Multi-data fusion method and system in airport station safety supervision and medium

Abstract

The invention relates to the technical field of data processing, and discloses a multi-data fusion method and system in airport station safety supervision and a medium. The method comprises the following steps: based on a target feature matrix, judging potential conflicts through a set safety distance threshold and a speed threshold, and constructing a risk assessment index in combination with historical risk event data to obtain a safety risk score; according to the security risk score, establishing a spatial association relationship and a risk propagation relationship between the apron targets, calculating regional risk density through a risk accumulation effect, and generating real-time security situation data; and for the real-time security situation data, calculating a target function value of security control, and setting a security constraint condition for optimization solution to form a control strategy parameter. According to the invention, effective fusion of multi-source heterogeneous data is realized, accurate awareness of the security situation of the station level is constructed, and an optimized management and control strategy is generated based on situation information, so that the accuracy and real-time performance of security supervision of the station level are improved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a multi-data fusion method, system and medium for airport apron safety supervision. Background Art

[0002] The airport apron is an important area for aircraft to carry out various support operations, involving multiple operations such as aircraft taxiing, towing, maintenance inspection, refueling, and catering. Existing apron safety supervision mainly relies on a single data source for management, such as video surveillance systems for target detection and tracking, radar systems for position measurement, and ADS-B systems for aircraft identification. These regulatory means operate independently, forming multiple information islands. At the same time, the data collected by different monitoring equipment have problems such as time delay, inconsistent spatial references, and uneven data quality, which brings many challenges to apron safety management.

[0003] The main problem with existing technologies is the lack of an effective fusion mechanism for multi-source heterogeneous data. The limitations of a single data source lead to the existence of monitoring blind spots, and there is a lack of coordination between different data sources, making it impossible to form a comprehensive understanding of the apron situation. In addition, existing risk assessment methods often only consider static safety interval requirements, ignoring the spatial propagation and cumulative effects of risks, and it is difficult to accurately reflect the dynamic risk distribution in the complex environment of the apron. In the formulation of management and control strategies, existing methods mostly use preset rules and lack consideration of multi-objective collaborative optimization, making it difficult to balance operational efficiency while ensuring safety. Summary of the invention

[0004] The present application provides a multi-data fusion method, system and medium for airport apron safety supervision, which is used to realize the effective fusion of multi-source heterogeneous data, build accurate apron safety situation awareness, and generate optimized management and control strategies based on situation information, thereby improving the accuracy and real-time performance of apron safety supervision.

[0005] In a first aspect, the present application provides a multi-data fusion method for airport apron safety supervision, and the multi-data fusion method for airport apron safety supervision includes: for video data, radar data and location data collected by multi-source sensors on the apron, a data credibility value is generated according to data integrity and time consistency, and the original data is filtered and denoised by the data credibility value to obtain a preprocessed data set; for the preprocessed data set, a unified spatiotemporal benchmark is established through timestamp and coordinate transformation, and the matching degree between different data sources is calculated according to the target feature similarity to obtain fused target data; according to the fused target data, feature parameters are extracted from three dimensions of position, speed and motion trajectory, and the feature parameters are screened and combined according to the correlation coefficient to form a target feature matrix; based on the target feature matrix, potential conflicts are judged by setting safety distance thresholds and speed thresholds, and risk assessment indicators are constructed in combination with historical risk event data to obtain a safety risk score; according to the safety risk score, a spatial correlation relationship and risk propagation relationship between apron targets are established, and the regional risk density is calculated by the risk accumulation effect to generate real-time safety situation data; for the real-time safety situation data, the objective function value of safety control is calculated, and safety constraints are set for optimization and solution to form control strategy parameters.

[0006] In a second aspect, the present application provides a multi-data fusion system for airport apron safety supervision, and the multi-data fusion system for airport apron safety supervision includes: A generation module is used to generate data credibility values ​​based on data integrity and time consistency for video data, radar data, and location data collected by multi-source sensors on the apron, and to filter and denoise the original data using the data credibility values ​​to obtain a preprocessed data set; The fusion module is used to establish a unified spatiotemporal benchmark for the preprocessed data set through timestamp and coordinate conversion, calculate the matching degree between different data sources based on the similarity of target features, and obtain the fused target data; The extraction module is used to extract feature parameters from the three dimensions of position, speed and motion trajectory according to the fused target data, and to screen and combine the feature parameters according to the correlation coefficient to form a target feature matrix; The judgment module is used to judge potential conflicts based on the target feature matrix, through the set safety distance threshold and speed threshold, and to construct risk assessment indicators in combination with historical risk event data to obtain a safety risk score; Establish a module to establish the spatial correlation and risk propagation relationship between apron targets based on the safety risk score, calculate the regional risk density through the risk accumulation effect, and generate real-time safety situation data; The solution module is used to calculate the objective function value of security control based on real-time security situation data, set security constraints for optimization and solution, and form control strategy parameters.

[0007] A third aspect of the present application provides a computer-readable storage medium, in which instructions are stored. When the computer-readable storage medium is executed on a computer, the computer executes the above-mentioned multi-data fusion method for airport apron safety supervision.

[0008] In the technical solution provided by the present application, the original data is filtered and denoised by the data credibility value, which solves the problem of quality differences in multi-source heterogeneous data and improves data reliability; secondly, a unified spatiotemporal benchmark is established by using timestamps and coordinate transformations, and the matching degree between data sources is calculated based on the target feature similarity, thereby achieving accurate alignment and association between different data sources; thirdly, feature parameters are extracted from the three dimensions of position, speed, and motion trajectory and correlation analysis is performed to construct a multi-dimensional target feature description, thereby enhancing the ability to express the target state; potential conflicts are judged by setting safety distance thresholds and speed thresholds, and an evaluation index system is constructed in combination with historical risk event data, thereby achieving accurate quantification of risks; at the same time, by establishing spatial correlation relationships and risk propagation relationships between apron targets, the concept of risk accumulation effect is introduced, and the spatial distribution and propagation characteristics of risks are accurately characterized; finally, the objective function value is calculated based on real-time safety situation data and constraints are set, and the optimization algorithm is used to solve the control strategy parameters, thereby achieving collaborative optimization of multiple objectives while ensuring safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] 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 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.

[0010] Figure 1 This is a schematic diagram of an embodiment of a multi-data fusion method for airport apron safety supervision in an embodiment of the present application; Figure 2 A radar multi-dimensional information waveform diagram provided in an embodiment of the present application; Figure 3 This is a schematic diagram of an embodiment of a multi-data fusion system for airport apron safety supervision in an embodiment of the present application. DETAILED DESCRIPTION

[0011] The embodiments of the present application provide a multi-data fusion method, system and medium for airport apron security supervision. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0012] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of the multi-data fusion method in the airport apron safety supervision includes: Step S101: for the video data, radar data and location data collected by the multi-source sensors of the apron, a data credibility value is generated according to the data integrity and time consistency, and the raw data is filtered and denoised by the data credibility value to obtain a preprocessed data set; Step S102: for the preprocessed data set, a unified spatiotemporal reference is established through timestamp and coordinate conversion, and the matching degree between different data sources is calculated according to the target feature similarity to obtain the fused target data; Step S103: extracting feature parameters from the three dimensions of position, speed and motion trajectory according to the fused target data, screening and combining the feature parameters according to the correlation coefficient to form a target feature matrix; Step S104: Based on the target feature matrix, potential conflicts are judged by setting safety distance thresholds and speed thresholds, and risk assessment indicators are constructed in combination with historical risk event data to obtain safety risk scores; Step S105: Based on the safety risk score, establish the spatial correlation relationship and risk propagation relationship between the apron targets, calculate the regional risk density through the risk accumulation effect, and generate real-time safety situation data; Step S106: Calculate the objective function value of security control based on the real-time security situation data, set security constraints for optimization and solution, and form control strategy parameters.

[0013] It is understandable that the execution subject of the present application may be a multi-data fusion system in airport apron safety supervision, or a terminal or a server, which is not limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0014] Specifically, video data, radar data and location data are processed. Video data is an image sequence collected by the video surveillance equipment in the apron. Radar data contains the distance, direction and speed information of the target. Location data records the latitude, longitude and timestamp of the target. Data integrity is obtained by calculating the data missing rate and sampling uniformity. The image sequence of video data is analyzed in units of 50 milliseconds, and the frame integrity in each time window is counted. Radar data determines data continuity by detecting the time interval between sampling points. The integrity of location data is determined by the uniformity of sampling time. Time consistency is based on the comparison of timestamps from different data sources, calculating the degree of deviation at the sampling time, and generating a time synchronization index. The data credibility value is obtained by weighting the integrity index and the time consistency index, and the weight coefficient is set according to the importance of different data sources. In the data preprocessing stage, data from different sources are unified to the same time and space reference. The time reference adopts the UTC time standard, and the space reference adopts the local coordinate system of the airport apron. Timestamp conversion involves time zone adjustment and millisecond-level precision alignment, and coordinate conversion requires the establishment of a mapping relationship between different coordinate systems. The calculation of target feature similarity includes three aspects: geometric features, motion features, and timing features. The Euclidean distance between feature vectors is used to measure whether different data sources describe the same target. The degree of matching is calculated using a similarity matrix, and the matrix elements represent the strength of the correspondence between different data sources. For the fused target data, features are extracted from three dimensions: position, speed, and motion trajectory. Position features contain the two-dimensional coordinates and height information of the target, speed features contain linear speed and angular velocity, and motion trajectory features describe the historical motion path of the target. The Pearson correlation coefficient is used for the correlation analysis of feature parameters to calculate the degree of linear correlation between different features and screen out feature combinations with significant correlation. Each row of the feature matrix represents a target, and each column corresponds to a feature dimension.

[0015] In the risk assessment stage, potential conflicts are judged based on the target feature matrix. The safety distance threshold is determined according to the operation specifications of different types of targets, and the speed threshold is set based on the speed limit requirements of the apron. The historical risk event data includes past safety event records such as collisions, boundary intrusions, and speeding. The risk assessment index system is constructed through statistical analysis. The safety risk score is normalized to unify the risk indicators of different dimensions into the range of 0-1. In the process of generating real-time safety situation data, the spatial correlation between targets is calculated, including distance, relative position, and movement trend. The risk propagation relationship describes the diffusion effect of risk between adjacent targets, and the transmission intensity of risk is represented by the risk propagation matrix. The calculation of regional risk density takes into account the risk accumulation effect, and the risk values ​​of all targets within a unit area are weighted and superimposed. In the control strategy generation stage, the real-time safety situation data is converted into an objective function, and the function value reflects the overall risk level under the current situation. Safety constraints include hard constraints such as minimum safety distance, maximum allowable speed, and prohibited areas. The optimization solution process adopts an iterative calculation method to gradually adjust the control parameters until the constraints are met and the objective function reaches the optimal.

[0016] For example, when multiple aircraft and ground vehicles are operating at the same time on the apron, video surveillance reveals that there is an intersection between the movement trajectories of target A and target B. The acquired data is processed to extract the position and speed information of the target, calculate the relative distance and speed of the two targets, and generate the current risk level by combining the risk event records in similar scenarios in historical data. Then, the spatial propagation law of the risk is analyzed. Considering that target C is also active in the area, the risk status of the three targets is correlated and analyzed to generate targeted management and control suggestions, such as adjusting the route of target B or reducing the running speed of target A.

[0017] In the embodiment of the present application, the original data is filtered and denoised by the data credibility value, which solves the problem of quality differences in multi-source heterogeneous data and improves data reliability; secondly, a unified spatiotemporal benchmark is established by using timestamps and coordinate transformations, and the matching degree between data sources is calculated based on the target feature similarity, thereby achieving accurate alignment and association between different data sources; thirdly, feature parameters are extracted from the three dimensions of position, speed, and motion trajectory and correlation analysis is performed to construct a multi-dimensional target feature description, thereby enhancing the expressive power of the target state; potential conflicts are judged by setting safety distance thresholds and speed thresholds, and an evaluation index system is constructed in combination with historical risk event data, thereby achieving accurate quantification of risks; at the same time, by establishing spatial correlation relationships and risk propagation relationships between apron targets, the concept of risk accumulation effect is introduced, thereby accurately characterizing the spatial distribution and propagation characteristics of risks; finally, the objective function value is calculated based on real-time safety situation data and constraints are set, and the optimization algorithm is used to solve the control strategy parameters, thereby achieving collaborative optimization of multiple objectives while ensuring safety.

[0018] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Segment the video data according to a fixed time window, divide the segmented video data into image frame sequences, calculate the data integrity based on the pixel grayscale values ​​in the image frame sequences, and obtain the video data integrity index; (2) Perform sampling rate detection on radar data, extract the signal waveform in the radar data, count the number of signal interruptions based on the continuity of the waveform, and obtain the radar data integrity index; (3) Perform time stamp analysis on the location data, calculate the sampling interval variance of the location data, determine the stability of data collection based on the variance, and obtain the location data integrity index; (4) Perform a weighted summation of the video data integrity index, radar data integrity index, and location data integrity index to generate a comprehensive integrity score; (5) Compare the time stamps of the video data, radar data, and location data, calculate the deviation value of the data arrival time, and generate a time consistency score; (6) The data credibility value is calculated based on the weighted combination of the comprehensive integrity score and the temporal consistency score. The data is filtered through the set threshold and the noise points are removed to obtain the preprocessed data set.

[0019] Specifically, when processing video data, a fixed time window means dividing the continuous video stream into fixed time lengths, such as dividing the video data into a time window every 50 milliseconds. In each time window, the video data is decomposed into a series of continuous image frames, each of which contains complete information about the apron scene at a specific moment. The following formula is used to calculate the data integrity for the pixel grayscale value in the image frame sequence: ; in, represents the video data integrity indicator, W represents the total number of frames in the time window, M and N represent the number of rows and columns of the image respectively, Represents the grayscale value of the pixel at the coordinate (x, y) in the i-th frame image, is the pixel quality evaluation function.

[0020] The processing of radar data focuses on the detection of sampling rate and the extraction of signal waveform from the original radar data. The waveform data contains the distance, direction and speed information of the target. Figure 2 As shown, it is a radar multi-dimensional information waveform diagram provided by an embodiment of the present invention, wherein: Figure 2 The figure includes three sub-figures, which represent the range signal waveform, azimuth signal waveform and speed signal waveform respectively. The range signal waveform is represented by the red curve, which is used to show the relative distance change between the target and the radar; the azimuth signal waveform is represented by the blue curve, which is used to show the azimuth angle change of the target relative to the radar; the speed signal waveform is represented by the green curve, which is used to show the target's moving speed change. The black dots on each waveform represent the actual sampling points, which are used to reflect the sampling rate of the radar data. The ordinate in the figure represents the amplitude of the range signal, the angle value of the azimuth signal and the speed value of the speed signal, respectively, and the abscissa represents time. By detecting the sampling rate of the three-dimensional signals and analyzing the waveform continuity, the integrity index of the radar data is calculated to evaluate the data quality. When the waveform is discontinuous or the sampling point is missing, it indicates that there is a signal interruption in the radar data, which needs to be supplemented by other data sources. The continuity of the waveform is judged by counting the time interval between adjacent sampling points. When the time interval exceeds the preset threshold, it is recorded as a signal interruption. The ratio of the number of signal interruptions to the total number of sampling points is used as the integrity index of the radar data. The time stamp parsing of the position data includes extracting the timestamp information from the data packet and converting it into a unified time format. The calculation of sampling interval variance is done by statistically analyzing the time difference between adjacent sampling points. The smaller the variance value, the more stable the data collection. The position data integrity index is obtained by normalizing the sampling interval variance.

[0021] The comprehensive integrity score is generated by weighted summation, with different weights assigned to the integrity indicators of the three data sources. The weights are set based on the importance of the data source in safety supervision, and the sum of the weights of video data, radar data, and location data is 1. The calculation of the temporal consistency score aligns the time stamps of the three data sources to the same time reference, and then calculates the deviation between the actual arrival time and the theoretical arrival time of each data source. The smaller the deviation value, the better the temporal consistency of the data. The temporal consistency score of the data is obtained by normalizing the deviation value.

[0022] The calculation of the data credibility value is a weighted combination of the comprehensive integrity score and the time consistency score. The setting of the weight coefficient reflects the relative importance of integrity and consistency in data quality assessment. During the data filtering process, the data credibility value is compared with the preset threshold. Data points below the threshold are marked as noise points and removed.

[0023] For example, taking the process of an aircraft sliding out of the parking space and arriving at the taxiway as an example, the grayscale value analysis of the image frame sequence collected by the video surveillance equipment found that some frames were blurred, the radar data had a brief signal loss when the aircraft turned, and the sampling interval of the position data was uneven due to occlusion. By calculating the comprehensive integrity score and time consistency score, the reliability of the information of the three data sources was evaluated, and it was found that the credibility of the radar data was low during the turning stage. Therefore, when generating the preprocessing data set, this stage mainly relies on the fusion results of the video data and the position data, thereby ensuring the accuracy of the data. The entire data preprocessing process effectively screens out reliable data sources and eliminates unreliable data points by quantitatively evaluating the quality of multi-source data, providing high-quality input data for subsequent data fusion.

[0024] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) Parse the timestamps in the preprocessed data set, align the sampling times of different data sources to a unified time axis, and generate a time-synchronized sequence; (2) Performing a benchmark transformation on the coordinate data in the time synchronization sequence, mapping the position information in different coordinate systems to a unified apron coordinate system, and obtaining spatially aligned data; (3) Extract the target’s geometric size, motion direction, and speed information from the spatial alignment data and construct a target feature description set; (4) Calculate the Euclidean distance between feature vectors based on the target feature description set to generate a similarity matrix between data sources; (5) Normalize the similarity matrix, select the best matching pair by setting the matching threshold, and form a target association table; (6) Information on multi-source data is merged based on the target association table, and the target parameters are fused using the weighted average method to obtain the fused target data.

[0025] Specifically, the timestamps in the preprocessed data set are parsed. The timestamp is an accurate record of the time when the data was collected, including year, month, day, hour, minute, second, and millisecond information. The sampling times of different data sources need to be aligned to a unified time axis, and the UTC time is uniformly used as the benchmark. The timestamps of each data source are converted to a time zone and aligned with millisecond accuracy to generate a time synchronization sequence. For the coordinate data in the time synchronization sequence, a benchmark conversion is required to unify the spatial reference system. The image coordinates in the video data use the pixel coordinate system, the radar data use the polar coordinate system, and the position data use the longitude and latitude coordinate system. The position information in these different coordinate systems is uniformly converted to the local coordinate system of the airport apron. The apron coordinate system uses the runway entrance as the origin to establish the northeast sky coordinate system. In the coordinate conversion process, the installation position and attitude parameters of different devices need to be considered, and various types of data are mapped to a unified spatial reference through the coordinate transformation matrix.

[0026] Target features are extracted from spatially aligned data. Geometric dimensions include the length, width, and height of the target, which are obtained through image segmentation and bounding box extraction. The direction of motion is obtained by calculating the change in the target position at adjacent moments, and the speed information is calculated by dividing the displacement distance by the time interval. The target feature description set organizes these features into a unified data structure, with one feature vector for each target.

[0027] The Euclidean distance between feature vectors is calculated using the following formula: ; in, represents the distance between the i-th eigenvector and the j-th eigenvector, and represent the kth components of the two eigenvectors, respectively. represents the weight coefficient of the kth feature, Represents the time decay factor, and n represents the feature dimension. By calculating the distance between all feature vector pairs, the similarity matrix between data sources is generated.

[0028] The normalization of the similarity matrix uses the maximum and minimum value normalization method to map the similarity value to the range of 0 to 1. The optimal matching pair is selected by setting the matching threshold. When the similarity exceeds the threshold, it is considered that the two data sources describe the same target, ensuring that the target of one data source matches at most one target of another data source, forming a target association table.

[0029] The multi-source data is merged according to the target association table. For the matched targets, the weighted average method is used to fuse their parameters. The weight is set based on the data credibility. The data source with high credibility has a larger weight. For the location information, the weight is determined by combining the GPS positioning accuracy and the radar ranging accuracy; for the speed information, the weight is set according to the Doppler speed measurement accuracy and the image target tracking accuracy.

[0030] For example, when an aircraft is taxiing on the apron, the video surveillance equipment collects the target's image sequence, the radar equipment records the target's distance and azimuth data, and the ADS-B receiver obtains the target's location broadcast information. The timestamps of these data are unified to UTC time, and the frame timestamps of the video data, the radar scanning cycle, and the ADS-B broadcast time form an aligned time series. Then the target pixel coordinates in the image, the polar coordinate measurement values ​​of the radar, and the longitude and latitude information of the ADS-B are converted to the apron coordinate system. The characteristic information of the aircraft is extracted from the converted data, including the fuselage size, taxiing direction, and ground speed. The distance between the feature vectors obtained from different data sources is calculated to generate a similarity matrix, and the threshold screening is used to confirm that these data describe the same aircraft. According to the accuracy characteristics of each data source, the position and velocity of the target are weighted fused to obtain a more accurate target state estimate.

[0031] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) Extract the position coordinates of the fused target data, calculate the instantaneous latitude, longitude and altitude of the target, and obtain the target position characteristic parameters; (2) Perform time difference operation on the position characteristic parameters, calculate the instantaneous velocity and acceleration values ​​of the target, and obtain the target velocity characteristic parameters; (3) Connect the position feature parameters in time series to construct a set of continuous motion trajectory points of the target and obtain the target trajectory feature parameters; (4) Combining the position feature parameters, velocity feature parameters, and trajectory feature parameters to generate a feature vector and construct an original feature data set; (5) Standardize the parameters in the original feature data set, calculate the Pearson correlation coefficient between the parameters, and obtain the feature correlation matrix; (6) Combine the features in the feature correlation matrix whose correlation coefficients are greater than the threshold, and reconstruct the features through dimensionality reduction through principal component analysis to form a target feature matrix.

[0032] Specifically, the fused target data needs to be processed to extract the target's location coordinates, which include three dimensions: longitude, latitude, and altitude. The longitude and latitude values ​​are expressed in decimal degrees, accurate to 6 decimal places, and the altitude value is in meters, based on the WGS84 ellipsoid datum. These coordinate values ​​are extracted from the fused target data to form a description of the target's spatial location at a specific moment.

[0033] Instantaneous speed against target and acceleration The calculation adopts the time difference method, and the calculation formula is: ; in, represents the position vector at time t, is the sampling time interval, and are the high-order correction coefficients, and is the time window weight, H and J are the historical data window sizes.

[0034] The continuous motion trajectory of the target is obtained by connecting the position feature parameters in time sequence. The trajectory point set contains the position information of the target at different times. Each trajectory point contains a timestamp, coordinate value and sampling sequence number. These points are connected in time sequence to form a complete motion trajectory.

[0035] The construction of the feature vector combines three types of feature parameters: position, speed, and trajectory. The position feature contains the three-dimensional coordinates of the current moment, the speed feature contains the linear velocity and angular velocity, and the trajectory feature contains the relative position relationship of the historical trajectory points. These features constitute a multidimensional vector, and each dimension represents an attribute feature of the target. When the original feature data set is standardized, the Z-score standardization method is used to convert features of different dimensions to the same scale. The Pearson correlation coefficient is calculated between the standardized features. The Pearson correlation coefficient reflects the degree of linear correlation between the features, and the value range is between -1 and 1. The calculation results of the correlation coefficient form a feature correlation matrix.

[0036] Principal component analysis is used for feature dimensionality reduction. Feature combinations with correlation coefficients greater than a threshold are selected, and then the eigenvalues ​​and eigenvectors of the feature covariance matrix are calculated. The principal component directions with the largest contribution rates are selected, and the original features are projected to obtain the feature representation after dimensionality reduction.

[0037] For example, the position coordinates of the aircraft at a certain moment are extracted from the fused data, including longitude, latitude and altitude. Based on the continuous position sampling data, the instantaneous speed of the aircraft is calculated by time difference, including the ground speed and vertical speed components, and then the acceleration value is calculated by the change in speed. These position points are connected in chronological order to form the aircraft's glide trajectory. The feature vector contains the aircraft's current position, movement speed, acceleration and trajectory characteristics. After these features are standardized, the correlation between the features is calculated, and it is found that the speed and trajectory curvature have a strong correlation, indicating that the aircraft will reduce speed when turning. Through principal component analysis, the original high-dimensional features are reduced to several main feature components, which comprehensively reflect the motion state of the aircraft.

[0038] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) Calculate the relative distance between targets based on the position data in the target feature matrix, compare the relative distance with the safety distance threshold, and obtain the distance risk index; (2) Calculate the relative speed of the target based on the speed data in the target feature matrix, compare the relative speed with the speed threshold, and obtain the speed risk index; (3) Combine the distance risk index and the speed risk index to generate the risk state vector at the current moment and obtain the potential conflict probability of the target; (4) Conduct statistical analysis on historical risk event data, extract the frequency and degree of harm of different types of risk events, and generate historical risk weights; (5) Perform weighted calculation on the target potential conflict probability and historical risk weight, construct a comprehensive risk assessment function, and obtain the risk quantification value; (6) The risk quantification value is normalized and converted into a risk level through hierarchical mapping to obtain a safety risk score.

[0039] Specifically, the position data in the target feature matrix is ​​processed. The target feature matrix contains the position information of all targets on the apron, and the relative distance value is obtained by calculating the Euclidean distance between two targets. The safety distance threshold is the minimum interval distance determined according to the safety operation specifications of different types of targets. Different thresholds are set for aircraft and aircraft, aircraft and ground vehicles, and ground vehicles. The calculated relative distance is compared with the corresponding safety distance threshold to generate a distance risk index, which reflects the spatial proximity between the targets. For the speed data in the target feature matrix, the relative speed between any two targets is calculated. The calculation of the relative speed needs to consider the directionality of the target movement. The speed vectors of the two targets are vector subtracted to obtain the relative speed vector. The speed threshold is set based on the safety management regulations of the airport apron, including speed limit requirements in different areas and speed limits under different weather conditions. The speed risk index is generated by comparing the relative speed value with the speed threshold.

[0040] The combination of distance risk index and speed risk index constitutes the risk state vector at the current moment. The risk state vector describes the possibility of conflict between two targets. The target potential conflict probability is obtained through the prediction of the target's motion state and the path extrapolation analysis. The calculation of the potential conflict probability comprehensively considers factors such as the target's current distance, relative speed, direction of movement, and acceleration. For the analysis of historical risk event data, risk events are classified according to type, including collision accidents, intrusion into restricted areas, speeding, etc. The frequency of occurrence of each type of risk event in different time periods and different regions is counted, and its proportion in the overall risk events is calculated. The assessment of the degree of harm is based on the consequences of the event, including casualties, property losses, flight delays and other dimensions. Through the mining and analysis of historical data, each type of risk event is assigned a corresponding weight value.

[0041] The weighted calculation of target potential conflict probability and historical risk weight adopts multi-level analysis method. A risk assessment index system is established, with the current conflict probability and historical statistical data as the first-level index, each containing several second-level indicators. The weight coefficient of each indicator is determined through hierarchical analysis, and the weighted results of all indicators are combined to obtain the risk quantification value. The risk quantification value reflects the comprehensive risk level in the current scenario. The normalization of the risk quantification value adopts the piecewise linear mapping method to map the risk value to the interval from 0 to 1. The hierarchical mapping divides the risk level into multiple levels according to the airport's safety management requirements, such as low risk, medium risk, high risk, etc. Each risk level corresponds to a score interval. The normalized risk quantification value is converted into a specific risk level through interval mapping to obtain a safety risk score.

[0042] For example, taking the intersection scenario of an aircraft and a shuttle bus on the airport apron as an example, the data processing process of risk assessment is described in detail. The position data of the aircraft and the shuttle bus are extracted from the target feature matrix, the relative distance between the two is calculated, and the distance is compared with the minimum safety interval specified between the aircraft and the ground vehicle. At the same time, the relative speed of the two targets is calculated. Considering that the aircraft is taxiing and the shuttle bus is crossing the taxiway laterally, it is necessary to analyze the velocity vectors of the two. Combining the distance and speed indicators, the movement trajectories of the two targets in the next few seconds are predicted to assess the potential intersection risk. By querying the historical database, it is found that similar crossing events have occurred in the area, some of which have caused flight delays, so this type of risk event has a higher weight. Combining the current conflict probability and historical risk weight, the risk quantification value is calculated and converted into the corresponding risk level, thereby giving an accurate risk assessment result for this intersection scenario.

[0043] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) Assign weights to the security risk scores, calculate the distance relationship between targets, and obtain the spatial impact coefficient matrix; (2) Perform distance attenuation calculation on the spatial impact coefficient matrix, analyze the risk transmission intensity between adjacent targets, and obtain the risk propagation matrix; (3) Iterate the risk propagation matrix, calculate the cumulative risk value on the risk propagation chain, and obtain the risk cumulative effect value; (4) Map the spatial distribution of the risk cumulative effect value, calculate the total amount of risk accumulation per unit area, and obtain the regional risk density; (5) Organize regional risk density according to time series, construct a dynamic risk distribution map, and generate a sequence of regional situation changes; (6) Conduct spatiotemporal correlation analysis on the regional situation change sequence, integrate data based on risk level distribution, and generate real-time security situation data.

[0044] Specifically, the safety risk score is weighted, and the weight value is determined based on the target type, activity area, and time period. Different types of targets (such as aircraft, ground vehicles, and personnel) have different risk impact weights. The distance relationship between targets is obtained by calculating the Euclidean distance to form a spatial impact coefficient matrix, which describes the spatial correlation strength between all targets on the apron.

[0045] The distance attenuation calculation for the spatial influence coefficient matrix uses the following formula: ; in, represents the risk transfer intensity of target i to target j, is the benchmark transmission coefficient of the kth risk, is the distance attenuation rate, is the distance between targets, is the direction influencing factor, For environmental factors, is the barrier coefficient of the mth type obstacle, is the obstacle distribution function.

[0046] The iterative calculation process of the risk propagation matrix takes into account the chain transmission of risks between multiple targets. The initial risk source is determined, and then the cumulative effect of the risk transmitted to other targets through different paths is calculated. The risk status of the target is updated at each iteration until the risk transmission intensity is lower than the threshold or the maximum number of iterations is reached. The risk cumulative effect value reflects the combined impact of all surrounding risk sources on a certain target. The spatial distribution mapping of the risk cumulative effect value divides the apron area into grid cells. For each grid cell, the risk cumulative effect values ​​of all targets within its range are counted, and the total risk per unit area is calculated to obtain the regional risk density. The calculation of risk density takes into account the spatial distribution characteristics of the target and the scope of risk impact.

[0047] The time series organization of regional risk density sorts the risk density data in continuous time periods by timestamp. The dynamic risk distribution map visualizes the risk density and shows the spatial distribution and temporal evolution characteristics of the risk level. The regional situation change sequence records the dynamic change process of risk distribution. The spatiotemporal correlation analysis focuses on discovering the correlation pattern of risk distribution in time and space dimensions. By analyzing the changing trend of risk distribution at adjacent moments and combining the risk level distribution in different regions, the risk situation data is integrated to form a complete real-time safety situation description.

[0048] For example, during a busy period at the airport apron, multiple aircraft are active at the same time, including an aircraft A being pushed out, an aircraft B being taxied, and several ground support vehicles. Based on their safety risk scores, aircraft A is given a higher weight because it is in the process of being pushed out, followed by aircraft B, and ground vehicles have a lower weight. The distance relationship between these targets is calculated to generate a spatial impact coefficient matrix. The matrix is ​​calculated for distance attenuation, and the actual risk propagation intensity is obtained by taking into account the barrier effect of fixed facilities on the apron on risk propagation. Through multiple rounds of iterative calculations, the process of risk propagation from aircraft A to surrounding targets is analyzed. The risk impact on aircraft B decays with increasing distance, but since it is also a risk source, a secondary risk accumulation area is formed around it. These risk accumulation effects are mapped to the apron grid, and the risk density distribution of each area is calculated. Over time, the change process of risk density is recorded, and it is found that the risk distribution shows obvious migration characteristics with the movement of aircraft. Through spatiotemporal correlation analysis, these dynamically changing information are integrated into complete safety situation data.

[0049] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) Analyze the real-time security situation data, extract the risk level and spatial distribution information, and obtain the current situation characteristics; (2) Substitute the current situation characteristics into the safety control function, calculate the risk control cost and benefit ratio, and obtain the objective function value; (3) Setting constraints on the objective function value, converting the safety distance requirement and speed limit into mathematical constraints to obtain a set of constraint equations; (4) Combining the constraint equations with the objective function value to construct an optimization problem, calculating the optimal solution through gradient descent, and obtaining the initial control parameters; (5) Conduct conflict tests on the initial control parameters, analyze the degree of mutual influence between the parameters, and obtain the feasible control range; (6) Refine and adjust the parameters within the feasible control range, select the optimal control scheme, and form the management and control strategy parameters.

[0050] Specifically, when processing real-time safety situation data, risk level information is extracted. The risk level includes three levels: high, medium, and low, and each level corresponds to a different numerical range. The spatial distribution information records the distribution status of the risk in the apron area, including the coordinate position and impact range of the risk point. The current situation characteristic quantity is composed of multiple indicators such as risk level, distribution density, and spatial aggregation. The construction of the safety control function needs to consider multiple factors, including the implementation cost of the control measures, the expected reduction in risk value, and the difficulty of implementation. The cost of risk control includes human resource investment, equipment scheduling cost, and operating efficiency impact. The benefits are reflected in the degree of risk reduction and the improvement of safety margin. The objective function value is obtained by calculating the cost-benefit ratio, which reflects the comprehensive advantages and disadvantages of the control plan.

[0051] The setting of constraints is based on airport operation specifications and safety management requirements. The safety distance requirement stipulates the minimum interval that must be maintained between different types of targets, and the speed limit sets the maximum allowable speed of various types of vehicles and aircraft in different areas. These physical constraints are converted into mathematical expressions to form a set of constraint equations. The set of constraint equations contains equality constraints and inequality constraints to ensure that the control plan meets safety requirements. The construction of the optimization problem combines the objective function and the set of constraint equations, and uses the gradient descent method to solve the optimal solution. During the gradient descent process, the control parameters are randomly initialized, and then the gradient of the objective function to each parameter is calculated, and the parameter value is updated along the gradient direction until it converges to the local optimal solution. The initial control parameters include specific control quantities such as speed adjustment and path offset.

[0052] During the conflict check of the initial control parameters, it is necessary to analyze the correlation between different control parameters. The speed adjustment of a certain target will affect its relative position relationship with other targets, and the path adjustment will change the intersection timing of multiple targets. By establishing a parameter influence relationship diagram, the positive promotion and negative inhibition relationship between parameters is identified, and the feasible value range of each parameter is determined. When making detailed adjustments to parameters within the feasible control range, the grid search method is used to divide the value range of each parameter into several discrete points, and the optimal parameter combination is found through exhaustive combination or heuristic search. The optimization objectives include multiple dimensions such as risk control effect, operating efficiency, and implementation difficulty. The plan with the highest comprehensive score is selected as the management and control strategy parameter.

[0053] For example, real-time safety situation data shows that there is a high-risk area near the No. 1 aircraft stand, where there is an aircraft about to be launched, a refueling truck and two luggage carts. The risk level and spatial distribution information show that the risk mainly comes from the conflict between the refueling operation and the aircraft launch sequence. According to the current situation characteristic quantity, the cost-benefit ratio of different control schemes is calculated, including options such as adjusting the operation time of the refueling truck and changing the direction of the aircraft launch. The constraint condition requires that the refueling truck and the aircraft maintain a safe distance of no less than 5 meters, and the aircraft launch speed does not exceed 5 kilometers per hour. After converting these constraints into mathematical expressions, the initial control parameters are obtained through optimization calculation: postpone the aircraft launch time. This initial scheme is tested for conflicts, and the impact on the subsequent flight support work is analyzed to obtain a feasible solution set containing the time adjustment range. Finally, through detailed adjustments, the coordination of multiple goals is comprehensively considered to form a specific control strategy: postpone the aircraft launch for 2 minutes, and adjust the working order of other support vehicles at the same time, which not only ensures the safety interval, but also reduces the impact on the overall operation efficiency.

[0054] In a specific embodiment, the process of performing the conflict check step on the initial control parameters may specifically include the following steps: (1) Perform numerical range analysis on the initial control parameters, count the upper and lower limits of each parameter, and obtain the parameter variation range; (2) Cross-combining the parameter change intervals, calculating the control effects under different parameter combinations, and obtaining the parameter correlation matrix; (3) Perform sensitivity analysis on the parameter correlation matrix, calculate the impact of parameter changes on the control effect, and obtain the parameter sensitivity value; (4) Arrange the parameter sensitivity values ​​in descending order, identify the key influencing parameters and secondary influencing parameters, and obtain the parameter importance sequence; (5) Taking the parameter importance sequence as the weight coefficient, weighted constraints are imposed on the parameter value range to obtain the parameter adjustment boundary; (6) Verify the feasibility of the parameter adjustment boundary, eliminate the value intervals that do not meet the constraints, and obtain the feasible control range.

[0055] Specifically, when analyzing the numerical range of the initial control parameters, the physical meaning and actual constraints of the parameters need to be considered. The control parameters include speed adjustment, distance threshold, time interval, etc., and each parameter has a reasonable range of values. The speed adjustment is determined by the airport operation specifications, such as the allowable range of taxiing speed; the distance threshold is determined based on the safety interval requirements; the time interval needs to consider the flight plan and the requirements of the support operation. The historical data of each parameter is statistically analyzed, the maximum and minimum values ​​are extracted, and the reasonable range of parameter changes is determined in combination with expert experience. The cross combination of parameter change intervals uses a grid search method to divide the range of each parameter into equal intervals to form a discrete set of value points. The value points of different parameters are fully combined, and each combination corresponds to a set of specific control parameter values. Simulation calculations are performed on each set of parameter combinations to evaluate their impact on the control objectives, such as the degree of reduction in risk levels and changes in operating efficiency. The evaluation results are organized into a parameter association matrix, and the matrix elements represent the contribution of the parameter combination to the control effect.

[0056] The sensitivity analysis of the parameter association matrix adopts the single factor change method, fixing other parameters unchanged, changing the value of one parameter alone, and observing the change range of the control effect. The sensitivity value of the parameter is obtained by calculating the partial derivative of the control effect with respect to the parameter change. The larger the sensitivity value, the more significant the influence of the parameter on the control effect. Repeat this process for all parameters to obtain a complete set of parameter sensitivity values. The descending order of parameter sensitivity values ​​sorts all parameters from large to small according to their influence intensity. The parameters ranked in front play a decisive role in the control effect and are defined as key influencing parameters; the parameters ranked behind have relatively small influences and are called secondary influencing parameters. This division helps to determine the priority of parameter adjustment in actual control and form a parameter importance sequence.

[0057] When the parameter importance sequence is converted into a weight coefficient, normalization is used so that the sum of all weights is equal to 1. Parameters with larger weight coefficients have higher priority in the adjustment process, and their value ranges are subject to stricter constraints. The original value range of each parameter is multiplied by the corresponding weight coefficient to obtain the weighted parameter adjustment boundary. The feasibility verification of the parameter adjustment boundary includes static constraint verification and dynamic conflict verification. The static constraint verification ensures that the parameter values ​​meet the physical limitations and operating specifications, and the dynamic conflict verification verifies the feasibility of the parameter combination in actual operation. By eliminating the value intervals that do not meet the constraints, the feasible control range is obtained.

[0058] Taking the optimization of control parameters for coordination between aircraft pushback and ground support vehicles as an example, the pushback time, pushback speed, and ground vehicle avoidance distance parameters are involved. According to the airport operation specifications, the pushback time adjustment range is plus or minus 5 minutes, the pushback speed range is 2-5 kilometers per hour, and the ground vehicle avoidance distance is not less than 5 meters. These restrictions constitute the initial range of parameter changes. These parameters are combined for analysis, such as a 2-minute pushback time delay with a pushback speed of 3 kilometers per hour, to evaluate the impact of each combination on the overall operation. Through sensitivity analysis, it is found that the pushback time adjustment has the greatest impact on conflict avoidance, followed by the avoidance distance, and the pushback speed has a relatively small impact. Based on this, the parameter importance ranking is obtained: pushback time adjustment, avoidance distance, and pushback speed. This importance sequence is converted into a weight coefficient to impose weighted constraints on the adjustment range of the parameters. Finally, through feasibility verification, the parameter adjustment plan is determined by comprehensively considering constraints such as the parking space occupancy time and taxiway congestion.

[0059] The above describes the multi-data fusion method in the airport apron safety supervision in the embodiment of the present application. The following describes the multi-data fusion system in the airport apron safety supervision in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of a multi-data fusion system for airport apron safety supervision includes: A generation module 201 is used to generate a data credibility value according to data integrity and time consistency for the video data, radar data and location data collected by multi-source sensors on the apron, and to filter and denoise the original data by the data credibility value to obtain a preprocessed data set; Fusion module 202, used to establish a unified spatiotemporal reference for the preprocessed data set through timestamp and coordinate conversion, calculate the matching degree between different data sources according to the similarity of target features, and obtain fused target data; The extraction module 203 is used to extract characteristic parameters from the three dimensions of position, speed and motion trajectory according to the fused target data, and to screen and combine the characteristic parameters according to the correlation coefficient to form a target characteristic matrix; The judgment module 204 is used to judge potential conflicts based on the target feature matrix by setting the safety distance threshold and speed threshold, and to construct a risk assessment index in combination with the historical risk event data to obtain a safety risk score; Establishing module 205, for establishing spatial correlation and risk propagation relationships between apron targets according to the safety risk scores, calculating regional risk density through risk accumulation effects, and generating real-time safety situation data; The solution module 206 is used to calculate the objective function value of security control based on real-time security situation data, set security constraints for optimization and solution, and form control strategy parameters.

[0060] Through the collaborative cooperation of the above-mentioned components, the original data is filtered and denoised through the data credibility value, which solves the problem of quality differences in multi-source heterogeneous data and improves data reliability; secondly, a unified spatiotemporal benchmark is established by using timestamps and coordinate transformations, and the matching degree between data sources is calculated based on the similarity of target features, so as to achieve accurate alignment and association between different data sources; thirdly, feature parameters are extracted from the three dimensions of position, speed, and motion trajectory, and correlation analysis is performed to construct a multi-dimensional target feature description, which enhances the expression ability of the target state; potential conflicts are judged by setting safety distance thresholds and speed thresholds, and an evaluation index system is constructed in combination with historical risk event data, so as to achieve accurate quantification of risks; at the same time, by establishing spatial correlation relationships and risk propagation relationships between apron targets, the concept of risk accumulation effect is introduced, and the spatial distribution and propagation characteristics of risks are accurately characterized; finally, the objective function value is calculated based on real-time safety situation data and constraints are set, and the optimization algorithm is used to solve the control strategy parameters, so as to achieve multi-objective collaborative optimization under the premise of ensuring safety.

[0061] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the multi-data fusion method in the airport apron safety supervision.

[0062] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0063] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0064] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A multi-data fusion method for airport apron safety supervision, characterized in that: The multi-data fusion method in the airport apron safety supervision includes: For the video data, radar data and location data collected by multi-source sensors on the apron, a data credibility value is generated according to the data integrity and time consistency. The raw data is filtered and denoised by the data credibility value to obtain a preprocessed data set. For the preprocessed data set, a unified spatiotemporal benchmark is established through timestamp and coordinate conversion, and the matching degree between different data sources is calculated based on the similarity of target features to obtain the fused target data; According to the fused target data, feature parameters are extracted from the three dimensions of position, speed, and motion trajectory, and the feature parameters are screened and combined according to the correlation coefficient to form a target feature matrix; Based on the target feature matrix, potential conflicts are judged by setting safety distance thresholds and speed thresholds, and risk assessment indicators are constructed in combination with historical risk event data to obtain safety risk scores; According to the safety risk score, the spatial correlation and risk propagation relationship between apron targets are established, the regional risk density is calculated through the risk accumulation effect, and real-time safety situation data is generated; Based on real-time security situation data, calculate the objective function value of security control, set security constraints for optimization and solution, and form control strategy parameters.

2. The multi-data fusion method for airport apron safety supervision according to claim 1 is characterized in that: The video data, radar data and location data collected by the multi-source sensors on the apron are used to generate a data credibility value according to data integrity and time consistency, and the raw data is filtered and denoised by the data credibility value to obtain a preprocessed data set, including: Segmenting the video data according to a fixed time window, dividing the segmented video data into an image frame sequence, calculating the data integrity according to the pixel grayscale value in the image frame sequence, and obtaining a video data integrity index; Performing sampling rate detection on the radar data, extracting the signal waveform in the radar data, counting the number of signal interruptions according to the continuity of the waveform, and obtaining a radar data integrity index; Perform time stamp analysis on the location data, count the sampling interval variance of the location data, determine the stability of data collection according to the variance, and obtain the location data integrity index; Performing a weighted summation of the video data integrity index, the radar data integrity index, and the location data integrity index to generate a comprehensive integrity score; Comparing the time stamps of the video data, radar data, and location data, calculating the deviation value of the data arrival time, and generating a time consistency score; The data credibility value is calculated according to the weighted combination of the comprehensive integrity score and the time consistency score, the data is filtered through a set threshold, and the noise points are removed to obtain a preprocessed data set.

3. The multi-data fusion method for airport apron safety supervision according to claim 1 is characterized in that: The preprocessed data set is used to establish a unified spatiotemporal reference through timestamp and coordinate conversion, and the matching degree between different data sources is calculated according to the similarity of target features to obtain the fused target data, including: Parsing the timestamps in the preprocessed data set, aligning the sampling times of different data sources to a unified time axis, and generating a time synchronization sequence; Performing a reference transformation on the coordinate data in the time synchronization sequence, mapping the position information in different coordinate systems to a unified apron coordinate system, and obtaining spatial alignment data; Extracting the geometric size, motion direction and speed information of the target from the spatial alignment data, and constructing a target feature description set; Calculate the Euclidean distance between feature vectors based on the target feature description set to generate a similarity matrix between data sources; The similarity matrix is ​​normalized, and the best matching pair is selected by setting a matching threshold to form a target association table; The multi-source data are merged according to the target association table, and the target parameters are fused using a weighted average method to obtain fused target data.

4. The multi-data fusion method for airport apron safety supervision according to claim 1 is characterized in that: According to the fused target data, feature parameters are extracted from the three dimensions of position, speed, and motion trajectory, and the feature parameters are screened and combined according to the correlation coefficient to form a target feature matrix, including: Extracting the position coordinates of the fused target data, calculating the instantaneous latitude, longitude and altitude of the target, and obtaining the target position characteristic parameters; Performing time difference operation on the position characteristic parameters, calculating the instantaneous speed and acceleration value of the target, and obtaining the target speed characteristic parameters; Performing temporal connection on the position characteristic parameters to construct a continuous motion trajectory point set of the target and obtain the target trajectory characteristic parameters; The position characteristic parameters, speed characteristic parameters and trajectory characteristic parameters are combined to generate a characteristic vector, and an original characteristic data set is constructed; Standardizing the parameters in the original feature data set, calculating the Pearson correlation coefficient between the parameters, and obtaining a feature correlation matrix; The features with correlation coefficients greater than a threshold in the feature correlation matrix are combined, and the features are reconstructed by reducing their dimensionality through principal component analysis to form a target feature matrix.

5. The multi-data fusion method for airport apron safety supervision according to claim 1 is characterized in that: Based on the target feature matrix, potential conflicts are judged by setting safety distance thresholds and speed thresholds, and risk assessment indicators are constructed in combination with historical risk event data to obtain safety risk scores, including: Calculating the relative distance between targets based on the position data in the target feature matrix, comparing the relative distance with the safety distance threshold, and obtaining a distance risk index; Calculating the relative speed of the target based on the speed data in the target feature matrix, comparing the relative speed with the speed threshold, and obtaining a speed risk index; The distance risk index and the speed risk index are combined to generate a risk state vector at the current moment to obtain a potential conflict probability of the target; Conduct statistical analysis on the historical risk event data, extract the occurrence frequency and degree of harm of different types of risk events, and generate historical risk weights; Perform weighted calculation on the target potential conflict probability and the historical risk weight, construct a comprehensive risk assessment function, and obtain a risk quantification value; The risk quantification value is normalized, converted into a risk level through hierarchical mapping, and a security risk score is obtained.

6. The multi-data fusion method for airport apron safety supervision according to claim 1 is characterized in that: According to the safety risk score, the spatial correlation relationship and risk propagation relationship between the apron targets are established, the regional risk density is calculated through the risk accumulation effect, and real-time safety situation data is generated, including: Assigning weights to the security risk scores, calculating distance relationships between targets, and obtaining a spatial impact coefficient matrix; Performing distance attenuation calculation on the spatial impact coefficient matrix, analyzing the risk transfer intensity between adjacent targets, and obtaining a risk propagation matrix; Iterate the risk propagation matrix, calculate the cumulative risk value on the risk propagation chain, and obtain the risk cumulative effect value; Perform spatial distribution mapping on the risk cumulative effect value, calculate the total risk accumulation within a unit area, and obtain the regional risk density; Organizing the regional risk density according to time series, constructing a dynamic risk distribution map, and generating a regional situation change sequence; The temporal and spatial correlation analysis of the regional situation change sequence is carried out, and data integration is performed in combination with the risk level distribution to generate real-time security situation data.

7. The multi-data fusion method for airport apron safety supervision according to claim 1 is characterized in that: The objective function value of security control is calculated for real-time security situation data, security constraints are set for optimization and solution, and control strategy parameters are formed, including: Analyze the real-time security situation data, extract risk level and spatial distribution information, and obtain current situation feature quantity; Substituting the current situation characteristic quantity into the safety control function, calculating the risk control cost and benefit ratio, and obtaining the objective function value; Setting constraint conditions for the objective function value, converting the safety distance requirement and the speed limit into mathematical constraints, and obtaining a set of constraint equations; The constraint equations are combined with the objective function value to construct an optimization problem, and the optimal solution is calculated by gradient descent to obtain initial control parameters; Performing a conflict check on the initial control parameters, analyzing the mutual influence between the parameters, and obtaining a feasible control range; The parameters within the feasible control range are refined and adjusted, the optimal control scheme is selected, and the management and control strategy parameters are formed.

8. The multi-data fusion method for airport apron safety supervision according to claim 7 is characterized in that: The conflict check of the initial control parameters is performed, the mutual influence between the parameters is analyzed, and the feasible control range is obtained, including: Performing numerical range analysis on the initial control parameters, counting the upper and lower limits of the values ​​of each parameter, and obtaining the parameter variation interval; Cross-combining the parameter variation intervals, calculating the control effects under different parameter combinations, and obtaining a parameter correlation matrix; Perform sensitivity analysis on the parameter association matrix, calculate the influence intensity of parameter changes on the control effect, and obtain the parameter sensitivity value; Arrange the parameter sensitivity values ​​in descending order, identify key influencing parameters and secondary influencing parameters, and obtain a parameter importance sequence; The parameter importance sequence is used as a weight coefficient to perform weighted constraints on the parameter value range to obtain the parameter adjustment boundary; The feasibility of the parameter adjustment boundary is verified, and the value intervals that do not meet the constraint conditions are eliminated to obtain a feasible control range.

9. A multi-data fusion system for airport apron safety supervision, used to implement the multi-data fusion method for airport apron safety supervision as described in any one of claims 1 to 8, characterized in that: The multi-data fusion system in the airport apron safety supervision includes: A generation module is used to generate data credibility values ​​based on data integrity and time consistency for video data, radar data, and location data collected by multi-source sensors on the apron, and to filter and denoise the original data using the data credibility values ​​to obtain a preprocessed data set; The fusion module is used to establish a unified spatiotemporal benchmark for the preprocessed data set through timestamp and coordinate conversion, calculate the matching degree between different data sources based on the similarity of target features, and obtain the fused target data; The extraction module is used to extract feature parameters from the three dimensions of position, speed and motion trajectory according to the fused target data, and to screen and combine the feature parameters according to the correlation coefficient to form a target feature matrix; The judgment module is used to judge potential conflicts based on the target feature matrix, through the set safety distance threshold and speed threshold, and to construct risk assessment indicators in combination with historical risk event data to obtain a safety risk score; Establish a module to establish the spatial correlation and risk propagation relationship between apron targets based on the safety risk score, calculate the regional risk density through the risk accumulation effect, and generate real-time safety situation data; The solution module is used to calculate the objective function value of security control based on real-time security situation data, set security constraints for optimization and solution, and form control strategy parameters.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the multi-data fusion method for airport apron safety supervision as described in any one of claims 1-8 is implemented.

Images