Intelligent automobile sensor data correction method and system based on safety monitoring

Abstract

The invention discloses an intelligent automobile sensor data correction method and system based on safety monitoring, and relates to the technical field of intelligent driving environment perception, and the method comprises the steps: carrying out the grading of safety feature codes based on a safety rule library, carrying out the mode matching through combining with a historical fault case library, and generating an abnormality diagnosis report of original monitoring data; a physical topology network of the intelligent automobile sensor is constructed, the node state of the physical topology network is dynamically adjusted according to the abnormity diagnosis report, and a physical topology network reconstruction instruction is obtained and executed; and correcting the original monitoring data twice based on the node state of the reconstructed physical topology network to generate vehicle safety monitoring data. According to the method, the weighted residual correction is combined with the iterative reweighted least square method, two-stage correction is performed on the monitoring data on the basis of the reconstructed node weight distribution table, progressive error compensation under the condition of dynamic performance degradation of the sensor network is realized, and the reliability and environmental adaptability of output data are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving environment perception technology, and in particular to a method and system for correcting intelligent automobile sensor data based on safety monitoring. Background Art

[0002] In the field of intelligent vehicle environmental perception, multi-sensor data fusion technology is a core component in ensuring driving safety. Current mainstream sensor data monitoring methods typically employ a technical approach that combines multi-source information acquisition, feature extraction, and state assessment. Typical approaches use sensors such as millimeter-wave radar and lidar to acquire raw data such as the target's spatial position, motion state, and reflectivity. After preprocessing through denoising, synchronization, and normalization, characteristic parameters such as the rate of change of point cloud density and signal-to-noise ratio fluctuations are extracted. Existing technologies generally employ a hierarchical analysis framework based on a rule base, classifying characteristic parameters using predefined safety thresholds and performing pattern matching based on historical failure cases to ultimately output anomaly diagnosis results.

[0003] Conventional methods still have two limitations in engineering practice: First, the dynamic adjustment mechanism of the physical topology network lacks a quantitative evaluation dimension, and node state switching relies on fixed threshold judgments, making it difficult to adapt to the gradual degradation of sensor performance under complex working conditions. In particular, in environmental interference scenarios such as rain and fog, static thresholds can easily lead to false isolation or downgraded delays. Second, the confidence assessment and spatiotemporal verification in the data correction process are coupled with defects. When transient anomalies occur simultaneously on multiple nodes, traditional sliding window statistical methods cannot distinguish between systematic deviations and random noise, resulting in residual errors in the data after secondary correction. Summary of the Invention

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

[0005] Therefore, the present invention provides a smart car sensor data correction method based on safety monitoring to solve the problems of insufficient dynamic adaptability and error coupling of sensor networks.

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

[0007] In a first aspect, the present invention provides a method for correcting sensor data for an intelligent vehicle based on safety monitoring, comprising collecting raw monitoring data through intelligent vehicle sensors and preprocessing the data, calculating the point cloud density mutation rate, the instantaneous decrease in signal-to-noise ratio, and the frequency of data continuity interruptions, and generating a safety feature code;

[0008] Based on the security rule library, the security feature codes are classified into different levels, and pattern matching is performed in combination with the historical fault case library to generate an abnormal diagnosis report of the original monitoring data;

[0009] Build the physical topology network of smart car sensors, dynamically adjust the node status of the physical topology network based on abnormal diagnosis reports, and obtain and execute physical topology network reconstruction instructions;

[0010] Based on the node status of the reconstructed physical topology network, the original monitoring data is corrected twice to generate vehicle safety monitoring data.

[0011] As a preferred solution of the intelligent automobile sensor data correction method based on safety monitoring described in the present invention, the original monitoring data includes the spatial position coordinates, motion speed value, electromagnetic wave reflection intensity value, three-dimensional point cloud distribution, optical reflectivity value, time synchronization mark, as well as close-range obstacle distance measurement value, vehicle acceleration value and angular velocity value of the target object.

[0012] As a preferred solution of the intelligent automobile sensor data correction method based on safety monitoring described in the present invention, the steps of calculating the point cloud density mutation rate, the instantaneous decrease in signal-to-noise ratio, and the frequency of data continuity interruption are as follows:

[0013] Based on the 3D point cloud distribution and time synchronization marking, the density mutation rate of the 3D point cloud distribution is calculated;

[0014] According to the density mutation rate of the three-dimensional point cloud distribution and the electromagnetic wave reflection intensity value, the instantaneous drop in the signal-to-noise ratio is calculated, and the frequency of statistical continuity interruption is analyzed in combination with the time synchronization mark statistics.

[0015] As a preferred solution of the intelligent automobile sensor data correction method based on safety monitoring of the present invention, wherein: the safety feature codes are classified into levels based on the safety rule library, pattern matching is performed in combination with the historical fault case library, and an abnormal diagnosis report of the original monitoring data is generated. The steps are as follows:

[0016] Based on the abnormal status of historical monitoring data, define abnormality judgment rules, associate and map the abnormality judgment rules with the risk level labels in the historical fault case library, and generate a safety rule library;

[0017] The security feature codes are classified into different levels based on the security rule base, and the fault types are matched from the historical fault case library according to the cosine similarity between the security feature codes and the historical fault feature vectors. An abnormal diagnosis report is generated and loaded into the historical fault case library.

[0018] As a preferred solution of the smart car sensor data correction method based on safety monitoring described in the present invention, the steps of constructing a physical topology network of smart car sensors and dynamically adjusting the node status of the physical topology network according to abnormal diagnosis reports are as follows:

[0019] Build a physical topology network of smart car sensors and dynamically adjust the node status of the physical topology network based on abnormal diagnosis reports;

[0020] Based on the risk level in the abnormal diagnosis report, the physical topology network nodes are marked for status, node control parameters are extracted and encapsulated, and node control instructions are generated;

[0021] Dynamically adjust the node status of the physical topology network according to node control instructions.

[0022] As a preferred solution of the smart car sensor data correction method based on safety monitoring described in the present invention, the obtaining and executing of physical topology network reconstruction instructions refers to collecting status feedback data of the adjusted physical topology network, predicting the health status score of each node in the physical topology network, and using the earliest deadline first algorithm to generate and execute the physical topology network reconstruction instructions of the smart car sensors.

[0023] As a preferred solution of the intelligent automobile sensor data correction method based on safety monitoring of the present invention, wherein: the original monitoring data is corrected twice based on the node status of the reconstructed physical topology network to generate vehicle safety monitoring data, the steps are as follows:

[0024] Based on the node status of the reconstructed physical topology network, the spatial position coordinates and weight distribution parameters of the normal state nodes are extracted, and the node weight distribution table of the physical topology network is generated by weighted least squares method;

[0025] Based on the node weight distribution table, the weighted residual correction method is used to calculate the deviation between the original monitoring data of each node and the weighted mean, and the original monitoring data is corrected for the first time according to the weight ratio compensation to generate the initial vehicle safety monitoring data;

[0026] Statistically analyze the confidence weights of each initial vehicle safety monitoring data, perform spatiotemporal consistency checks on the confidence weights, and mark invalid confidence weights;

[0027] The iterative reweighted least squares method is used to perform secondary correction on the initial vehicle safety monitoring data with invalid confidence weights to generate vehicle safety monitoring data.

[0028] As a preferred solution of the intelligent automobile sensor data correction method based on security monitoring of the present invention, the steps of generating the security feature code are as follows:

[0029] Identify abnormal conditions in raw monitoring data based on density mutation rate, instantaneous drop in signal-to-noise ratio, and frequency of data continuity interruption;

[0030] Through the binary bit field encoding method, the abnormal state of the original monitoring data is mapped into the security feature code.

[0031] As a preferred solution of the intelligent automobile sensor data correction method based on safety monitoring described in the present invention, the preprocessing includes normalization, filtering and denoising, time synchronization calibration and validity verification processing.

[0032] In a second aspect, the present invention provides a smart car sensor data correction system based on safety monitoring, comprising: a data acquisition module that collects raw monitoring data from smart car sensors and preprocesses it, calculates the point cloud density mutation rate, the instantaneous signal-to-noise ratio drop, and the frequency of data continuity interruption, and generates a safety feature code;

[0033] The abnormality diagnosis module classifies the security feature codes based on the security rule library, performs pattern matching based on the historical fault case library, and generates an abnormality diagnosis report for the original monitoring data;

[0034] The topology network adjustment module builds the physical topology network of smart car sensors, dynamically adjusts the node status of the physical topology network based on abnormal diagnosis reports, and obtains and executes physical topology network reconstruction instructions;

[0035] The data correction module corrects the original monitoring data twice based on the node status of the reconstructed physical topology network to generate vehicle safety monitoring data.

[0036] The beneficial effects of the present invention include: using multi-level threshold criteria to jointly analyze the point cloud density mutation rate, instantaneous signal-to-noise ratio drop, and data continuity interruption frequency, and generating structured safety feature codes through binary bit field coding, achieving accurate quantitative characterization of multi-dimensional abnormal states, effectively improving the accuracy and efficiency of fault feature identification under complex working conditions. By combining weighted residual correction with iterative reweighted least squares, a two-level correction of monitoring data is performed based on the reconstructed node weight distribution table, achieving progressive error compensation under conditions of dynamic performance degradation of the sensor network, and improving the reliability and environmental adaptability of the output data. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 The flowchart of the intelligent vehicle sensor data correction method based on safety monitoring.

[0039] Figure 2 Schematic diagram of the intelligent vehicle sensor data correction system based on safety monitoring.

[0040] Figure 3 Flowchart generated for security feature coding.

[0041] Figure 4 Flowchart for secondary correction of vehicle safety monitoring data. DETAILED DESCRIPTION

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0043] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. 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.

[0044] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0045] Reference Figures 1 to 4 , is an embodiment of the present invention, which provides a method for correcting sensor data of an intelligent vehicle based on safety monitoring, comprising the following steps:

[0046] S1. Collect and pre-process raw monitoring data through smart car sensors, calculate the point cloud density mutation rate, the instantaneous drop in signal-to-noise ratio, and the frequency of data continuity interruption, and generate a security feature code;

[0047] The original monitoring data includes the spatial position coordinates of the target object, movement speed value, electromagnetic wave reflection intensity value, three-dimensional point cloud distribution, optical reflectivity value, time synchronization mark, as well as close-range obstacle distance measurement value, vehicle acceleration value and angular velocity value.

[0048] Furthermore, the collection of raw monitoring data is achieved through the collaborative work of multiple sensors: the lidar generates the three-dimensional point cloud distribution and spatial position coordinates of the target object by emitting laser pulses and receiving reflected signals, while recording the optical reflectivity value; the millimeter wave radar uses the reflection characteristics of electromagnetic waves to measure the movement speed and reflection intensity values; the ultrasonic sensor is responsible for measuring the distance to close-range obstacles; and the inertial measurement unit (IMU) continuously collects the vehicle's acceleration and angular velocity values.

[0049] Preprocessing includes normalization, filtering and denoising, time synchronization calibration and validity verification.

[0050] Furthermore, normalization is first performed to uniformly map raw monitoring data of different dimensions (such as millimeter-wave radar reflection intensity and lidar point cloud density) to a standard interval, eliminating the impact of dimensional differences on subsequent analysis. Filtering and denoising are then performed. A statistical outlier removal algorithm is used on the lidar data to eliminate anomalous point clouds, a Kalman filter is applied to the millimeter-wave radar signal to suppress multipath interference, and a sliding window median filter is applied to the IMU data to eliminate transient noise. Time synchronization and calibration utilizes hardware clocks and the PTP (Precision Time Protocol) protocol to achieve microsecond-level time alignment of multi-sensor data, ensuring time consistency. Finally, validity verification is performed, filtering the raw monitoring data using physical plausibility rules to remove significant outliers.

[0051] It should be noted that the physical rationality rules are verification standards established based on statistical analysis of historical vehicle operation data (such as the vehicle's speed, acceleration, and position trajectory on actual roads) and environmental perception data (such as lidar point cloud distribution and millimeter-wave radar reflection intensity).

[0052] Based on the 3D point cloud distribution and time synchronization marking, the density mutation rate of the 3D point cloud distribution is calculated by spatial grid density analysis, and the expression is:

[0053] ;

[0054] in, express The density mutation rate of the three-dimensional point cloud distribution at the moment, express The 3D point cloud density at the moment, express 3D point cloud density at the moment;

[0055] According to the density mutation rate and electromagnetic wave reflection intensity value of the 3D point cloud distribution, the sliding window standard deviation is used to calculate the instantaneous signal-to-noise ratio drop, and the continuity interruption frequency of statistical data is combined with time synchronization mark statistics;

[0056] Furthermore, the instantaneous drop in signal-to-noise ratio is calculated based on the density mutation rate of the three-dimensional point cloud distribution and the electromagnetic wave reflection intensity value. First, a sliding window with a fixed time length is used to segment the electromagnetic wave reflection intensity value, and the instantaneous fluctuation amplitude of the signal-to-noise ratio is calculated through the standard deviation of the electromagnetic wave reflection intensity value in the window; the density mutation rate of the three-dimensional point cloud distribution is used as a dynamic weight factor, and the signal-to-noise ratio fluctuation amplitude in the window is weightedly corrected to generate a normalized instantaneous drop in signal-to-noise ratio value; the time synchronization mark analyzes the timestamp intervals of adjacent data packets, and counts the number of intervals exceeding the timestamp interval threshold as the frequency of data continuity interruption.

[0057] The expression for calculating the instantaneous decrease in signal-to-noise ratio is:

[0058] ;

[0059] in, is the mean signal-to-noise ratio in the sliding window, express The instantaneous decrease in the signal-to-noise ratio at the moment, express The signal-to-noise ratio at the moment, Indicates the standard deviation of the electromagnetic wave reflection intensity value within the current sliding window;

[0060] It should be noted that the time interval threshold is defined based on statistical analysis of network transmission stability requirements (such as maximum allowable delay and average round-trip time), and is usually in the range of 0.05 to 0.2 seconds.

[0061] Based on the density mutation rate, instantaneous decrease in signal-to-noise ratio, and frequency of data continuity interruption, a multi-level threshold criterion is used to identify the abnormal state of the original monitoring data (abnormal point cloud density mutation, abnormal instantaneous decrease in signal-to-noise ratio, and abnormal data continuity interruption flags).

[0062] Furthermore, to identify abnormal conditions based on density mutation rate, instantaneous signal-to-noise ratio (SNR) drop, and data continuity interruption frequency, a multi-level threshold judgment is first established. The first level targets the density mutation rate, setting a density mutation threshold (typically ranging from 5% to 15%) based on statistical analysis of historical monitoring data. When the density mutation rate exceeds the density mutation threshold, a point cloud density mutation anomaly flag is triggered. The second level uses a sliding window mean comparison method to flag an SNR drop anomaly when the instantaneous SNR drop consistently exceeds three standard deviations of the historical mean within the window. The third level uses time series analysis to activate the data continuity interruption anomaly flag when the data continuity interruption frequency reaches the reciprocal of the hardware sampling period per unit time. These three anomaly flags are combined and output using binary bitfield encoding.

[0063] Through the binary bit field encoding method, the abnormal state of the original monitoring data is mapped into the security feature code.

[0064] Furthermore, based on the density mutation rate, the instantaneous decrease in the signal-to-noise ratio, and the frequency of data continuity interruption, an 8-bit binary coding structure is defined: the lowest three bits correspond to the point cloud density mutation anomaly flag (bit0), the signal-to-noise ratio instantaneous decrease anomaly flag (bit1), and the data continuity interruption anomaly flag (bit2); the middle three bits are reserved for extension bits (bit3-bit5); the highest two bits represent the risk level (bit6-bit7), and finally an 8-bit security feature code is synthesized through bit operations.

[0065] S2. Classify the security feature codes based on the security rule library, perform pattern matching based on the historical fault case library, and generate an abnormal diagnosis report for the original monitoring data;

[0066] Based on the abnormal status of historical monitoring data, support vector machine is used to define abnormality judgment rules, and the abnormality judgment rules are associated and mapped with the risk level labels in the historical fault case library through K-nearest neighbor algorithm to generate a safety rule library.

[0067] Furthermore, based on the abnormal state of historical monitoring data, support vector machines are used to classify and train the original monitoring data with abnormal states. By maximizing the category interval to find the optimal hyperplane, the boundary conditions that can distinguish different types of abnormalities (such as density mutation threshold, three times the standard deviation of the historical mean) are determined, thereby forming a multi-dimensional decision boundary for the point cloud density mutation rate, the instantaneous drop in signal-to-noise ratio, and the frequency of data continuity interruption, and obtaining specific abnormality judgment rules; then the K nearest neighbor algorithm is used to calculate the distance between the current safety feature code and each historical fault feature vector in the historical fault case library, and the K historical fault feature vectors with the closest distance are found. According to the majority category of the risk level labels corresponding to the K historical fault feature vectors, an association mapping relationship is established between the abnormality judgment rules and the risk level labels, and finally a safety rule library containing abnormality judgment rules and corresponding risk level labels is generated.

[0068] Support vector machines are trained using supervised learning methods based on historical monitoring data labeled with a specific state category. It should be noted that the historical fault case database is a database that stores historical fault feature vectors, their corresponding risk level labels, and fault type information. This database is constructed by collecting various sensor failure cases that occurred during the actual operation of smart cars and storing them as historical fault feature vectors.

[0069] The security feature codes are classified into different levels based on the security rule base, and the fault types are matched from the historical fault case library according to the cosine similarity between the security feature codes and the historical fault feature vectors. An abnormal diagnosis report is generated and loaded into the historical fault case library.

[0070] Furthermore, the security feature code is input into the security rule library, and the risk level corresponding to the security feature code is determined based on the association mapping relationship between the abnormality judgment rules and the risk level labels in the security rule library, and the level division is completed; then, the cosine similarity between the security feature code and each historical fault feature vector in the historical fault case library is calculated by weighted dot product, and by comparing the cosine similarity values, the fault type corresponding to the historical fault feature vector with the highest cosine similarity is matched from the historical fault case library; an abnormality diagnosis report is generated by combining the level division result and the matched fault type, and the generated abnormality diagnosis report is loaded into the historical fault case library as a new record.

[0071] The cosine similarity between the safety feature code and the historical fault feature vector is calculated by weighted dot product. The expression is:

[0072] ;

[0073] in, is the cosine similarity between the safety feature encoding and the historical fault feature vector, is the number of bits in the security feature code, Is the security feature code The weight coefficient of the bit (usually in the range of 0.1 to 1.0), Is the first security feature code Place value, is the first character vector of the historical fault case Place value;

[0074] Weight coefficient It is defined based on the statistical analysis of the importance of the historical fault feature vector dimensions;

[0075] It should be explained that Before participating in the calculation, the dimension is eliminated by Z-score normalization. S3. Build the physical topology network of the smart car sensor, and dynamically adjust the node status of the physical topology network according to the abnormal diagnosis report, and obtain and execute the physical topology network reconstruction instruction;

[0076] Based on the installation location coordinates and communication link relationships of smart car sensors, a physical topology network of smart car sensors is constructed;

[0077] Furthermore, the three-dimensional installation position coordinates of each smart car sensor on the car body are obtained, and the communication connection status (such as normal connection, connection interruption and signal strength level) and link bandwidth information (such as maximum transmission rate, current available bandwidth and data packet loss rate) between each sensor are collected. Each smart car sensor is used as a node and the communication link relationship as an edge. The spatial layout of the nodes is determined according to the installation position coordinates, and the connection structure between the nodes is established based on the communication link relationship. The nodes and edges are topologically connected through the graph structure representation method, and finally the physical topology network of the smart car sensors is constructed.

[0078] It should be noted that the communication link relationship refers to the communication connection and its properties between smart car sensors, for example, the connection between sensor A and sensor B is normal and has a maximum transmission rate of 10 Mbps and a packet loss rate of 5%.

[0079] Based on the risk level in the abnormal diagnosis report, the physical topology network nodes are marked as isolated / demoted / normal. Node control parameters are extracted through hash table query and encapsulated using DDS to generate node control instructions.

[0080] Furthermore, based on the risk level in the abnormal diagnosis report, the risk level is matched with the node status mapping table to determine the status mark that the corresponding physical topology network node should execute. The status mark includes isolation, demotion or normal state; then, using the node identifier as the key, the control parameters corresponding to the current node are extracted from the node control parameter database through the hash table query method, including power-off delay time, weight adjustment value or recovery instruction; the obtained node control parameters are serialized and encapsulated according to DDS (Data Distribution Service) to generate node control instructions containing the target node identifier, status mark and control parameters.

[0081] It should be noted that node control instructions are hardware-level control protocols used to dynamically adjust the working status of nodes in the physical topology network of smart car sensors. They achieve real-time optimization of the network topology by encapsulating node control parameters (such as isolation flags, weight reduction coefficients, sampling frequency, etc.).

[0082] Dynamically adjust the node status of the physical topology network according to node control instructions (such as the power-off delay time of isolated nodes or the weight value of demoted nodes);

[0083] Furthermore, the target node identifier, status mark and control parameters contained in the node control instruction are parsed through the DDS serialization parsing method to confirm whether the physical topology network node that needs to be adjusted and its corresponding status mark are isolated or downgraded. For example, if the risk level value is greater than or equal to the isolation threshold, the status mark is isolated; if the risk level value is between the downgrade threshold and the isolation threshold, the status mark is downgraded. If the status mark is isolated, the power-off delay time of the isolated node is obtained from the control parameter, and the power supply of the current node is cut off after the power-off delay time is reached; if the status mark is downgraded, the weight value of the downgraded node is obtained from the control parameter, and the original monitoring data of the current node in the physical topology network is reduced;

[0084] It should be noted that the demotion threshold and isolation threshold are set based on statistical analysis of historical failure cases. The demotion threshold usually ranges from 0.5 to 0.8, and the isolation threshold usually ranges from 0.8 to 1.0.

[0085] Collect the state feedback data of the adjusted physical topology network, predict the health status score of each node in the physical topology network through fuzzy comprehensive evaluation method, and use the earliest deadline first algorithm to generate and execute the physical topology network reconstruction instructions of the smart car sensors;

[0086] Furthermore, the state feedback data of the adjusted physical topology network is collected, and the state feedback data is used as an input parameter. Combined with the preset evaluation factor weights and membership functions, the operating status of each node is quantitatively evaluated through the fuzzy comprehensive evaluation method, and the health status score of each node in the physical topology network is output; according to the health status score of each node and its priority in the physical topology network, the reconstruction task deadline of each node is determined, the reconstruction tasks are arranged in ascending order of deadline, and the earliest deadline first algorithm is used to schedule the reconstruction tasks, and the physical topology network reconstruction instructions of the smart car sensors are generated and executed.

[0087] It should be noted that the weights of evaluation factors are set based on the vehicle's historical operating data, and usually range from 0.1 to 1.0.

[0088] Status feedback data includes node online status, communication delay, and data packet loss rate.

[0089] S4. Based on the node status of the reconstructed physical topology network, the original monitoring data is corrected twice to generate vehicle safety monitoring data.

[0090] Based on the node status (isolation / degraded / normal state) of the reconstructed physical topology network, principal component analysis is used to extract the spatial position coordinates and weight distribution parameters of the normal state nodes, and the node weight distribution table of the physical topology network is generated through weighted least squares method;

[0091] Furthermore, based on the node status (isolation / degraded / normal status) of the reconstructed physical topology network, the nodes in the normal state are screened out through the state label matching method, the spatial position coordinates and weight distribution parameters of the normal state nodes are collected, the spatial position coordinates and weight distribution parameters are combined into a multidimensional data matrix, the multidimensional data matrix is ​​standardized, the covariance matrix is ​​calculated and the eigenvalues ​​and eigenvectors are solved, the principal components with a cumulative contribution rate greater than the principal component cumulative contribution rate threshold are selected, and the spatial position coordinates and weight distribution parameters in the main feature directions are extracted through principal component analysis, retaining the information with the greatest impact on the overall distribution, and then the extracted spatial position coordinates and weight distribution parameters are used as input, and the corresponding initial weights are assigned according to the spatial distribution differences of each node. The relative reliability between nodes is optimized and calculated through the weighted least squares method to generate a node weight distribution table of the physical topology network.

[0092] It should be noted that the weight distribution parameter refers to a value determined according to the position of the node in the physical topology network and the degree of its impact on the performance of the physical topology network, and is used to adjust the contribution and priority of the original monitoring data of each node.

[0093] The principal component cumulative contribution rate threshold is defined based on the statistical analysis of the variance distribution characteristics of the original monitoring data under historical normal operating conditions, and is usually in the range of 0.85 to 0.95.

[0094] Based on the node weight distribution table, the weighted residual correction method is used to calculate the deviation between the original monitoring data of each node and the weighted mean, and the original monitoring data is corrected for the first time according to the weight ratio compensation to generate the initial vehicle safety monitoring data;

[0095] Furthermore, a weighted mean is calculated based on the original monitoring data of each node and its corresponding weight distribution parameters; then, a weighted residual correction method is used to calculate the deviation between the monitoring data of each node and the corresponding weighted mean to reflect the degree of deviation of the node monitoring data from the overall average state; then, the deviation is compensated according to the weight ratio of the node, that is, the monitoring data of the node is adjusted according to its importance to reduce the difference with the corresponding weighted mean; the first correction of the monitoring data of all nodes is achieved, thereby generating initial vehicle safety monitoring data.

[0096] The confidence weights of each initial vehicle safety monitoring data are statistically analyzed through a sliding window, and the confidence weights are checked for spatiotemporal consistency, marking the invalid confidence weights.

[0097] Furthermore, each initial vehicle safety monitoring data is continuously sampled in the time dimension through a sliding window to form a time series data segment, and the statistical characteristics of the built-in confidence weight of each time series data segment are obtained, including the mean, variance and rate of change; at the same time, the confidence weights of adjacent sensor nodes at the same time are compared in the spatial dimension, and the spatial neighborhood difference method is used to analyze the spatial distribution differences; based on the preset spatiotemporal consistency threshold, it is judged whether the confidence weight of each initial vehicle safety monitoring data is continuous and stable in time and conforms to the proximity relationship in space. If the confidence weight suddenly changes in the time series or exceeds the spatiotemporal consistency threshold, the confidence weight is marked as an invalid confidence weight.

[0098] It should be noted that the spatiotemporal consistency threshold is defined based on road environment parameters and sensor historical performance data, including the time dimension threshold (usually ranging from 0.1 to 0.3) and the space dimension threshold (usually ranging from 0.15 to 0.25).

[0099] The iterative reweighted least squares method is used to perform secondary correction on the initial vehicle safety monitoring data with invalid confidence weights to generate vehicle safety monitoring data.

[0100] Furthermore, an iterative reweighted least squares method is used, and the initial vehicle safety monitoring data with invalid confidence weights is used as input. The weight of each data point is initialized to the same value, and the weighted least squares estimate under the current weight is calculated to obtain a preliminary correction result; based on the absolute value of the residual of the preliminary correction result and the initial vehicle safety monitoring data obtained by the weighted residual correction method, the Huber weight function is used to update the weight of each data point, and the data points with larger residuals are assigned lower weights; the weighted least squares calculation and weight update process are repeated until the weight change is less than the preset convergence threshold or the maximum number of iterations is reached, and finally the corrected vehicle safety monitoring data is output.

[0101] It should be noted that the convergence threshold is defined based on the mean and standard deviation of the L2 norm of two consecutive weight differences in historical iterative calculations, and is usually in the range of: .

[0102] This embodiment also provides a smart car sensor data correction system based on safety monitoring, including:

[0103] The data acquisition module collects raw monitoring data through intelligent vehicle sensors and performs preprocessing to calculate the point cloud density mutation rate, the instantaneous drop in signal-to-noise ratio, and the frequency of data continuity interruptions to generate a security feature code;

[0104] The abnormality diagnosis module classifies the security feature codes based on the security rule library, performs pattern matching based on the historical fault case library, and generates an abnormality diagnosis report for the original monitoring data;

[0105] The topology network adjustment module builds the physical topology network of smart car sensors, dynamically adjusts the node status of the physical topology network based on abnormal diagnosis reports, and obtains and executes physical topology network reconstruction instructions;

[0106] The data correction module corrects the original monitoring data twice based on the node status of the reconstructed physical topology network to generate vehicle safety monitoring data.

[0107] This embodiment also provides a computer device suitable for the case of a smart car sensor data correction method based on safety monitoring, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the smart car sensor data correction method based on safety monitoring proposed in the above embodiment.

[0108] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0109] This embodiment further provides a storage medium having a computer program stored thereon. When the program is executed by a processor, the method for correcting sensor data of an intelligent vehicle based on safety monitoring as 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 (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0110] In summary, the present invention achieves accurate quantitative characterization of multi-dimensional abnormal states by using multi-level threshold criteria to jointly analyze the point cloud density mutation rate, instantaneous signal-to-noise ratio drop, and data continuity interruption frequency, and generates structured safety feature codes through binary bit field coding, effectively improving the accuracy and efficiency of fault feature recognition under complex working conditions. By combining weighted residual correction with iterative reweighted least squares, a two-level correction is performed on the monitoring data based on the reconstructed node weight distribution table, achieving progressive error compensation under conditions of dynamic performance degradation of the sensor network, and improving the reliability and environmental adaptability of the output data.

[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. 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. A method for correcting sensor data of an intelligent vehicle based on safety monitoring, characterized by: include, The original monitoring data is collected and pre-processed through smart car sensors to calculate the point cloud density mutation rate, the instantaneous drop in signal-to-noise ratio, and the frequency of data continuity interruption to generate a security feature code; Based on the security rule library, the security feature codes are classified into different levels, and pattern matching is performed in combination with the historical fault case library to generate an abnormal diagnosis report of the original monitoring data; Build the physical topology network of smart car sensors, dynamically adjust the node status of the physical topology network based on abnormal diagnosis reports, and obtain and execute physical topology network reconstruction instructions; Based on the node status of the reconstructed physical topology network, the original monitoring data is corrected twice to generate vehicle safety monitoring data.

2. The method for correcting sensor data of an intelligent vehicle based on safety monitoring according to claim 1, wherein: The raw monitoring data includes the spatial position coordinates of the target object, motion speed value, electromagnetic wave reflection intensity value, three-dimensional point cloud distribution, optical reflectivity value, time synchronization mark, as well as close-range obstacle distance measurement value, vehicle acceleration value and angular velocity value.

3. The method for correcting sensor data of an intelligent vehicle based on safety monitoring according to claim 2, wherein: The steps for calculating the point cloud density mutation rate, the instantaneous decrease in signal-to-noise ratio, and the frequency of data continuity interruption are as follows: Based on the 3D point cloud distribution and time synchronization marking, the density mutation rate of the 3D point cloud distribution is calculated; According to the density mutation rate of the three-dimensional point cloud distribution and the electromagnetic wave reflection intensity value, the instantaneous drop in the signal-to-noise ratio is calculated, and the frequency of statistical continuity interruption is analyzed in combination with the time synchronization mark statistics.

4. The method for correcting sensor data of an intelligent vehicle based on safety monitoring according to claim 1, wherein: The steps of classifying the security feature codes based on the security rule library, performing pattern matching in combination with the historical fault case library, and generating an abnormal diagnosis report of the original monitoring data are as follows: Based on the abnormal status of historical monitoring data, define abnormality judgment rules, associate and map the abnormality judgment rules with the risk level labels in the historical fault case library, and generate a safety rule library; The security feature codes are classified into different levels based on the security rule base, and the fault types are matched from the historical fault case library according to the cosine similarity between the security feature codes and the historical fault feature vectors. An abnormal diagnosis report is generated and loaded into the historical fault case library.

5. The method for correcting sensor data of an intelligent vehicle based on safety monitoring according to claim 4, characterized in that: The steps of constructing the physical topology network of smart car sensors and dynamically adjusting the node status of the physical topology network according to the abnormal diagnosis report are as follows: Build a physical topology network of smart car sensors and dynamically adjust the node status of the physical topology network based on abnormal diagnosis reports; Based on the risk level in the abnormal diagnosis report, the physical topology network nodes are marked for status, node control parameters are extracted and encapsulated, and node control instructions are generated; Dynamically adjust the node status of the physical topology network according to node control instructions.

6. The method for correcting sensor data of an intelligent vehicle based on safety monitoring according to claim 5, characterized in that: The obtaining and executing of physical topology network reconstruction instructions refers to collecting status feedback data of the adjusted physical topology network, predicting the health status score of each node in the physical topology network, and using the earliest deadline first algorithm to generate and execute physical topology network reconstruction instructions for smart car sensors.

7. The method for correcting sensor data of an intelligent vehicle based on safety monitoring according to claim 6, characterized in that: The node status of the reconstructed physical topology network is used to modify the original monitoring data twice to generate vehicle safety monitoring data. The steps are as follows: Based on the node status of the reconstructed physical topology network, the spatial position coordinates and weight distribution parameters of the normal state nodes are extracted, and the node weight distribution table of the physical topology network is generated by weighted least squares method; Based on the node weight distribution table, the weighted residual correction method is used to calculate the deviation between the original monitoring data of each node and the weighted mean, and the original monitoring data is corrected for the first time according to the weight ratio compensation to generate the initial vehicle safety monitoring data; Statistically analyze the confidence weights of each initial vehicle safety monitoring data, perform spatiotemporal consistency checks on the confidence weights, and mark invalid confidence weights; The iterative reweighted least squares method is used to perform secondary correction on the initial vehicle safety monitoring data with invalid confidence weights to generate vehicle safety monitoring data.

8. The method for correcting sensor data of an intelligent vehicle based on safety monitoring according to claim 1, wherein: The steps for generating the security feature code are as follows: Identify abnormal conditions in raw monitoring data based on density mutation rate, instantaneous drop in signal-to-noise ratio, and frequency of data continuity interruption; Through the binary bit field encoding method, the abnormal state of the original monitoring data is mapped into the security feature code.

9. The method for correcting sensor data of an intelligent vehicle based on safety monitoring according to claim 1, wherein: The preprocessing includes normalization, filtering and denoising, time synchronization calibration and validity verification processing.

10. A smart car sensor data correction system based on safety monitoring, based on the smart car sensor data correction method based on safety monitoring according to any one of claims 1 to 9, characterized in that: include, The data acquisition module collects raw monitoring data through intelligent vehicle sensors and performs preprocessing to calculate the point cloud density mutation rate, the instantaneous drop in signal-to-noise ratio, and the frequency of data continuity interruptions to generate a security feature code; The abnormality diagnosis module classifies the security feature codes based on the security rule library, performs pattern matching based on the historical fault case library, and generates an abnormality diagnosis report for the original monitoring data; The topology network adjustment module builds the physical topology network of smart car sensors, dynamically adjusts the node status of the physical topology network based on abnormal diagnosis reports, and obtains and executes physical topology network reconstruction instructions; The data correction module corrects the original monitoring data twice based on the node status of the reconstructed physical topology network to generate vehicle safety monitoring data.

Images