[0057] Define a policy effectiveness satisfaction level X. By comparing the size of the policy evaluation function Ev and the policy effectiveness satisfaction level X, judge the effectiveness of the audit policy;

[0058] When Ev ≥ X, it indicates that the execution effect of the audit policy is good, meeting or exceeding the expected goal, and no immediate adjustment is required;

[0059] When Ev < X, it indicates that the audit policy performs poorly in the current situation and fails to achieve the expected effect. Analyze and optimize it, and send the optimized audit policy to the intelligent vehicle through the wireless communication protocol Comp.

[0060] In the second aspect, the present invention provides an intelligent vehicle audit system based on a big data edge computing device, including a data collection and preprocessing module, an edge computing node module, a context recognition module, and a policy execution and optimization module; the data collection and preprocessing module is used to collect raw data from the intelligent vehicle and implement data cleaning, format conversion, and preliminary anomaly detection; the edge computing node module is used to receive and process the preprocessed data from the intelligent vehicle and perform preliminary analysis to quickly screen key information. The context recognition module is used to identify the characteristics of the driving context based on the preliminary analysis results by applying machine learning algorithms; the policy execution and optimization module is used to adjust the policy in real time, optimize the operation of the intelligent vehicle, and ensure efficiency and safety.

[0061] In the third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, it implements any step of the intelligent vehicle audit method based on a big data edge computing device as described in the first aspect of the present invention.

[0062] In the fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, it implements any step of the intelligent vehicle audit method based on a big data edge computing device as described in the first aspect of the present invention.

[0063] The beneficial effects of the present invention are as follows: the content of the present invention performs preliminary analysis and situational perception through the fog nodes of edge computing, and generates and optimizes audit strategies with the support of cloud databases. This process not only significantly improves the adaptability and safety of smart vehicles to complex environments, but also greatly improves the accuracy and efficiency of driving decisions, reduces potential risks, and improves user experience. Through the deep integration of smart vehicles and advanced algorithms, the present invention has brought revolutionary changes to the field of modern transportation and opened a new chapter in smart travel. Its beneficial effects are reflected in enhancing road safety, optimizing traffic flow, reducing energy consumption and other aspects, and have a far-reaching positive impact on social and economic development and environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0065] Figure 1 This is a flow chart of the intelligent vehicle audit method based on big data edge computing equipment in Example 1.

[0066] Figure 2 This is a flow chart of the intelligent vehicle audit system based on big data edge computing equipment in Example 1. DETAILED DESCRIPTION

[0067] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0068] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0069] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0070] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides an intelligent vehicle audit method based on a big data edge computing device, comprising the following steps:

[0071] S1. Collect environmental information and vehicle operation status data through sensors on smart vehicles and pre-process the data.

[0072] Furthermore, the high-precision sensors equipped in smart vehicles, including but not limited to millimeter-wave radar, laser radar (LiDAR), visual cameras, ultrasonic sensors, GPS positioning modules, accelerometers, and gyroscopes, are used to collect environmental information and vehicle operation status data;

[0073] The Z-score is used to identify and remove measurements that are significantly deviated from the normal range. The expression is:

[0074]

[0075] Where X is the observed value, μ is the mean, σ is the standard deviation, and if |Z|>3, the value is considered an outlier;

[0076] For missing data caused by signal interference and hardware failure, the nearest neighbor algorithm is used to predict the missing values;

[0077] Use UTC time to align all raw data collected by sensors to the same time base, so that the time of data across sensors remains consistent;

[0078] Convert the encoding format of the original data to UTF-8 to improve data compatibility, and apply the maximum and minimum scaling method to scale the data to the specified range.

[0079] It should also be noted that in an intelligent vehicle system, different sensors may collect data at different time points. In order to ensure the accuracy of data analysis, it is necessary to align the data of all sensors to a unified time base so that the data can be correctly associated and analyzed;

[0080] UTC is an accurate and widely accepted time standard used for global time synchronization. Using UTC time as the time reference can ensure global consistency of data and facilitate data exchange and comparison between different regions and systems. Usually, the central processing unit of a smart vehicle records the local timestamp of each sensor data and converts it to UTC time, which may involve the calculation of time offset and time zone conversion to ensure that all data is marked and stored according to UTC time.

[0081] S2. After receiving the preprocessed data, the fog node performs preliminary analysis.

[0082] Furthermore, an adaptive weighted fusion algorithm is used to dynamically adjust the weight of the preprocessed data for data fusion. The expression is:

[0083]

[0084] Among them, W i represents the weight of the i-th sensor data, (σi) 2 is the variance of the sensor measurement error, d i is the distance between the sensor and the target;

[0085] Define the environmental complexity C. Using the data after adaptive weighted fusion, evaluate the complexity of the current environment to guide the depth of subsequent data analysis and resource allocation. The expression is:

[0086]

[0087] Among them, n is the total number of sensors, N i is the number of targets detected by the i-th sensor, N ag is the average number of targets detected by all sensors;

[0088] Define the risk index R. Based on the environmental complexity and vehicle state, the expression is:

[0089]

[0090] Among them, v is the vehicle speed, σ2 is the standard deviation of the speed, α is the absolute value of the acceleration, and A is the average value of the acceleration;

[0091] According to the risk index R, the fog node generates decision suggestions, such as adjusting the vehicle speed, starting the warning system or recommending a driving route. At the same time, the analysis results are fed back to the vehicle control system and the cloud server for optimizing global traffic management and vehicle autonomous decision-making.

[0092] It should also be noted that by introducing the adaptive weighted fusion algorithm and the complexity evaluation formula, the refined management of the intelligent vehicle environment perception is realized. The value range of the risk index R is [0, ∞), where R < 1 indicates low risk and is suitable for normal driving; 1 < R < 10 indicates medium risk and requires cautious driving, and R ≥ 10 indicates high risk and preventive measures should be taken immediately.

[0093] S3. Through the preliminary analysis results, the fog node conducts scenario perception analysis, constructs a driving scenario view in combination with external data sources, and uses a machine learning model to identify scenario features.

[0094] Furthermore, based on the preprocessed sensor data and preliminary analysis results, including the environmental complexity C and the risk index R, the fog node further integrates external data sources, such as weather forecasts, traffic flow, road construction information, etc., to form a comprehensive driving scenario score J. The expression is:

[0095] J = w W Wt +w F F t +w S S t +C+R;

[0096] Among them, w w is the weight coefficient of weather conditions, w F is the traffic flow weight coefficient, w s is the mass coefficient of road construction information, W t , which is a quantitative index of weather conditions, F t Represents the quantitative index of traffic flow, S t Refers to the impact degree of road construction information;

[0097] Convert the scenario score J into a specific scenario view, such as "moderate rain, low visibility, slight traffic congestion, no road construction, medium environmental complexity, and high risk index", to help the autonomous driving system better understand the current driving environment;

[0098] The multi-dimensional feature fusion network MFFN is used to extract the comprehensive feature vector. MFFN contains multiple branches. Each branch extracts features for a data type (such as vision, radar, GPS, etc.) through convolutional neural networks and long short-term memory networks, and then merges them into a unified feature representation through a gating mechanism. The expression is:

[0099] F=MFFN(V 1 ,V 2 ,…,V m );

[0100] Among them, V i represents the input vector of the i-th data type, and F is the final scenario feature vector.

[0101] It should also be noted that in intelligent vehicle technology, the multimodal feature fusion network (MFFN) plays a vital role. It improves the situational awareness of the autonomous driving system by integrating data from various sensors. The MFFN is designed as a multi-branch structure, and each branch specializes in processing one type of data, such as visual images, radar signals, GPS positioning, etc. These branches use specialized data processing techniques, such as convolutional neural networks (CNN) for image data to extract visual features, and long short-term memory networks for processing sequence data such as time series radar signals. The features extracted by each branch are finally synthesized into a comprehensive feature vector through a gating mechanism. This mechanism not only effectively combines information from different data sources, but also optimizes the combination of features through the regulation of gating, so that the final feature vector can comprehensively and accurately characterize the current driving environment, thereby greatly enhancing the autonomous driving system's ability to understand and judge the environment.

[0102] S4. Based on the identified situational features, the fog nodes collaborate with the cloud database to generate audit strategies for specific situations.

[0103] Furthermore, a feature enhancement function E is introduced to enhance the scenario feature vector F so that it contains richer driving environment details. This function takes into account factors such as time, location, and weather changes. The expression is:

[0104] F'=E(F,J,t,l,w);

[0105] Among them, F' is the enhanced scene feature vector, t is the current timestamp, and l is the vehicle position coordinate;

[0106] By designing the situational relationship matrix A, the mutual influence between different situational characteristics is evaluated, and the expression is:

[0107]

[0108] Among them, a ik is the correlation coefficient of the i-th situation feature in the k-th sensor data, A ij represents the correlation strength between the i-th and j-th situational features, m is the dimension of the vector, and α ik is the vector α i The kth element of jk is the vector α j The kth element of ;

[0109] Based on the situational association matrix A and the enhanced situational feature vector F', the audit strategy generation function G is constructed, and the expression is:

[0110]

[0111] Among them, w i ′ is the personalized weight of the i-th situational feature, G is the score of the audit strategy, which reflects the pertinence and effectiveness of the strategy, and n represents the number of features;

[0112] Design a strategy adjustment function Φ to dynamically adjust the strategy based on the characteristics of the current situation. The expression is:

[0113]

[0114] Among them, Φ is the adjusted audit strategy score, β controls the influence of G on the output of the entire function, γ controls the influence of R on the output of the entire function, and δ controls the influence of C on the output of the entire function.

[0115] It should also be noted that in order to formulate an effective audit strategy, the scenario feature vector F is first enriched through the feature enhancement function E to include more detailed driving environment information, such as time, location and weather changes. Subsequently, the scenario association matrix A is used to evaluate the mutual influence between different scenario features. This matrix is ​​constructed by examining the correlation between the features, which helps to understand how different driving scenarios work together in the driving environment. Based on these enhanced features and scenario correlations, an audit strategy generation function G is constructed. This function combines the personalized weights of the scenario features to generate a score that reflects the strategy's targeting and effectiveness. Finally, a strategy adjustment function Φ is introduced to dynamically adjust the strategy according to the current environmental characteristics to ensure the real-time and adaptability of the strategy. This process ensures that the autonomous driving system can formulate and implement the most appropriate audit strategy for various complex driving scenarios through precise mathematical modeling and algorithm adjustment.

[0116] S5. Send the audit strategy instruction set back to the intelligent vehicle, and the vehicle implements strategy adjustments and feeds back effect data.

[0117] Furthermore, based on the audit strategy score Φ, the audit strategy is converted into an executable instruction set I, which is expressed as:

[0118] I=T(Φ,V s ,E);

[0119] Among them, T is the instruction conversion function, V s is the current state information of the vehicle, and E is the environmental information;

[0120] The wireless communication protocol Comp is used to send the instruction set I to the intelligent vehicle. The vehicle control system adjusts the vehicle behavior, such as speed control and route planning, based on the received instruction set I. The vehicle's built-in monitoring system collects the effect data D after the execution of the strategy in real time. e , the expression is:

[0121] D e =Q(I,V' s );

[0122] Among them, V s ' is the vehicle status information after the strategy is executed, Q is the effect data collection function;

[0123] The collected effect data De is uploaded to the fog nodes and the cloud database through the wireless communication protocol Comp.

[0124] It should also be noted that in view of the security and real-time requirements of instruction set transmission, an optimized wireless communication protocol Comp is adopted, which combines the advantages of 5G and satellite communications. The expression is:

[0125]

[0126] where RT i is the round-trip time of the i-th communication path, and ω i is the reliability weight of path i, and N is the total number of available communication paths;

[0127] The instruction set I is sent to the intelligent vehicle through the optimized wireless communication protocol Comp, and after the vehicle receives it, it sends a reception confirmation signal ASK. The expression is:

[0128] ACK = r(I, Comp);

[0129] where r is the reception confirmation function;

[0130] S6. The fog node receives the feedback effect data to evaluate the policy performance. Based on the evaluation results, the existing auditing policy is optimized, and the optimized auditing policy is sent to the intelligent vehicle.

[0131] Furthermore, the fog node evaluates and analyzes the feedback effect data. The expression is:

[0132]

[0133] where Ev is the policy evaluation function, and C s is the current policy configuration, ρ i is the evaluation weight of the i-th effect data, M is the total number of effect data items, and D ei represents the specific value of the i-th effect data;

[0134] Define a policy effectiveness satisfaction level X. By comparing the magnitudes of the policy evaluation function Ev and the policy effectiveness satisfaction level X, the effectiveness of the auditing policy is judged;

[0135] When Ev ≥ X, it indicates that the execution effect of the auditing policy is good, meeting or exceeding the expected goal, and no immediate adjustment is required;

[0136] When Ev < X, it indicates that the auditing policy performs poorly in the current situation and fails to achieve the expected effect. Analyze and optimize it, and send the optimized auditing policy to the intelligent vehicle through the wireless communication protocol Comp.

[0137] It should also be noted that the fog node receives the effect data returned from the vehicle to evaluate the performance of the current auditing policy. Through the policy evaluation function Ev, it combines the current policy configuration Cs, the weights ρ i of each item of effect data ei, calculate a quantitative evaluation result. If this result reaches or exceeds the preset strategy effectiveness satisfaction level X, it is considered that the strategy execution effect is good and no adjustment is required; if it does not reach X, it indicates that the strategy needs to be optimized. The fog node will adjust and optimize the strategy based on these analysis results, and then send the optimized strategy back to the smart vehicle through the wireless communication protocol Comp to ensure that the vehicle can adapt to the changing driving environment and improve driving safety and efficiency. This process ensures that the audit strategy can be adjusted in real time according to the actual effect data, thereby improving the adaptability and overall performance of the smart vehicle system.

[0138] This embodiment also provides an intelligent vehicle audit system based on big data edge computing devices, including:

[0139] Data collection and preprocessing module, edge computing node module, situation recognition module, strategy execution and optimization module; the data collection and preprocessing module is used to collect raw data from smart vehicles, and realize data cleaning, format conversion and preliminary anomaly detection; the edge computing node module is used to receive and process preprocessed data from smart vehicles, perform preliminary analysis, and quickly screen key information. The situation recognition module is used to identify the characteristics of driving situations based on preliminary analysis results and apply machine learning algorithms; the strategy execution and optimization module is used to adjust strategies in real time, optimize the operation of smart vehicles, and ensure efficiency and safety.

[0140] This embodiment also provides a computer device, which is suitable for the case of an intelligent vehicle audit method based on a big data edge computing device, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the intelligent vehicle audit method based on a big data edge computing device as proposed in the above embodiment.

[0141] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0142] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the intelligent vehicle audit method based on a big data edge computing device proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0143] In summary, the present invention achieves data quality improvement and ensures a solid analysis basis through: efficient collection and preprocessing of intelligent vehicle sensor data, preliminary analysis stage, dynamic data fusion and environmental risk assessment, generating instant decision-making suggestions, enhancing driving safety, situational awareness analysis combined with external data, using MFFN to extract comprehensive features, deepening the system's understanding of the environment, improving decision-making intelligence, auditing strategy generation and dynamic adjustment, ensuring strategy targeting and real-time, adapting to complex situations, closed-loop feedback mechanism to optimize strategy performance, continuously improving the environmental adaptability and driving safety of intelligent vehicles, and promoting the efficiency and safety of intelligent transportation systems.

[0144] Example 2, referring to Table 1, is the second example of the present invention. In order to further verify the advancement of the present invention, experimental simulation data of the intelligent vehicle audit method based on big data edge computing equipment is provided.

[0145] Four smart vehicles of different models were selected, namely smart vehicle A, smart vehicle B, smart vehicle C and smart vehicle D. Each smart vehicle was independently tested under the preset conditions of environmental complexity C, risk index R and situational score J.

[0146] First, define the objectives and hypotheses of the experiment, clarify the questions you want to solve or the concepts you want to verify through the experiment. This step ensures the direction and effectiveness of the experimental design.

[0147] Secondly, four representative intelligent vehicles, A, B, C, and D, were selected, and the basic configuration and characteristics of each vehicle were recorded in detail to ensure the comparability and accuracy of the experimental results.

[0148] Next, different experimental conditions were set, including environment complexity C, risk index R, etc., to simulate various situations that smart vehicles may encounter in the real world. Each smart car was tested independently under these preset conditions.

[0149] Finally, the collected data is analyzed to compare the performance of different vehicles under various conditions, the effectiveness of personalized strategy adjustments is evaluated, and suggestions for improving the performance and strategies of intelligent vehicles are proposed based on the experimental results.

[0150] The details are shown in Table 1:

[0151] Table 1 Experimental record table

[0152]

[0153] Through the analysis of the data in the above table, it can be clearly seen that the effectiveness and superiority of the method of the present invention. By integrating advanced sensor technology, deep learning algorithms and personalized strategy adjustment mechanisms, the present invention effectively improves the decision-making ability and driving safety of smart cars in complex environments. For example, by comparing the data of smart vehicles A and B, it can be clearly seen that when the environmental complexity C is high, vehicle B, when the risk index R (4.1) is relatively high, still maintains a high situation score J (7.9) and strategy execution effect score Ev (9.0), which fully reflects the remarkable results of the present invention in improving the ability of smart cars to adapt to complex environments.

[0154] The intelligent vehicle audit method based on big data edge computing equipment of the present invention shows significant innovation and advantages in intelligent vehicle environmental perception and decision optimization, effectively improving the driving safety and efficiency of vehicles in complex environments. Compared with the existing technology, the present invention can more accurately assess environmental risks and intelligently adjust driving strategies, thus providing strong technical support for the future development of intelligent transportation systems.

[0155] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent vehicle audit method based on big data edge computing equipment, characterized in that: include, Collect environmental information and vehicle operating status data through sensors on smart vehicles and pre-process the data; After receiving the preprocessed data, the fog node performs preliminary analysis; Based on the preliminary analysis results, the fog node conducts situational awareness analysis, builds a driving situation view in combination with external data sources, and uses machine learning models to identify situational features; Based on the identified situational features, the fog nodes collaborate with the cloud database to generate audit strategies for specific situations; The audit strategy instruction set is sent back to the intelligent vehicle, and the vehicle implements the strategy adjustment and provides feedback on the effect data; The fog node receives feedback data to evaluate the strategy performance, optimizes the existing audit strategy based on the evaluation results, and sends the optimized strategy to the smart vehicle.

2. The intelligent vehicle audit method based on big data edge computing equipment according to claim 1, characterized in that: The environmental information and vehicle operation status data are collected through sensors on the smart vehicle, and the data is preprocessed. The specific steps are as follows: Collect environmental information and vehicle operation status data through high-precision sensors equipped on smart vehicles; Use Z-score to identify and remove measurements that are significantly out of the normal range. For missing data caused by signal interference and hardware failure, use the nearest neighbor algorithm to predict missing values. Use UTC time to align all raw data collected by sensors to the same time base, so that the time of data across sensors remains consistent; Convert the encoding format of the original data to UTF-8 to improve data compatibility, and apply the maximum and minimum scaling method to scale the data to the specified range.

3. The intelligent vehicle audit method based on big data edge computing equipment according to claim 2, characterized in that: After receiving the preprocessed data, the fog node performs preliminary analysis. The specific steps are as follows: Adaptive weighted fusion algorithm is used to dynamically adjust the weight of pre-processed data for data fusion. The expression is: Among them, W i represents the weight of the i-th sensor data, (σ i )2 is the variance of the sensor measurement error, d i is the distance between the sensor and the target; Define the environment complexity C, and use the adaptive weighted fusion data to evaluate the complexity of the current environment to guide the depth of subsequent data analysis and resource allocation. The expression is: Where n is the total number of sensors, N i is the number of targets detected by the i-th sensor, N ag is the average number of targets detected by all sensors; Define the risk index R, based on the complexity of the environment and the vehicle status, and the expression is: Where v is the vehicle speed, σ2 is the standard deviation of the speed, α is the absolute value of the acceleration, and A is the average value of the acceleration; According to the risk index R, the fog node generates decision recommendations and feeds back the analysis results to the vehicle control system and cloud server to optimize global traffic management and vehicle autonomous decision-making.

4. The intelligent vehicle audit method based on big data edge computing equipment according to claim 3 is characterized in that: Based on the preliminary analysis results, the fog node performs situational awareness analysis, combines external data sources to build a driving situation view, and uses a machine learning model to identify situational features. The specific steps are as follows: Based on the preprocessed sensor data and preliminary analysis results, including the environmental complexity C and the risk index R, the fog node further integrates external data sources to form a comprehensive driving scenario score J, which is expressed as: J=w W W t +w F F t +w S S t +C+R; Among them, w w is the weight coefficient of weather conditions, w F is the traffic flow weight coefficient, w s is the mass coefficient of road construction information, W t , which is a quantitative index of weather conditions, F t represents the quantitative index of traffic flow, St refers to the impact of road construction information; Convert the scenario score J into a specific scenario view to help the autonomous driving system better understand the current driving environment; The multi-dimensional feature fusion network MFFN is used to extract the comprehensive feature vector. MFFN contains multiple branches. Each branch extracts features for a data type through a convolutional neural network and a long short-term memory network, and then merges them into a unified feature representation through a gating mechanism. The expression is: F=MFFN(V1,V2,…,V m ); Among them, V i represents the input vector of the i-th data type, and F is the final scenario feature vector.

5. The intelligent vehicle audit method based on big data edge computing equipment according to claim 4, characterized in that: Based on the identified situational features, the fog node collaborates with the cloud database to generate an audit strategy for a specific situation. The specific steps are as follows: The feature enhancement function E is introduced to enhance the scene feature vector F, and the expression is: F'=E(F,J,t,l,w); Among them, F' is the enhanced scene feature vector, t is the current timestamp, and l is the vehicle position coordinate; By designing the situational relationship matrix A, the mutual influence between different situational characteristics is evaluated, and the expression is: Among them, a ik is the correlation coefficient of the i-th situation feature in the k-th sensor data, A ij represents the correlation strength between the i-th and j-th situational features, m is the dimension of the vector, and α ik is the vector α i The kth element of jk is the vector α j The kth element of ; Based on the situational association matrix A and the enhanced situational feature vector F', the audit strategy generation function G is constructed, and the expression is: Among them, w i ′ is the personalized weight of the i-th situational feature, G is the score of the audit strategy, which reflects the pertinence and effectiveness of the strategy, and n represents the number of features; Design a strategy adjustment function Φ to dynamically adjust the strategy based on the characteristics of the current situation. The expression is: Among them, Φ is the adjusted audit strategy score, β controls the influence of G on the output of the entire function, γ controls the influence of R on the output of the entire function, and δ controls the influence of C on the output of the entire function.

6. The intelligent vehicle audit method based on big data edge computing equipment according to claim 5, characterized in that: The audit strategy instruction set is sent back to the intelligent vehicle, and the vehicle implements the strategy adjustment and feeds back the effect data. The specific steps are as follows: Based on the audit strategy score Φ, the audit strategy is converted into an executable instruction set I, which is expressed as: I=T(Φ,Vs,E); Among them, T is the instruction conversion function, V s is the current state information of the vehicle, and E is the environmental information; The wireless communication protocol Comp is used to send the instruction set I to the intelligent vehicle. The vehicle control system adjusts the vehicle behavior according to the received instruction set I and collects the effect data D after executing the strategy in real time through the vehicle's built-in monitoring system. e , the expression is: D e =Q(I,V' s ); Among them, V s ' is the vehicle status information after the strategy is executed, Q is the effect data collection function; The collected effect data D e , uploaded to the fog nodes and cloud database through the wireless communication protocol Comp.

7. The intelligent vehicle audit method based on big data edge computing equipment according to claim 6, characterized in that: The fog node receives the feedback effect data to evaluate the strategy performance, optimizes the existing audit strategy based on the evaluation results, and sends the audit-optimized strategy to the smart vehicle. The specific steps are as follows: The fog node evaluates and analyzes the feedback effect data, and the expression is: Among them, Ev is the strategy evaluation function, C s is the current policy configuration, ρ i is the evaluation weight of the i-th effect data, M is the total number of effect data, D ei Represents the specific value of the i-th effect data; Define a strategy effectiveness satisfaction level X, and judge the effectiveness of the audit strategy by comparing the strategy evaluation function Ev and the strategy effectiveness satisfaction level X; When Ev ≥ X, it indicates that the audit strategy is being implemented well, meeting or exceeding the expected goals, and no immediate adjustment is required; If Ev<X, it indicates that the audit strategy performs poorly in the current situation and fails to achieve the expected results. Analysis and optimization are performed, and the optimized audit strategy is sent to the smart vehicle through the wireless communication protocol Comp.

8. An intelligent vehicle audit system based on big data edge computing equipment, based on the intelligent vehicle audit method based on big data edge computing equipment according to any one of claims 1 to 7, characterized in that: Including data collection and preprocessing module, edge computing node module, situation recognition module, strategy execution and optimization module; The data acquisition and preprocessing module is used to collect raw data from smart vehicles and implement data cleaning, format conversion and preliminary anomaly detection; The edge computing node module is used to receive and process pre-processed data from smart vehicles, perform preliminary analysis, and quickly filter key information. The situation recognition module is used to identify the characteristics of the driving situation based on the preliminary analysis results and apply machine learning algorithms; The strategy execution and optimization module is used to adjust strategies in real time, optimize the operation of intelligent vehicles, and ensure efficiency and safety.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent vehicle audit method based on big data edge computing equipment described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent vehicle audit method based on big data edge computing equipment described in any one of claims 1 to 7 are implemented.

Images