Rapid and accurate decision method and system based on training space terrain environment, and electronic equipment

Abstract

The invention relates to the technical field of training, in particular to a rapid and accurate decision method and system based on a training space terrain environment and electronic equipment, and the method comprises the steps: collecting elevation point cloud, a ground cover contour line and underground blindage structure data of a training space; generating a terrain grid model containing the gradient, the intervisibility and the masking coefficient; establishing a decision rule engine based on the terrain grid model; beidou positioning data and millimeter wave radar scanning data of the training equipment are received, and data matching is carried out; and calling a decision rule engine to calculate the motion feasibility of the equipment, the fire strike effective radius and the electromagnetic interference attenuation value. According to the invention, by collecting and processing the terrain data and combining the Kalman filtering algorithm to optimize equipment positioning, efficient fusion of the terrain information and the equipment state is realized, so that the real-time performance, the accuracy and the reliability of commanding and deciding are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of military training, and in particular to a method and system for rapid and accurate decision-making based on terrain environment of training space, and an electronic device. BACKGROUND

[0002] In the process of military training, the position information and environmental state of training equipment are continuously and dynamically changing, which puts higher requirements on the real-time response capability of the system. However, the existing training system generally has a large time delay between real-time data acquisition, analysis and decision output, which is difficult to meet the demand for rapid decision-making in a dynamic environment. In addition, due to the lack of high-precision modeling and dynamic adaptation mechanism for complex terrain, the system cannot effectively integrate the influence of terrain, equipment state and external environmental factors, resulting in deviation of the decision result, and even misleading tactical instructions due to data errors.

[0003] Therefore, how to use modern geographic information technology (such as GIS, remote sensing, three-dimensional terrain modeling, etc.) and real-time equipment data to build an efficient, accurate and dynamically adaptive behavior decision-making method for training equipment has become a key direction to improve the intelligent level and command decision-making accuracy of the training system. SUMMARY

[0004] In view of the above, the present application provides a method and system for rapid and accurate decision-making based on terrain environment of training space, and an electronic device, which improves the real-time and preparedness of behavior decision-making of training equipment in complex and dynamically changing training scenarios.

[0005] The method for rapid and accurate decision-making based on terrain environment of training space provided by the present application comprises the following steps:

[0006] S1: using laser radar and infrared remote sensing technology to collect elevation point cloud, ground cover contour line and underground shelter structure data of the training space, and generating a terrain feature data set;

[0007] S2: inputting the terrain feature data set into a spatial interpolation algorithm to generate a terrain grid model containing slope, visibility rate and masking coefficient, wherein each grid cell stores its corresponding terrain influence factor matrix;

[0008] S3: based on the terrain grid model, establishing a decision rule engine containing a fire coverage correction algorithm, a path passability discrimination function and an electromagnetic environment interference compensation value;

[0009] S4: receiving Beidou positioning data and millimeter wave radar scanning data of the training equipment, and matching the data through Kalman filtering algorithm to realize registration with the terrain grid model;

[0010] S5: According to the matched spatial coordinates, the terrain influence factor matrix of the corresponding grid cell is extracted, and a decision rule engine is called to calculate the movement feasibility, fire attack effective radius and electromagnetic interference attenuation value of the equipment.

[0011] Preferably, the S1 specifically comprises:

[0012] S11: Scanning the training space using laser radar technology to obtain elevation point cloud data, wherein the laser radar measures the three-dimensional coordinates of the point position by emitting laser pulses and receiving reflected signals to generate a ground point cloud data set, and the point cloud data contains the elevation information of the terrain surface;

[0013] S12: Using infrared remote sensing technology, the ground cover contour line data of the training space is obtained by an infrared sensor, which uses the principle of thermal radiation to detect the temperature difference of the object surface, identify the different characteristics of the ground cover, and generate the distribution and contour line data of the cover;

[0014] S13: Combined with underground detection technology, the underground shelter structure data is obtained by ground penetrating radar, which generates an image or data of the underground structure by emitting electromagnetic waves and receiving reflected signals, providing the spatial position and structural characteristics of the underground shelter;

[0015] S14: Merge and process the elevation point cloud, ground cover contour line and underground shelter structure data collected in steps S11, S12 and S13, and convert them into a standard format to generate a terrain feature data set.

[0016] Preferably, the S2 specifically comprises:

[0017] S21: Input the terrain feature data set generated in S1 into the spatial interpolation algorithm, and use the Kriging interpolation method to perform spatial interpolation processing on the terrain data to convert the discrete terrain feature data into continuous spatial data;

[0018] S22: During the interpolation process, the elevation value and terrain feature data of each grid cell are determined according to the spatial distribution and local variation of the terrain feature data;

[0019] S23: Based on the interpolated elevation data, the slope of each grid cell in the terrain grid model is calculated, and the slope value is determined by calculating the elevation difference between adjacent grid cells, and the triangular subdivision method is used to calculate the slope of each grid cell;

[0020] S24: Calculate the visibility rate of each grid cell, and determine whether the line of sight of each grid cell is blocked by other terrain cells through the ray tracing algorithm, and thus calculate the visibility rate value of the cell;

[0021] S25: Calculate the masking coefficient of each grid cell, and calculate the masking coefficient value of each grid cell by using the shielding algorithm according to the slope of the grid cell and the elevation data of the adjacent area;

[0022] S26: Integrate the slope, visibility rate and masking coefficient calculated in steps S23 to S25 into the terrain grid model, and assign a corresponding terrain influence factor matrix to each grid cell to form a final terrain grid model.

[0023] Preferably, the S21 specifically comprises:

[0024] S211: Organize the discrete point data in the terrain feature data set generated in S1 according to the spatial coordinates, assign an elevation value to each data point, and preprocess the data points to remove abnormal data and noise;

[0025] S212: Determine the weight function of spatial interpolation, and use the covariance function in the Kriging interpolation method to quantify the spatial correlation between different data points. The calculation formula of the covariance function is: wherein, represents the variogram value at a distance of h, represents the elevation difference between data points and ;

[0026] S213: Determine the spatial weight of each point to the surrounding points through the calculation of the variogram. Specifically, for each pair of data points and , calculate the spatial distance and obtain the weight through the covariance function. The weight calculation formula is: wherein, is the spatial weight between data points and , is the variogram value at a distance of h;

[0027] S214: Use the weighted average method to perform interpolation calculation on the elevation value to obtain continuous elevation data after spatial interpolation. The formula is: wherein, is the elevation value at the point to be interpolated, is the spatial weight of the point to be interpolated and the known point , is the elevation value of the known point , is the number of data points.

[0028] Preferably, the S3 specifically comprises: ​​​

[0029] S31: Based on the slope, visibility rate and masking coefficient of each grid cell in the terrain grid model, a fire coverage correction algorithm is established, and the correction formula is represented as: R 修正 =R 原始 ×(1-α×slope), wherein R 修正 is the corrected fire coverage, R 原始 is the uncorrected fire coverage, and a is the correction coefficient of the slope on the fire coverage;

[0030] S32: A path passability discrimination function is established to determine whether the passable path of the training equipment in the terrain is feasible, and the discrimination function expression is: , wherein is the path passability score, is the weight of each grid cell, is the number of grid cells, are the slope, visibility rate and masking coefficient of the cell respectively, is the path passability function;

[0031] S33: Combined with the influence of electromagnetic environment, a calculation method of electromagnetic environment interference compensation value is established, and the electromagnetic interference compensation formula is: E 补偿 =E 原始 ×(1-β⋅visibility rate), wherein E 补偿 is the compensation value of the electromagnetic environment, E 原始 is the electromagnetic interference value without considering the terrain, β is the electromagnetic interference correction coefficient, and the visibility rate is the visibility rate value of the cell in the terrain grid model;

[0032] S34: The fire coverage correction algorithm, path passability discrimination function and electromagnetic environment interference compensation value are combined to establish a complete decision rule engine; and the comprehensive decision formula is: C 裁决 =α1⋅R 修正 +α2⋅P 通行 +α3⋅E 补偿 , wherein C_ decision is the comprehensive decision result, and α1, α2 and α3 are the weighting coefficients of the factors.

[0033] Preferably, the S4 specifically comprises:

[0034] S41: Receiving Beidou positioning data and millimeter wave radar scanning data of the training equipment, wherein the Beidou positioning data includes real-time coordinates and attitude information of the equipment, and the millimeter wave radar scanning data includes obstacle positions around the equipment and radar echo information of the surrounding environment;

[0035] S42: Preprocessing the received Beidou positioning data to remove noise and outliers;

[0036] S43: processing the received millimeter wave radar scan data, converting the radar echo signal into distance and angle information, and converting into spatial data in the standard Cartesian coordinate system through polar coordinate conversion;

[0037] S44: inputting the processed Beidou positioning data and millimeter wave radar scan data into the Kalman filtering algorithm, fusing the two by using the Kalman filtering algorithm, and optimizing the estimation result by the Kalman filtering algorithm through the prediction and update steps according to the weighted fusion of prior information and measurement data;

[0038] S45: matching the Beidou positioning data and the millimeter wave radar scan data according to the estimated position and state information updated by the Kalman filtering algorithm, and registering them with the terrain grid model.

[0039] Preferably, the S45 specifically comprises:

[0040] S451: obtaining the estimated position and state information of the equipment updated by the Kalman filtering algorithm, including the position of the equipment and its motion state;

[0041] S452: matching the updated equipment position and state information with the terrain grid model, specifically, converting the accurate coordinates obtained by the Kalman filtering into the index of the terrain grid model, and mapping the estimated position of the equipment into the grid cell by using the space mapping algorithm to ensure the alignment of the position with the terrain grid model;

[0042] S453: associating the millimeter wave radar scan data with the updated equipment position, and judging the relationship between the equipment and the surrounding environment through the obstacle position and echo information in the radar data;

[0043] S454: registering the matched Beidou positioning data, millimeter wave radar scan data and terrain grid model by mapping the equipment position and radar data into the corresponding terrain grid cell to complete the registration of the equipment and the terrain grid model;

[0044] S455: after completing the registration of the equipment and the terrain grid model, outputting the final registration result to provide the equipment positioning information for the subsequent decision-making process.

[0045] Preferably, the S5 specifically comprises:

[0046] S51: determining the specific position of the equipment in the terrain grid model according to the matched spatial coordinates obtained in S4, and determining the grid cell where the position is located;

[0047] S52: extracting the terrain influence factor matrix of the grid cell corresponding to the equipment position, including the slope, visibility rate and masking coefficient information of the corresponding cell grid cell;

[0048] S53: Call the decision rule engine, calculate the movement feasibility, fire attack effective radius and electromagnetic interference attenuation value of the equipment respectively based on the extracted terrain influence factor matrix, and provide support for subsequent command decision.

[0049] The rapid and accurate decision system based on the terrain environment of the training space provided by the application is used to realize the rapid and accurate decision method based on the terrain environment of the training space, and comprises the following modules.

[0050] The information acquisition module is used to acquire the elevation point cloud, ground cover contour line and underground shelter structure data of the training space in real time through laser radar and infrared remote sensing technology, and generate terrain feature data set from the acquired data.

[0051] The data processing module is used to input the terrain feature data set into the spatial interpolation algorithm, and generate the terrain grid model containing slope, visibility rate and masking coefficient through the Kriging interpolation method.

[0052] The decision rule engine module is used to calculate the movement feasibility, fire attack effective radius and electromagnetic interference attenuation value of the equipment in turn based on the terrain influence factor matrix in the terrain grid model, combine the real-time positioning data of the training equipment, and through the fire coverage range correction algorithm, path trafficability discrimination function and electromagnetic interference compensation algorithm.

[0053] The data fusion module is used to receive the Beidou positioning data and millimeter wave radar scanning data of the training equipment, match the data through the Kalman filtering algorithm, calculate the position of the equipment and register with the terrain grid model, and ensure the accuracy of the position data.

[0054] The decision output module is used to output accurate decision information according to the calculation result of the decision rule engine and the real-time state of the equipment, and support the command decision of the equipment.

[0055] An electronic device comprises a processor and a storage device, and the storage device stores a computer program, which is used to execute the steps of the rapid and accurate decision method based on the terrain environment of the training space when the computer program is run by the processor.

[0056] The application has the advantages that the terrain data of the training space is acquired in real time by combining laser radar and infrared remote sensing technology, and a continuous terrain grid model is generated by using the Kriging interpolation method, so that the problem of inaccurate terrain modeling and update lag in the traditional method is solved; through fine processing of the terrain data, the change of complex terrain can be accurately captured, more accurate geographic information support is provided for subsequent decision, and the reliability and accuracy of the command decision are significantly improved.

[0057] This invention fuses equipment positioning data and millimeter-wave radar scanning data using the Kalman filtering algorithm, optimizing position and status estimation and solving the problem of large position errors in traditional methods. By combining high-precision terrain information with real-time equipment status, it can adjust command decisions in real time in dynamically changing training environments, improving operational efficiency and responsiveness, and effectively avoiding erroneous decisions caused by terrain or environmental errors. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of a rapid and accurate adjudication method according to an embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram of a rapid and accurate adjudication system according to an embodiment of the present invention. Detailed Implementation

[0060] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0061] like Figure 1 , 2 As shown, the rapid and accurate adjudication method based on the terrain environment of the training space includes the following steps:

[0062] S1: Use lidar and infrared remote sensing technology to collect elevation point cloud, surface cover outline and underground bunker structure data of the training space to generate terrain feature dataset;

[0063] S2: Input the terrain feature dataset into the spatial interpolation algorithm to generate a terrain grid model containing slope, visibility and masking coefficient, where each grid cell stores its corresponding terrain influence factor matrix;

[0064] S3: Based on the terrain grid model, establish a decision rule engine that includes a fire coverage range correction algorithm, a path traversability discrimination function, and an electromagnetic environment interference compensation value;

[0065] S4: Receives BeiDou positioning data and millimeter-wave radar scanning data from training equipment, and performs data matching through the Kalman filtering algorithm to achieve registration with the terrain grid model;

[0066] S5: Based on the matched spatial coordinates, extract the terrain influence factor matrix of the corresponding grid cell, and call the adjudication rule engine to calculate the equipment's movement feasibility, effective fire strike radius, and electromagnetic interference attenuation value.

[0067] S1 specifically includes:

[0068] S11: Scanning the training space using LiDAR technology to obtain elevation point cloud data, wherein the LiDAR measures the three-dimensional coordinates of the point by emitting laser pulses and receiving reflected signals, generates a high-precision ground point cloud dataset, and the point cloud data contains the elevation information of the terrain surface;

[0069] S12: Using infrared remote sensing technology, obtaining the ground cover contour line data of the training space through infrared sensors, the infrared sensors use the principle of thermal radiation to detect the temperature difference of the object surface, identify the different characteristics of the ground cover, and generate the distribution and contour line data of the cover;

[0070] S13: Combining the underground detection technology, obtaining the underground shelter structure data through the ground penetrating radar (GPR), the ground penetrating radar generates an image or data of the underground structure by emitting electromagnetic waves and receiving reflected signals, providing the spatial position and structural characteristics of the underground shelter;

[0071] S14: Merging and processing the elevation point cloud, ground cover contour line and underground shelter structure data collected in steps S11, S12 and S13, and converting them into a standard format to generate a complete terrain feature dataset, providing accurate basic data for the generation of subsequent terrain grid models; the above steps combine the use of laser radar, infrared remote sensing and underground detection technology, which can accurately collect the elevation, ground and underground information of the training space, providing an efficient and reliable data collection process, providing comprehensive and accurate data support for the subsequent terrain grid model generation, which helps to improve the accuracy and efficiency of the entire decision-making process.

[0072] S2 specifically includes:

[0073] S21: Inputting the terrain feature dataset generated in S1 into the spatial interpolation algorithm, and using the Kriging interpolation method to perform spatial interpolation processing on the terrain data, the Kriging method converts discrete terrain feature data into continuous spatial data by considering the spatial autocorrelation between sample points;

[0074] S22: In the interpolation process, according to the spatial distribution and local variation law of the terrain feature data, the elevation value and terrain feature data of each grid cell are determined, which provides the basis for subsequent calculation of slope, visibility rate and masking coefficient;

[0075] S23: Based on the interpolated elevation data, the slope of each grid cell in the terrain grid model is calculated, the slope value is determined by calculating the elevation difference between adjacent grid cells, and the triangular subdivision method is used to calculate the slope of each grid cell; the slope calculation formula is as follows: slope , wherein, is the elevation difference between adjacent grid cells, The slope value is obtained by calculating the elevation difference and horizontal distance between two grid cells using the triangulation method to determine the horizontal distance between them.

[0076] S24: Calculate the view fetch of each grid cell. View fetch refers to the visible area between grid cells from a specific viewpoint. Using a ray tracing algorithm, determine whether the view of each grid cell is obstructed by other terrain cells, and then calculate the view fetch value for that cell. The view fetch calculation formula is as follows: View Fetch ,in, This represents the number of grid cells in the visible area within a single grid cell. This represents the total number of grid cells.

[0077] S25: Calculate the masking coefficient of each grid cell. The masking coefficient represents the degree to which the grid cell is occluded by the surrounding terrain. Based on the slope of the grid cell and the elevation data of the adjacent area, the masking coefficient value of each grid cell is calculated using a masking algorithm. A higher masking coefficient indicates a stronger masking effect. The formula for calculating the masking coefficient is as follows: Masking Coefficient ,in, For the first The slope angle of the adjacent area The number of neighboring areas;

[0078] S26: Integrate the slope, visibility, and masking coefficient calculated in steps S23 to S25 into the terrain grid model, and assign a corresponding terrain influence factor matrix to each grid cell to form the final terrain grid model. Through the precise calculation of slope, visibility, and masking coefficient, the terrain grid model can accurately reflect the terrain features in the training space, thereby providing high-precision data support for subsequent fire coverage, path accessibility judgment, and electromagnetic interference analysis, ensuring the accuracy and reliability of the adjudication process.

[0079] S21 specifically includes:

[0080] S211: Organize the discrete point data in the terrain feature dataset generated in S1 according to spatial coordinates, assign an elevation value to each data point, and preprocess the data points to remove outliers and noise to ensure the quality of the input data.

[0081] S212: Determine the weighting function for spatial interpolation. The covariance function (also known as the variogram) in Kriging interpolation is used to quantify the spatial correlation between different data points. The formula for calculating the covariance function is: ,in, Indicates distance The value of the variogram at that location, Representing data points and The variance of the elevation differences between points is used to assess the degree of spatial autocorrelation between different points.

[0082] S213: By calculating the variogram, determine the spatial weight of each point relative to its surrounding points. Specifically, for each pair of data points... and Calculate its spatial distance and obtain its weight through the covariance function. The weight calculation formula is as follows: ,in, For data points and Spatial weights between them Distance The value of the variogram at the specified point is used to determine the weight relationship between each pair of data points, thus affecting the interpolation result;

[0083] S214: The weighted average method is used to interpolate the elevation values ​​to obtain spatially interpolated continuous elevation data. The formula is as follows: ,in, Points to be interpolated The elevation value at that location, For the points to be interpolated and the known points Spatial weights, Given points The elevation value at that location, Given the number of data points, the formula calculates the elevation of the points to be interpolated based on spatial weights. Finally, interpolation is performed on all the points to be interpolated, and the discrete terrain data points are transformed into continuous spatial data using the Kriging interpolation method. Through accurate interpolation calculations, the above steps can generate continuous elevation data from discrete terrain data, providing accurate spatial data support for the subsequent construction of terrain grid models.

[0084] S3 specifically includes:

[0085] S31: Based on the slope, visibility, and concealment coefficient of each grid cell in the terrain grid model, a fire coverage range correction algorithm is established. This algorithm corrects the traditional fire coverage calculation model by considering the influence of terrain on the fire coverage range. Specifically, the correction formula is expressed as follows: ,in, The revised fire coverage area. The original data represents the uncorrected fire coverage area. This is the correction factor for slope on fire coverage. Slope is the slope value of this cell in the terrain grid model. This formula is used to calculate the effective range of fire strikes in complex terrain.

[0086] S32: Establish a path passability judgment function for judging whether the passable path of the training equipment in the terrain is feasible, the path passability function is evaluated based on the slope, visibility rate and masking coefficient of each grid element in the terrain grid model, the judgment function expression is: wherein, is the path passability score, is the weight of each grid element, is the number of grid elements, is the slope, visibility rate and masking coefficient of the element respectively, is the path passability function, which represents the influence of each grid element in the passable path, the function scores the path passability according to different terrain conditions to judge whether the path is feasible;

[0087] S33: Combined with the influence of electromagnetic environment, a calculation method of electromagnetic environment interference compensation value is established, the propagation attenuation value of electromagnetic wave in different terrain conditions is calculated according to the visibility rate and masking coefficient in the terrain grid model, the electromagnetic interference compensation formula is: wherein, is the compensation value of electromagnetic environment, is the electromagnetic interference value without considering the terrain, is the electromagnetic interference correction coefficient, the visibility rate is the visibility rate value of the element in the terrain grid model, the formula is used to compensate the electromagnetic interference according to the terrain change, so as to ensure the correction of electromagnetic environment interference more accurate;

[0088] S34: The fire coverage correction algorithm, the path passability judgment function and the electromagnetic environment interference compensation value are combined to establish a complete decision rule engine, through comprehensive consideration of terrain, fire coverage, path passability and electromagnetic environment factors, the comprehensive decision result is generated, so as to support the command decision of the training equipment; the comprehensive decision formula is: wherein, is the comprehensive decision result, , and are the weighting coefficients of each factor, which are dynamically adjusted according to the task demand, the engine dynamically adjusts the decision result according to different training conditions, so as to provide accurate instructions for the equipment; the above steps are based on the terrain grid model, combined with the factors of fire coverage, path passability and electromagnetic environment interference, to establish a comprehensive decision rule engine, through these calculations, the tactical environment in the training space can be accurately evaluated, and accurate basis is provided for the command decision.

[0089] S4 specifically comprises:

[0090] S41: Receives BeiDou positioning data and millimeter-wave radar scanning data from training equipment. The BeiDou positioning data includes the equipment's real-time coordinates and attitude information, while the millimeter-wave radar scanning data includes the location of obstacles around the equipment and radar echo information of the surrounding environment.

[0091] S42: Preprocess the received BeiDou positioning data to remove noise and outliers to ensure the accuracy of the positioning data. Preprocessing methods include data filtering and interpolation. Use appropriate filters to remove possible high-frequency noise and generate smooth positioning data.

[0092] S43: Process the received millimeter-wave radar scan data, convert the radar echo signal into range and angle information, and then convert it into spatial data in the standard Cartesian coordinate system using polar coordinates. Specifically, each point of the radar scan is converted into Cartesian coordinates using the following formula: ; ;in, The range value is the radar echo. For scanning angle, and These are the transformed Cartesian coordinates;

[0093] S44: Input the processed BeiDou positioning data and millimeter-wave radar scanning data into the Kalman filtering algorithm, and use the Kalman filtering algorithm to fuse the two. The Kalman filtering algorithm optimizes the estimation result by weighted fusion of prior information and measurement data through prediction and update steps.

[0094] The specific steps are as follows:

[0095] The prediction process involves using a system model to predict the current position and state. The current position is predicted based on the position and velocity from the previous moment. The prediction formula is as follows: ,in, For a moment The predicted state, Here is the state transition matrix. This is the estimated state from the previous moment. To control the input matrix, For control input;

[0096] The update process involves using BeiDou positioning data and millimeter-wave radar scanning data as observation data. The prediction results are then fused with the observation results using a weighted average. The update formula is as follows: ,in, The updated estimated state, For the observed values, For the observation matrix, This is the Kalman gain, used to balance the weights of predicted and observed values.

[0097] S45: According to the updated estimated position and state information of Kalman filtering algorithm, the matching of Beidou positioning data and millimeter wave radar scanning data is realized, and it is matched with the terrain grid model. Through the accurate position data of fusion, the position of equipment can be located in the terrain grid model, and the accurate alignment with the terrain data is ensured, which provides high-precision geographic position data support for the subsequent decision-making process.

[0098] S45 specifically includes:

[0099] S451: Obtain the updated estimated position and state information of Kalman filtering algorithm, including the position of equipment and its motion state;

[0100] S452: Match the updated equipment position and state information with the terrain grid model. Specifically, the accurate coordinates obtained by Kalman filtering are converted into the index of the terrain grid model, and the estimated position of equipment is mapped into the grid cell by using the space mapping algorithm, so as to ensure the alignment of position with the terrain grid model;

[0101] S453: Associate the millimeter wave radar scanning data with the updated equipment position. Through the obstacle position and echo information in the radar data, the relationship between equipment and surrounding environment is judged. In this process, the spatial data of millimeter wave radar scanning is aligned with the estimated equipment position through coordinate conversion, so as to ensure the spatio-temporal consistency of data;

[0102] S454: Match the matched Beidou positioning data, millimeter wave radar scanning data and terrain grid model. By mapping the equipment position and radar data into the corresponding terrain grid cell, the matching of equipment and terrain grid model is completed. This matching is realized by minimizing the positioning error, and the least square method is used for optimization;

[0103] S455: After completing the matching of equipment and terrain grid model, the final matching result is output, which provides accurate equipment positioning information for the subsequent decision-making process. This information is used to support the accurate calculation of firepower attack, path planning and other decision-making processes.

[0104] S5 specifically includes:

[0105] S51: According to the matched spatial coordinates obtained in S4, the specific position of equipment in the terrain grid model is determined, and the grid cell where the position is located is determined. Specifically, according to the matched coordinates, the equipment position is mapped into the corresponding terrain grid model by using the space mapping algorithm, so as to accurately locate the grid cell where the equipment is located in the terrain grid model;

[0106] S52: Extract the terrain influence factor matrix corresponding to the grid unit of the equipment position, including the slope, visibility rate and masking coefficient information of the corresponding unit grid unit; by reading the corresponding data structure in the terrain grid model, the terrain parameters of the grid unit are extracted to provide the necessary terrain data for the subsequent decision calculation;

[0107] S53: Call the decision rule engine, respectively calculate the movement feasibility, fire attack effective radius and electromagnetic interference attenuation value of the equipment based on the extracted terrain influence factor matrix, and provide support for the subsequent command decision.

[0108] As shown in Figure 2 The rapid and accurate decision system based on the terrain environment of the training space is used to realize the rapid and accurate decision method based on the terrain environment of the training space, which includes the following modules:

[0109] The information collection module is used to collect the elevation point cloud, ground cover contour line and underground shelter structure data of the training space in real time through laser radar and infrared remote sensing technology, and generate the terrain feature data set from the collected data;

[0110] The data processing module is used to input the terrain feature data set into the spatial interpolation algorithm, and generate the terrain grid model containing slope, visibility rate and masking coefficient through the Kriging interpolation method;

[0111] The decision rule engine module is based on the terrain influence factor matrix in the terrain grid model, combined with the real-time positioning data of the training equipment, through the fire coverage correction algorithm, path trafficability discrimination function and electromagnetic interference compensation algorithm, to calculate the movement feasibility, fire attack effective radius and electromagnetic interference attenuation value of the equipment in turn;

[0112] The data fusion module is used to receive the Beidou positioning data and millimeter wave radar scanning data of the training equipment, and match the data through the Kalman filtering algorithm, calculate the equipment position and register with the terrain grid model to ensure the accuracy of the position data;

[0113] The decision output module outputs accurate decision information according to the calculation result of the decision rule engine combined with the real-time state of the equipment, supporting the command decision of the equipment.

[0114] An electronic device includes a processor and a storage device, the storage device stores a computer program, and the computer program is used to execute the steps of the rapid and accurate decision method based on the terrain environment of the training space when the processor runs.

Claims

1. A rapid and accurate adjudication method based on the terrain environment of the training space, characterized in that, Includes the following steps: S1: Use lidar and infrared remote sensing technology to collect elevation point cloud, surface cover outline and underground bunker structure data of the training space to generate terrain feature dataset; S2: Input the terrain feature dataset into the spatial interpolation algorithm to generate a terrain grid model containing slope, visibility and masking coefficient, where each grid cell stores its corresponding terrain influence factor matrix; S3: Based on the terrain grid model, establish a decision rule engine that includes a fire coverage range correction algorithm, a path traversability discrimination function, and an electromagnetic environment interference compensation value; S4: Receives BeiDou positioning data and millimeter-wave radar scanning data from training equipment, and performs data matching through the Kalman filtering algorithm to achieve registration with the terrain grid model; S5: Based on the matched spatial coordinates, extract the terrain influence factor matrix of the corresponding grid cell, and call the adjudication rule engine to calculate the equipment's movement feasibility, effective fire strike radius, and electromagnetic interference attenuation value.

2. The rapid and accurate adjudication method based on the terrain environment of the training space according to claim 1, characterized in that, S1 specifically includes: S11: Use lidar technology to scan the training space and acquire elevation point cloud data. Lidar measures the three-dimensional coordinates of points by emitting laser pulses and receiving reflected signals, generating a ground point cloud dataset. The point cloud data contains elevation information of the terrain surface. S12: Using infrared remote sensing technology, the infrared sensor acquires the outline data of the ground cover in the training space. The infrared sensor uses the principle of thermal radiation to detect the temperature difference of the object surface, identify the different characteristics of the ground cover, and generate the distribution and outline data of the cover. S13: Combining underground detection technology, ground-penetrating radar is used to obtain data on the structure of underground bunkers. The ground-penetrating radar generates images or data of underground structures by emitting electromagnetic waves and receiving reflected signals, providing the spatial location and structural characteristics of underground bunkers. S14: Merge the elevation point cloud, surface cover outline and underground bunker structure data collected in steps S11, S12 and S13, and convert them into a standard format to generate a terrain feature dataset.

3. The rapid and accurate adjudication method based on the terrain environment of the training space according to claim 1, characterized in that, S2 specifically includes: S21: Input the terrain feature dataset generated in S1 into the spatial interpolation algorithm, and use the Kriging interpolation method to perform spatial interpolation processing on the terrain data, transforming the discrete terrain feature data into continuous spatial data; S22: During the interpolation process, the elevation value and terrain feature data of each grid cell are determined based on the spatial distribution and local variation patterns of the terrain feature data; S23: Based on the interpolated elevation data, calculate the slope of each grid cell in the terrain grid model. The slope value is determined by calculating the elevation difference between adjacent grid cells. The triangulation method is used to calculate the slope of each grid cell. S24: Calculate the visibility of each grid cell. Using a ray tracing algorithm, determine whether the line of sight of each grid cell is blocked by other terrain cells, and then calculate the visibility value of that cell. S25: Calculate the masking coefficient of each grid cell. Based on the slope of the grid cell and the elevation data of the adjacent area, use the masking algorithm to calculate the masking coefficient value of each grid cell. S26: Integrate the slope, visibility, and camouflage coefficient calculated in steps S23 to S25 into the terrain grid model, and assign a corresponding terrain influence factor matrix to each grid cell to form the final terrain grid model.

4. The rapid and accurate adjudication method based on the terrain environment of the training space according to claim 3, characterized in that, S21 specifically includes: S211: Organize the discrete point data in the terrain feature dataset generated in S1 according to spatial coordinates, assign an elevation value to each data point, and preprocess the data points to remove outliers and noise; S212: Determine the weighting function for spatial interpolation. The covariance function in Kriging interpolation is used to quantify the spatial correlation between different data points. The formula for calculating the covariance function is: ,in, Indicates distance The value of the variogram at that location, Representing data points and The variance of the elevation difference between them; S213: By calculating the variogram, determine the spatial weight of each point relative to its surrounding points. Specifically, for each pair of data points... and Calculate its spatial distance and obtain its weight through the covariance function. The weight calculation formula is as follows: ,in, For data points and Spatial weights between them Distance The value of the variogram at that location; S214: The weighted average method is used to interpolate the elevation values ​​to obtain spatially interpolated continuous elevation data. The formula is as follows: ,in, Points to be interpolated The elevation value at that location, For the points to be interpolated and the known points Spatial weights, Given points The elevation value at that location, This represents the number of data points.

5. The rapid and accurate adjudication method based on the terrain environment of the training space according to claim 1, characterized in that, S3 specifically includes: S31: Based on the slope, visibility, and concealment coefficient of each grid cell in the terrain grid model, a fire coverage range correction algorithm is established. The correction formula is expressed as: R 修正 =R 原始 ×(1-α×slope), where R 修正 For the corrected fire coverage, R 原始 The uncorrected fire coverage area is represented by α, which is the correction factor for fire coverage due to slope. S32: Establish a path feasibility discrimination function to determine whether the travel path of training equipment in the terrain is feasible. The expression of the discrimination function is: ,in, The path accessibility score is given. The weight for each grid cell, For the number of grid cells, These are the unit's slope, visibility, and shading coefficient, respectively. This is a path traversability function; S33: Considering the influence of the electromagnetic environment, establish a method for calculating the electromagnetic interference compensation value. The electromagnetic interference compensation formula is: E 补偿 =E 原始 ×(1-β⋅viewability), where E 补偿 E is the compensation value for the electromagnetic environment. 原始 The electromagnetic interference value is not considered based on terrain, β is the electromagnetic interference correction coefficient, and the visibility rate is the visibility rate value of this cell in the terrain mesh model. S34: Combining the fire coverage correction algorithm, the path traversability discrimination function, and the electromagnetic environment interference compensation value, a complete adjudication rule engine is established; the comprehensive adjudication formula is: C 裁决 =α1⋅R 修正 +α2⋅P 通行 +α3⋅E 补偿 Wherein, C_ruling is the comprehensive ruling result, and α1, α2 and α3 are the weighting coefficients of each factor.

6. The rapid and accurate adjudication method based on the terrain environment of the training space according to claim 1, characterized in that, S4 specifically includes: S41: Receives BeiDou positioning data and millimeter-wave radar scanning data from training equipment. The BeiDou positioning data includes the equipment's real-time coordinates and attitude information, while the millimeter-wave radar scanning data includes the location of obstacles around the equipment and radar echo information of the surrounding environment. S42: Preprocess the received BeiDou positioning data to remove noise and outliers; S43: Process the received millimeter-wave radar scan data, convert the radar echo signal into range and angle information, and convert it into spatial data in the standard Cartesian coordinate system through polar coordinates; S44: Input the processed BeiDou positioning data and millimeter-wave radar scanning data into the Kalman filtering algorithm, and use the Kalman filtering algorithm to fuse the two. The Kalman filtering algorithm optimizes the estimation result by weighted fusion of prior information and measurement data through prediction and update steps. S45: Based on the estimated position and state information updated by the Kalman filtering algorithm, the BeiDou positioning data and millimeter-wave radar scanning data are matched and registered with the terrain grid model.

7. The rapid and accurate adjudication method based on the terrain environment of the training space according to claim 6, characterized in that, Specifically, S45 includes: S451: Obtain the estimated position and state information updated by the Kalman filter algorithm, including the position of the equipment and its motion state; S452: Match the updated equipment position and status information with the terrain grid model. Specifically, convert the precise coordinates obtained by Kalman filtering into the index of the terrain grid model, and use a spatial mapping algorithm to map the estimated position of the equipment into the grid cells to ensure the alignment of the position with the terrain grid model. S453: Associate millimeter-wave radar scan data with the updated equipment location, and determine the relationship between the equipment and the surrounding environment by using obstacle locations and echo information in the radar data. S454: Register the matched BeiDou positioning data, millimeter-wave radar scanning data and terrain grid model. By mapping the equipment position and radar data to the corresponding terrain grid cells, the registration of the equipment and the terrain grid model is completed. S455: After completing the registration of the equipment with the terrain mesh model, output the final registration result to provide equipment positioning information for the subsequent adjudication process.

8. The rapid and accurate adjudication method based on the terrain environment of the training space according to claim 1, characterized in that, S5 specifically includes: S51: Based on the matched spatial coordinates obtained in S4, determine the specific location of the equipment in the terrain grid model and determine the grid cell where the location is located. S52: Extract the terrain influence factor matrix of the grid cell corresponding to the equipment location, including the slope, visibility and masking coefficient information of the corresponding grid cell; S53: Calls the adjudication rule engine and, based on the extracted terrain influence factor matrix, calculates the equipment's movement feasibility, effective fire strike radius, and electromagnetic interference attenuation value, providing support for subsequent command and adjudication.

9. A rapid and accurate adjudication system based on the terrain environment of the training space, characterized in that: Includes the following modules: Information acquisition module: used to collect elevation point cloud, surface cover outline and underground bunker structure data of the training space in real time through lidar and infrared remote sensing technology, and generate terrain feature dataset from the collected data; Data processing module: Used to input terrain feature datasets into spatial interpolation algorithms, and generate terrain grid models containing slope, visibility, and masking coefficients through the Kriging interpolation method; The adjudication rule engine module: Based on the terrain influence factor matrix in the terrain grid model, combined with the real-time positioning data of the training equipment, it calculates the equipment's movement feasibility, effective fire strike radius, and electromagnetic interference attenuation value in sequence through the fire coverage range correction algorithm, path traversability discrimination function, and electromagnetic interference compensation algorithm. Data fusion module: used to receive BeiDou positioning data and millimeter-wave radar scanning data from training equipment, and to perform data matching through Kalman filtering algorithm, calculate equipment position and register it with terrain grid model; Decision Output Module: Based on the calculation results of the decision rule engine and combined with the real-time status of the equipment, the module outputs decision information to support the command and decision-making of the equipment.

10. An electronic device, characterized in that, It includes a processor and a storage device, wherein the storage device stores a computer program, which, when run by the processor, executes the rapid and accurate decision-making method based on the terrain environment of the training space as described in any one of claims 1-8.

Images