Unmanned position firepower distribution optimization method and system based on three-dimensional strike efficiency

Abstract

The invention provides an unmanned position firepower distribution optimization method and system based on three-dimensional strike efficiency, and relates to the technical field of unmanned position firepower distribution optimization, and the method comprises the steps: obtaining a preset resolution digital elevation value of an area where an unmanned position is located and discrete data of vegetation coverage, and calculating a slope value based on the digital elevation value; based on the discrete data, a Gaussian process machine learning model is adopted to carry out space fitting on the digital elevation value, the vegetation coverage index and the slope value to obtain high-precision data, and a high-precision regular grid geographic parameter model is generated according to the high-precision data; constructing an initial joint firepower distribution model, and calculating a terrain influence factor set based on a regular grid geographic parameter model; and according to the terrain influence factor set, correcting a striking efficiency parameter of the combined firepower distribution model so as to establish a three-dimensional terrain adaptive combined firepower striking model considering the terrain influence, and performing iteration so as to optimize an unmanned position firepower distribution scheme. The actual combat feasibility and the environmental adaptability are improved.

Description

Technical Field

[0001] This invention relates to the field of unmanned position fire allocation optimization technology, and in particular to an unmanned position fire allocation optimization method and system based on three-dimensional strike effectiveness. Background Technology

[0002] In recent years, with the widespread application of unmanned combat systems in complex battlefield environments such as field defense, mountain warfare, and island and reef defense, fire coordination and dynamic allocation systems based on unmanned platforms have become a core component of modern battlefield fire support systems. These battlefield environments are characterized by significant terrain complexity and dynamic uncertainty. Factors such as three-dimensional terrain undulations, vegetation cover, building obstruction, and elevation changes directly impact the firing range, ballistic trajectory, and target visibility of fire units. Simultaneously, the diverse types of threats on the battlefield (e.g., infantry, vehicles, drones, ships), their changing motion states, and the dynamic evolution of threat levels place extremely high demands on the real-time performance, accuracy, and adaptability of fire allocation systems.

[0003] In existing technologies, the fire allocation process for unmanned defense positions generally suffers from the following two core technical bottlenecks: First, the lack of three-dimensional terrain environment modeling leads to fire allocation schemes being detached from actual battlefield conditions. Traditional fire allocation systems often rely on two-dimensional planar maps or simplified terrain models for target location and fire unit deployment, failing to fully consider the constraints of key geographical factors such as three-dimensional terrain undulations, concealment effects, and line-of-sight obstruction on firing effectiveness. In actual complex environments such as mountains, jungles, and islands, there are often terrain barriers, elevation differences, and curvature effects between fire units and targets, resulting in the allocated fire units failing to effectively hit targets in actual execution. Second, there is a lack of refined modeling and quantification capabilities for the "three-dimensional strike effectiveness" determined by three-dimensional terrain. Existing methods often use simplified global correction coefficients or empirical discounts, failing to construct a dynamic mathematical model of the relationship between terrain features (such as elevation differences, visibility, slope, and vegetation concealment) and weapon firing accuracy, effective range, and probability of destruction based on three-dimensional spatial geometric relationships. This "planar" or "static" effectiveness assessment method cannot truly reflect the actual coverage and firing feasibility of fire units in three-dimensional space. As a result, there is a significant deviation between the strike effectiveness predicted by the allocation plan and the actual effectiveness under complex terrain, thereby affecting the scientific nature and reliability of the overall combat decision. Summary of the Invention

[0004] Based on this, the purpose of this invention is to provide a method and system for optimizing the fire allocation of unmanned positions based on three-dimensional strike effectiveness, in order to solve the technical problems in the prior art that ignore the influence of complex terrain factors, resulting in schemes being out of touch with actual combat and inaccurate effectiveness assessments.

[0005] This invention provides, in one aspect, a method for optimizing the firepower allocation of unmanned positions based on three-dimensional strike effectiveness, comprising: By integrating multi-source remote sensing data into a geospatial information platform, the preset resolution digital elevation values ​​and discrete vegetation cover data of the area where the unmanned site is located are obtained, and the slope value is calculated based on the digital elevation values. Based on the discrete data, a Gaussian process machine learning model is used to spatially fit the digital elevation value, vegetation cover index and slope value to obtain high-precision data, and a high-precision regular grid geographic parameter model is generated based on the high-precision data. An initial joint fire allocation model is constructed, and a set of terrain influence factors is calculated based on the high-precision regular grid geographic parameter model. The strike effectiveness parameters of the joint fire allocation model are corrected according to the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model that considers terrain influence, and the model is iterated to output an unmanned position fire allocation scheme that considers terrain influence.

[0006] The aforementioned method for optimizing fire allocation in unmanned positions based on three-dimensional strike effectiveness overcomes the technical limitations of traditional methods that rely on coarse two-dimensional terrain data by constructing a refined three-dimensional terrain-adaptive joint fire strike model. This lays a reliable data foundation for accurately quantifying the impact of terrain. Secondly, it systematically calculates a set of terrain influence factors, including visibility, slope effect, and vegetation obstruction, transforming complex terrain spatial features into quantitative parameters that can be directly coupled to fire allocation. This enables dynamic and refined modeling of weapon strike effectiveness in complex terrain environments. By integrating the set of terrain influence factors into the initial joint fire allocation model, the core effectiveness parameters such as weapon hit probability and effective range are dynamically corrected, constructing a three-dimensional terrain-adaptive joint fire strike model. This allows the fire allocation scheme to realistically reflect the constraints of terrain undulations and obstruction on strike feasibility and effectiveness, significantly improving the scheme's practical feasibility and environmental adaptability. This solves the technical problem of traditional fire allocation methods ignoring the influence of complex terrain factors, leading to schemes deviating from actual combat and distorted effectiveness predictions.

[0007] In addition, the unmanned position fire allocation optimization method based on three-dimensional strike effectiveness according to the present invention may also have the following additional technical features: Further, the steps of correcting the strike effectiveness parameters of the joint fire allocation model based on the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model considering terrain influence and iterating thereafter to output an unmanned position fire allocation scheme considering terrain influence include: The strike effectiveness parameters of the joint fire allocation model are modified according to the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model that takes into account the influence of terrain. Based on the aforementioned three-dimensional terrain-adaptive joint fire strike model, the NSGA-II multi-objective optimization algorithm is used to iteratively optimize the fire allocation decision variables to output an unmanned position fire allocation scheme that takes into account the influence of terrain. The fire allocation decision variables are the corresponding allocation relationship between weapon platforms and strike targets.

[0008] Furthermore, the objective function expression for the three-dimensional terrain-adaptive joint fire strike model is: ; In the formula, V’ ( x () represents the attack value gain after terrain correction; C ( x This indicates the overall cost of the strike; d 3D ( i , j () indicates a three-dimensional terrain range constraint; R’ i This indicates the dynamically adjusted range of the weapon platform. x ij Indicates the decision variables for firepower allocation; LOS ij Indicates visibility factor; The number of weapon platforms allocated to the target; For the purpose of striking the target; m i For the first i The current number of weapon-like platforms; m Indicates the weapon platform category; i Indicates the first i Weapon-like platform; j Indicates the first j One objective; st represents constraints; in: ; ; In the formula, v j Indicates damage to the first The expected value of each goal; This indicates the corrected hit probability; c i Indicates the first The cost of a single strike by a weapon-type platform, where i = 1, 2, ..., m , m Indicates the weapon platform category.

[0009] Furthermore, the set of topographic influence factors includes a vegetation shading coefficient, the calculation formula for which is: ; In the formula, V j Indicates the vegetation shading coefficient; G Indicates the blade projection index; LAI j Indicates leaf area index; Indicates the angle between the line of sight and the vertical direction; i Indicates the number of weapon platform categories; j Indicates the number of targets; where, LAI j The calculation formula is: ; In the formula, k represents the extinction coefficient; q Indicates the leaf tilt angle distribution parameters; NDVI j Represents the normalized vegetation index at the j-th target location; NDVI soil Indicates the NDVI baseline value for bare soil; NDVI veg This represents the baseline NDVI value for pure vegetation.

[0010] Furthermore, the set of terrain influence factors includes terrain correction factors and slope-range correction coefficients, wherein: the calculation formula for the terrain correction factor is: ; In the formula, TF ij Indicates the terrain correction factor; LOS ij Indicates visibility factor; α V represents the vegetation shading weighting coefficient; j Indicates the vegetation shading coefficient; The formula for calculating the slope range correction coefficient is as follows: ; In the formula, η ij This represents the slope range correction factor; γ Indicates the uphill range attenuation coefficient; β θ represents the downhill range gain coefficient; crit Indicates the critical slope angle; η max Indicates the maximum range gain limit; θ ij θ represents the slope angle between the i-th weapon platform and the j-th target; where θ ij The calculation formula is: ; In the formula, ( m xi ,m yi , E i ) represents the three-dimensional coordinates of the i-th type of weapon platform; (t) xj , t yj , E j ) represents the three-dimensional coordinates of the j-th target; Digital elevation values ​​of weapon deployment points extracted from a regular grid digital elevation model; This refers to the digital elevation values ​​of the target points extracted from the regular grid digital elevation model.

[0011] Furthermore, in the step of iteratively optimizing the fire allocation decision variables using the NSGA-II multi-objective optimization algorithm based on the aforementioned three-dimensional terrain-adaptive joint fire strike model to output an unmanned position fire allocation scheme that considers terrain influence, the iterative optimization method includes: Initialize the parameters of the NSGA-II multi-objective optimization algorithm and randomly generate an initial population representing the fire allocation scheme; use the modified objective function in the three-dimensional terrain adaptive joint fire strike model as the fitness function to evaluate the strike effectiveness and strike cost of each individual in the population. Perform fast nondominated sorting, crowding calculation, tournament selection, simulated binary crossover, and polynomial mutation operations to produce offspring populations; An elite retention strategy is used to generate a new generation of population. The process is iterated until convergence, and a set of Pareto optimal solutions is output. Based on the Pareto solution set, a joint fire allocation scheme that takes into account the influence of terrain is output.

[0012] Furthermore, the step of correcting the strike effectiveness parameters of the joint fire allocation model based on the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model considering terrain influence includes: Calculate the three-dimensional Euclidean distance between the weapon platform and the target; The hit probability is adjusted based on the terrain correction factor to obtain the corrected hit probability; The range is corrected according to the slope range correction factor to obtain the corrected range. Based on the three-dimensional Euclidean distance, the corrected hit probability, and the obtained corrected range, a three-dimensional terrain-adaptive joint fire strike model considering the influence of terrain is established.

[0013] Another aspect of the present invention provides an unmanned position firepower allocation optimization system based on three-dimensional strike effectiveness, the system comprising: The data acquisition module is used to acquire the preset resolution digital elevation value and discrete vegetation cover data of the area where the unmanned site is located through a geospatial information platform that integrates multi-source remote sensing data, and to calculate the slope value based on the digital elevation value. The spatial fitting module is used to perform spatial fitting on the digital elevation value, vegetation cover index and slope value based on the discrete data using a Gaussian process machine learning model to obtain high-precision data, and generate a high-precision regular grid geographic parameter model based on the high-precision data. The fire allocation module is used to construct an initial joint fire allocation model and calculate a set of terrain influence factors based on the high-precision regular grid geographic parameter model; it corrects the strike effectiveness parameters of the joint fire allocation model according to the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model that considers terrain influence and iterates it to output a fire allocation scheme for unmanned positions that considers terrain influence.

[0014] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for optimizing the fire allocation of unmanned positions based on three-dimensional strike effectiveness.

[0015] In another aspect, the present invention provides a data processing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-described method for optimizing the fire allocation of unmanned positions based on three-dimensional strike effectiveness. Attached Figure Description

[0016] Figure 1 This is a flowchart of the unmanned position firepower allocation optimization method based on three-dimensional strike effectiveness in the first embodiment of the present invention; Figure 2 This is a high-resolution three-dimensional digital elevation model of the experimental island and reef area in the first embodiment of the present invention; Figure 3 This is a top view of the vegetation cover in the experimental island / reef area in the first embodiment of the present invention; Figure 4 This is a map showing the locations of weapon platforms and targets to be attacked in the experimental island and reef area in the first embodiment of the present invention; Figure 5 This is the optimized Pareto front solution set in the first embodiment of the present invention; Figure 6 This is a diagram of a weapon target allocation scheme for selecting a certain front solution on the Pareto front solution set in the first embodiment of the present invention; The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0017] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0019] To facilitate understanding of the present invention, several embodiments are given below. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of the present invention will be more thorough and complete.

[0020] Example 1 Please see Figure 1 The figure shows the method for optimizing the firepower allocation of unmanned positions based on three-dimensional strike effectiveness in the first embodiment of the present invention, the method including steps S101 to S105: S101. By integrating multi-source remote sensing data into a geospatial information platform, the preset resolution digital elevation value and discrete vegetation cover data of the area where the unmanned site is located are obtained, and the slope value is calculated based on the digital elevation value.

[0021] Specifically, utilizing a geospatial information platform with cloud computing and massive remote sensing data management capabilities, an automated processing flow is written and executed to systematically extract geographic information from the target island and reef area to obtain discrete data. In this embodiment, the discrete data includes digital elevation values ​​and normalized difference vegetation index (NDVI). The automated processing flow includes: First, based on the deployment range of the unmanned site, a precise research area is defined in the platform by inputting its geographical coordinate boundaries. Within the research area, a regular sampling point grid is created with a spatial resolution of 30 meters. Each node of the grid corresponds to a geographical location with latitude and longitude coordinates, resulting in discrete and uniformly distributed geospatial sampling points.

[0022] Secondly, the platform calls the Digital Elevation Model (DEM) dataset covering the area and reads the digital elevation value corresponding to each sampling point. Taking each sampling point as the center point, the slope value at the center point is calculated by numerical differentiation using the digital elevation values ​​of the eight neighboring sampling points around it.

[0023] Furthermore, the multispectral remote sensing image dataset covering the area in the platform is called up. Preferably, the near-infrared and red light bands are used to calculate the Normalized Difference Vegetation Index (NDVI) corresponding to each sampling point location, so as to quantitatively characterize the vegetation coverage at each sampling point location.

[0024] S102. Based on discrete data, a Gaussian process machine learning model is used to spatially fit digital elevation values, vegetation cover index, and slope values ​​to obtain high-precision data. A high-precision regular grid geographic parameter model is then generated based on the high-precision data.

[0025] In this embodiment, a Gaussian process machine learning model is constructed for each geographic parameter to establish a nonlinear mapping framework between geospatial coordinates and geographic parameters, providing a basic probabilistic model for subsequent statistical extrapolation. Specifically, the geographic parameters include digital elevation values, normalized difference vegetation index (NDV), and slope values. To achieve this nonlinear mapping, the spatial distribution surface of the parameters to be reconstructed is modeled as a Gaussian stochastic process, expressed as:

[0026] in, The function values ​​of the surface containing the spatial distribution of parameters to be reconstructed; For spatial coordinate vectors, This is a regression function vector used to describe the deterministic trend between geographic parameters and spatial coordinates; This is a vector of regression coefficients; A random process with zero mean has its covariance defined by a kernel function.

[0027] Within the aforementioned Gaussian process machine learning model architecture, a kernel function is defined and a covariance matrix is ​​constructed. The kernel function, as the core evolutionary mechanism of this Gaussian process machine learning model, is used to characterize the correlation between different geographical locations in space. This correlation is then used to construct the covariance matrix, thereby quantifying the smoothness and evolution of terrain features as spatial distance changes. Specifically, a radial basis function is used as the kernel function to measure the similarity between spatial points and construct the covariance structure. The radial basis function is:

[0028] in,l The correlation coefficient controls the rate at which spatial correlation decays. For two data points and The square of the Euclidean distance between them, along with the relevant length, is used to measure similarity.

[0029] A Gaussian process machine learning model is trained based on known geographic parameter point data, enabling the model to adaptively fit the specific terrain undulation trend of the target area. The discrete data obtained in step S101 is then substituted into the aforementioned covariance matrix, and the kernel function hyperparameters are optimized by maximizing the marginal likelihood function, thus estimating the regression coefficients. This allows the abstract kernel function features to achieve the best fit with the actual terrain undulations of the target area. Specifically, the regression coefficients... The calculation formula is:

[0030] in, This is the basis function matrix corresponding to the known geographic parameter data points; This is the spatial correlation matrix between known geographic parameter data points; This is a vector of observation values ​​at known geographic parameter data points. Based on the trained optimal model, spatial prediction is performed on unsampled areas to generate a regular grid digital elevation model, achieving high-precision reconstruction from discrete data to a continuous regular terrain surface. For each point to be predicted in the regular grid digital elevation model... The posterior predicted distribution is calculated by iterating through all locations in the entire regular grid and calculating the predicted mean value for each location as an estimate of the geographic parameters for that point. This is then integrated with the geographic parameter data of all grid points to finally generate a high-precision regular grid geographic parameter model. The formula for the posterior predicted mean is:

[0031] in, This is the covariance vector between the predicted point and all known points.

[0032] S103. Construct an initial joint fire allocation model and calculate the set of terrain influence factors based on a high-precision regular grid geographic parameter model.

[0033] As a concrete example, firstly, a multi-objective constrained optimization model is constructed with the objective function of maximizing the value gain of the strike and minimizing the cost of the strike, and with the destructive capability of the weapon platform, the number of weapon platforms, and the strike effect of the weapon platform on the target as constraints, which serves as the benchmark for computational optimization; then, based on the regular grid geographic parameter model, the multi-dimensional influence factors of terrain on fire strikes are calculated.

[0034] Furthermore, the initial joint fire allocation model, based on the classic weapon-target allocation problem framework, defines the core elements and mathematical relationships of the model, specifically: Setting up unmanned positions m Weapon platform, No. The number of weapon platforms is ,in, i =1,2,…, m They are distributed in different locations on the battlefield, with coordinates as ( , ), with destructive power at maximum kill range This indicates that the probability of hitting is... The average cost of carrying out a single strike is Furthermore, there are Several potential targets were included in the target list, also distributed in different locations on the battlefield, with coordinates ( ). , ),in j Indicates the first j One goal, j =1,2,…,n, damage number The expected value of each target is .

[0035] 1) Decision variables Define firepower allocation decision variables x ij , ,in Indicates the assignment of the first i Weapon-type platform strike One goal, Indicates the first Weapon platforms do not strike the first Given several targets, the firepower distribution matrix between the weapon platform and the targets is as follows: ; 2) Objective function The objective function is to maximize the value gain from the strike and minimize the cost of the strike; the formula for calculating the value gain from the strike is as follows: ; In the formula, V ( x To combat value gains; v j To destroy the first The expected value of each goal; P j Indicates the first j The overall probability of damage when striking a target; among which... P j The calculation formula is: ; In the formula, P j Indicates the first j The overall damage probability when striking a target; p i For the first i The original hit probability of weapon-like platforms, among which, i Indicates the first i Weapon-like platform, i=1,2,…, m , m Indicates the weapon platform category; j Indicates the first One goal, j =1,2,…, n , n Indicates the number of targets; x ij This represents the decision variable for firepower allocation.

[0036] The formula for calculating the overall cost of the strike is as follows:

[0037] in, C ( x This indicates the overall cost of the strike; c i For the first i The cost of a single strike by a weapon-like platform, of which, i Indicates the first i Weapon-like platform i =1,2,…, m , m Indicates the weapon platform category; j Indicates the first One goal, j =1,2,…, n , n Indicates the number of targets; x ij This represents the decision variable for firepower allocation.

[0038] 3) Constraints Using the destructive capability of the weapon platform as a constraint, the distance between the weapon platform and the target cannot exceed its maximum kill range. The range constraint expression is as follows:

[0039] in, R i Indicates the maximum kill range; i Indicates the first i Weapon-like platform i =1,2,…,m , m Indicates the weapon platform category; j Indicates the first One goal, j =1,2,…, n , n Indicates the number of targets; d ( i , j ) represents the two-dimensional distance between the weapon platform and the target, where, .

[0040] Using the number of weapon platforms as a constraint, the number of weapon platforms allocated to a target cannot exceed its existing number. The calculation formula is as follows:

[0041] in, M i Indicates the number of weapon platforms assigned to the target; Indicates the current number of weapon platforms; i Indicates the first i Weapon-like platform i =1,2,…, m , m Indicates the weapon platform category; j Indicates the first j One goal, j =1,2,…, n , n Indicates the number of targets; x ij This represents the decision variable for firepower allocation.

[0042] Using the target strike effect as a constraint, each target is assigned to at least one weapon platform for strike. The formula for calculating the target strike effect is as follows:

[0043] Indicates the effect of a target strike; i Indicates the first i Weapon-like platform i =1,2,…, m , m Indicates the weapon platform category; j Indicates the first One goal, j =1,2,…, n , n Indicates the number of targets; x ij This represents the decision variable for firepower allocation.

[0044] 4) Optimize the model Based on the above objective function and constraint functions, an initial joint fire allocation model is established. The expression of the initial joint fire allocation model is as follows:

[0045] Combining the initial joint fire allocation model and based on a regular grid geographic parameter model, multi-dimensional influence factors of terrain on fire strikes are calculated. A set of terrain influence factors is then constructed based on these multi-dimensional influence factors. Specifically: The set of topographic influence factors includes visibility factors, vegetation shading coefficients, topographic correction factors, and slope-range correction factors, among which: (1) The visibility factor is a binary variable. Based on the DEM, a ray tracing algorithm is used to analyze the 3D terrain data to determine whether the line of sight from the i-th weapon platform to the j-th target is obstructed by the intermediate terrain. The expression for the visibility factor is: ; In the formula, LOS ij This indicates the visibility factor. If the line of sight is unobstructed, it means that the line of sight is unobstructed; if the line of sight is blocked, it means that the line of sight is obstructed.

[0046] (2) The formula for calculating the vegetation shading coefficient is: ; In the formula, V j Indicates the vegetation shading coefficient; G Indicates the blade projection index; LAI j Indicates leaf area index; Indicates the angle between the line of sight and the vertical direction; i Indicates the number of weapon platform categories; j Indicates the number of targets; where, LAI j The calculation formula is: ; In the formula, k represents the extinction coefficient; q Indicates the leaf tilt angle distribution parameters; NDVI j Represents the normalized vegetation index at the j-th target location; NDVI soil Indicates the NDVI baseline value for bare soil; NDVI veg This represents the baseline NDVI value for pure vegetation.

[0047] (3) The formula for calculating the terrain correction factor is: ; In the formula, TFij Indicates the terrain correction factor; LOS ij Indicates visibility factor; α V represents the vegetation shading weighting coefficient; j Indicates the vegetation shading coefficient; (4) The formula for calculating the slope range correction coefficient is: ; In the formula, η ij This represents the slope range correction factor; γ Indicates the uphill range attenuation coefficient; β θ represents the downhill range gain coefficient; crit Indicates the critical slope angle; η max Indicates the maximum range gain limit; θ ij θ represents the slope angle between the i-th weapon platform and the j-th target; where θ ij The calculation formula is: ; In the formula, ( m xi , m yi , E i ) represents the three-dimensional coordinates of the i-th type of weapon platform; t xj , t yj , E j () represents the three-dimensional coordinates of the j-th target; Digital elevation values ​​of weapon deployment points extracted from a regular grid digital elevation model; This refers to the digital elevation values ​​of the target points extracted from the regular grid digital elevation model.

[0048] S104. Based on the set of terrain influence factors, modify the strike effectiveness parameters of the joint fire allocation model to establish a three-dimensional terrain-adaptive joint fire strike model that takes into account the influence of terrain.

[0049] In this embodiment, the three-dimensional Euclidean distance between the weapon platform and the target is calculated; the hit probability is corrected according to the terrain correction factor to obtain the corrected hit probability; the range is corrected according to the slope range correction coefficient to obtain the corrected range; and a three-dimensional terrain-adaptive joint fire strike model considering the terrain influence is established based on the three-dimensional Euclidean distance, the corrected hit probability, and the corrected range.

[0050] Traditional two-dimensional planar distance models produce significant errors in undulating terrain such as mountains. Using three-dimensional Euclidean distance can accurately represent the actual spatial distance between the weapon platform and the target. Specifically, for the... Weapon platforms and the first The formula for calculating the three-dimensional Euclidean distance between the weapon platform and the target is as follows:

[0051] in: d 3D ( i , j ) represents the three-dimensional Euclidean distance; , ) indicates the first i The three-dimensional coordinates of a weapon-like platform, , ) indicates the first j The three-dimensional coordinates of the target; Digital elevation values ​​of weapon deployment points extracted from a regular grid digital elevation model; This refers to the digital elevation values ​​of the target points extracted from the regular grid digital elevation model.

[0052] The hit probability is adjusted based on the terrain correction factor; the formula for calculating the hit probability adjustment is as follows:

[0053] in, This represents the original hit probability; This is the terrain correction factor, which can be obtained according to the calculation formula mentioned above; This represents the dynamically adjusted hit probability.

[0054] The range is corrected based on the slope range correction factor; the range correction formula is as follows:

[0055] in, This refers to the original firing range of the weapon platform. The range correction factor is calculated above. This refers to the dynamically corrected range of the weapon platform.

[0056] Based on the above-mentioned corrected parameters, a three-dimensional terrain-adaptive joint fire strike model considering terrain influences is established; specifically: 1) Objective function The objective function is to maximize the strike value benefit after terrain correction and minimize the overall strike cost; the formula for maximizing the strike value benefit after terrain correction is as follows:

[0057] in, The attack value gain after terrain correction; v j Indicates damage to the first j The expected value of each goal; This indicates the corrected hit probability.

[0058] The formula for minimizing the overall cost of an attack is as follows:

[0059] c i Indicates the first The cost of a single strike by a weapon-like platform i =1,2,…, m , m Indicates the weapon platform category; j =1,2,…, n , n Indicates the number of targets.

[0060] 2) Constraint Functions The model constraint function is established using constraints such as 3D terrain range, visibility constraint, weapon platform quantity constraint, and target strike effect constraint. The expression for the 3D terrain range constraint is as follows:

[0061] This expression means that the three-dimensional Euclidean distance is less than or equal to the dynamically corrected weapon platform range, where... d 3D ( i , j () represents the three-dimensional Euclidean distance between the weapon platform and the target; The range of the weapon platform is dynamically corrected; i =1,2,…, m , m Indicates the weapon platform category; j Indicates the first One goal, j =1,2,…, n , n Indicates the number of targets.

[0062] The formula for calculating the visibility factor constraint is as follows:

[0063] in, The visibility factor can be calculated using the ray tracing algorithm based on 3D terrain data mentioned above. i =1,2,…, m ,m Indicates the weapon platform category; j Indicates the first One goal, j =1,2,…, n , n Indicates the number of targets.

[0064] The formula for calculating the constraint on the number of weapon platforms is as follows:

[0065] in, Indicates the number of weapon platforms assigned to the target; For the first The current number of weapon-like platforms; i =1,2,…, m , m Indicates the weapon platform category.

[0066] The constraint calculation formula for the target strike effect is as follows:

[0067] in, Indicates the effect of a target strike; j Indicates the first One goal, j =1,2,…, n .

[0068] 3) Optimization Model Based on the above objective function and constraint function, a three-dimensional terrain-adaptive joint fire strike model considering the influence of terrain is established, as follows:

[0069] In the formula: V’ ( x () represents the attack value gain after terrain correction; C ( x This indicates the overall cost of the strike; d 3D ( i , j () indicates a three-dimensional terrain range constraint; R’ i This indicates the dynamically adjusted range of the weapon platform. x ij Indicates the decision variables for firepower allocation; LOS ij Indicates visibility factor; The number of weapon platforms allocated to the target; For the purpose of striking the target; m i For the first iThe current number of weapon-like platforms; m Indicates the current number of weapon platforms; i Indicates the first i Weapon-like platform; j Indicates the first j One objective; st represents constraints.

[0070] S105. Based on the three-dimensional terrain adaptive joint fire strike model, the NSGA-II multi-objective optimization algorithm is used to iteratively optimize the fire allocation decision variables to output an unmanned position fire allocation scheme that takes into account the influence of terrain.

[0071] Iterative optimization methods include: First, the parameters of the NSGA-II multi-objective optimization algorithm are initialized, and an initial population representing the fire allocation scheme is randomly generated. Second, the modified objective function in the 3D terrain-adaptive joint fire strike model is used as the fitness function to evaluate the strike effectiveness and overall strike cost of each individual in the initial population. Third, fast non-dominated sorting, crowding calculation, tournament selection, simulated binary crossover, and polynomial mutation operations are performed on the current population to generate offspring populations, explore new solution spaces, and maintain population diversity. Fourth, an elite retention strategy is adopted to merge, sort, and select parent and offspring populations to generate a new generation population. The above evaluation, sorting, genetic, and elite selection processes are iterated until convergence, and a set of Pareto optimal solutions is output. Finally, a joint fire allocation scheme considering the influence of terrain is output based on the Pareto solution set.

[0072] Specifically: First, the parameters of the NSGA-II multi-objective optimization algorithm are initialized by randomly generating an initial population representing the fire allocation scheme. In this embodiment, the core parameters of the NSGA-II algorithm are set, including the population size. Maximum number of iterations Crossover probability Probability of mutation Randomly generate the initial population. Each individual corresponds to a feasible fire allocation scheme, i.e., a fire allocation matrix that satisfies all constraints. The encoding representation.

[0073] Secondly, the modified objective function in the three-dimensional terrain-adaptive joint fire strike model is used as the fitness function to evaluate the strike effectiveness and overall strike cost of each individual in the initial population. In this embodiment, for each individual in the initial population, according to the three-dimensional terrain-adaptive joint fire strike model established in step S104, its corresponding two objective function values ​​are calculated, and these two function values ​​together constitute the fitness vector of that individual.

[0074] Secondly, fast non-dominated sorting, crowding calculation, tournament selection, simulated binary crossover, and polynomial mutation operations are performed on the current population to generate offspring, explore new solution spaces, and maintain population diversity. In this embodiment, fast non-dominated sorting is performed on the entire current population, dividing it into different non-dominated frontier levels based on the dominance relationships between individuals. Within the same non-dominated frontier, to further maintain solution diversity, the crowding of each individual is calculated to measure its degree of clustering with other solutions in the target space. Based on the fitness evaluation and sorting results, genetic operations are performed to generate offspring. Genetic operations include selection, crossover, and mutation. Selection employs a binary tournament selection mechanism, randomly selecting two individuals from the parent population each time, prioritizing those with higher non-dominant front ranks; if they are on the same front, the more crowded individual is selected, thus choosing the paternal line for reproduction. Crossover involves applying a probability-based selection method to the selected paternal individuals. A simulated binary crossover operation is performed, exchanging some of the encoded information to explore new regions of the solution space. The mutation operation applies a probability-based formula to the individuals produced after the crossover. P m Perform polynomial mutation operations to randomly perturb parts of the encoding in order to maintain population diversity and avoid premature convergence.

[0075] Furthermore, an elite preservation strategy is employed to merge, sort, and select the parent and offspring populations, generating a new generation population. This process of evaluation, sorting, genetics, and elite selection is iterated until convergence, outputting a set of Pareto optimal solutions. In this embodiment, the parent population... With offspring population Merge to form a size of joint population .right All individuals are re-ranked using the fast non-dominated ordering and crowding calculation. Individuals are then selected for the next generation of the population in descending order of their non-dominated frontier status. When selecting a certain frontier, adding all individuals from that frontier would cause the population size to exceed [a certain threshold]. Then, individuals within the frontier are selected from highest to lowest crowding level until the frontier is filled. This elite retention strategy ensures the inheritance of superior individuals while promoting a more even distribution of the Pareto frontier. Individual fitness assessment and ranking, genetic operations to generate offspring, elite selection, and population updates are repeated until the preset maximum number of iterations is reached. Or it may meet the convergence condition. When the algorithm terminates, the final population will be... All individuals belonging to the first non-dominated frontier are extracted to form the Pareto optimal solution set of the fire strike model.

[0076] Finally, a joint fire allocation scheme considering terrain influences is output based on the Pareto solution set. In this embodiment, each solution in the solution set represents a fire allocation scheme that achieves an optimal balance between strike value and overall cost, and fully adapts to three-dimensional terrain constraints.

[0077] The NSGA-II multi-objective optimization algorithm is used to efficiently solve the three-dimensional terrain-adaptive joint fire strike model. It can quickly generate a set of Pareto schemes that represent the optimal trade-off relationship under multiple objectives and complex constraints, such as maximizing strike effectiveness and minimizing resource consumption. This provides decision-makers with a decision space that is scientific, flexible and interpretable, and effectively supports rapid and intelligent decision-making under complex constraints.

[0078] To further illustrate this embodiment, an experiment was conducted using a certain island / reef area as the research object for verification: Figure 2 This is a high-resolution three-dimensional terrain digital elevation model of the region constructed using a Gaussian process model based on a multi-source remote sensing dataset. Figure 3 This is an aerial view of the vegetation cover of the region, constructed using a Gaussian process model based on a multi-source remote sensing dataset. Figure 4 This map shows the locations of 10 weapon platforms and 4 targets to be engaged in this terrain environment. Triangles represent weapon positions, and circles represent target positions. Figure 5 This is the optimized Pareto front solution set. Figure 6 This diagram illustrates the weapon target allocation scheme for selecting a specific front solution on the Pareto front solution set.

[0079] In summary, the unmanned position fire allocation optimization method based on three-dimensional strike effectiveness in the above embodiments of the present invention, by constructing a refined three-dimensional terrain-adaptive joint fire strike model, overcomes the technical limitations of traditional methods that rely on coarse two-dimensional terrain data, laying a reliable data foundation for accurately quantifying the impact of terrain. Secondly, it systematically calculates a set of terrain influence factors, including visibility, slope effect, and vegetation obstruction, transforming complex terrain spatial features into quantitative parameters that can be directly coupled to fire allocation, realizing dynamic and refined modeling of weapon strike effectiveness in complex terrain environments. By integrating the set of terrain influence factors into the initial joint fire allocation model, the core effectiveness parameters such as weapon hit probability and effective range are dynamically corrected, constructing a three-dimensional terrain-adaptive joint fire strike model. This enables the fire allocation scheme to truly reflect the constraints of terrain undulations, obstruction, and other factors on strike feasibility and effectiveness, significantly improving the scheme's practical feasibility and environmental adaptability, thereby solving the technical problem of traditional fire allocation methods ignoring the influence of complex terrain factors, leading to schemes deviating from actual combat and distorted effectiveness predictions.

[0080] Example 2 The second embodiment of the present invention provides an unmanned position firepower allocation optimization system based on three-dimensional strike effectiveness, the system comprising: The data acquisition module is used to acquire the preset resolution digital elevation value and discrete vegetation cover data of the area where the unmanned site is located through a geospatial information platform that integrates multi-source remote sensing data, and to calculate the slope value based on the digital elevation value. The spatial fitting module is used to perform spatial fitting on the digital elevation value, vegetation cover index and slope value based on the discrete data using a Gaussian process machine learning model to obtain high-precision data, and generate a high-precision regular grid geographic parameter model based on the high-precision data. The fire allocation module is used to construct an initial joint fire allocation model and calculate a set of terrain influence factors based on the high-precision regular grid geographic parameter model; it corrects the strike effectiveness parameters of the joint fire allocation model according to the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model that considers terrain influence and iterates it to output a fire allocation scheme for unmanned positions that considers terrain influence.

[0081] In summary, the unmanned position fire allocation optimization system based on three-dimensional strike effectiveness in the above embodiments of the present invention, by constructing a refined three-dimensional terrain-adaptive joint fire strike model, overcomes the technical limitations of traditional methods that rely on coarse two-dimensional terrain data, laying a reliable data foundation for accurately quantifying the impact of terrain. Secondly, it systematically calculates a set of terrain influence factors, including visibility, slope effect, and vegetation obstruction, transforming complex terrain spatial features into quantitative parameters that can be directly coupled to fire allocation, realizing dynamic and refined modeling of weapon strike effectiveness in complex terrain environments. By integrating the set of terrain influence factors into the initial joint fire allocation model, it dynamically corrects core effectiveness parameters such as weapon hit probability and effective range, constructing a three-dimensional terrain-adaptive joint fire strike model. This enables the fire allocation scheme to truly reflect the constraints of terrain undulations, obstruction, and other factors on strike feasibility and effectiveness, significantly improving the scheme's practical feasibility and environmental adaptability, thereby solving the technical problem of traditional fire allocation methods ignoring the influence of complex terrain factors, leading to schemes deviating from actual combat and distorted effectiveness predictions.

[0082] Furthermore, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the methods described above.

[0083] Furthermore, embodiments of the present invention also propose a data processing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the methods described above.

[0084] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0085] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0086] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0087] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0088] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for optimizing the firepower allocation of unmanned positions based on three-dimensional strike effectiveness, characterized in that, include: By integrating multi-source remote sensing data into a geospatial information platform, the preset resolution digital elevation values ​​and discrete vegetation cover data of the area where the unmanned site is located are obtained, and the slope value is calculated based on the digital elevation values. Based on the discrete data, a Gaussian process machine learning model is used to spatially fit the digital elevation value, vegetation cover index and slope value to obtain high-precision data, and a high-precision regular grid geographic parameter model is generated based on the high-precision data. An initial joint fire allocation model is constructed, and a set of terrain influence factors is calculated based on the high-precision regular grid geographic parameter model. The strike effectiveness parameters of the joint fire allocation model are corrected according to the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model that considers terrain influence, and the model is iterated to output an unmanned position fire allocation scheme that considers terrain influence.

2. The method for optimizing the allocation of unmanned field firepower based on three-dimensional strike effectiveness according to claim 1, characterized in that, The steps of correcting the strike effectiveness parameters of the joint fire allocation model based on the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model that considers terrain influence, and iterating thereafter to output an unmanned position fire allocation scheme that considers terrain influence, include: The strike effectiveness parameters of the joint fire allocation model are modified according to the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model that takes into account the influence of terrain. Based on the aforementioned three-dimensional terrain-adaptive joint fire strike model, the NSGA-II multi-objective optimization algorithm is used to iteratively optimize the fire allocation decision variables to output an unmanned position fire allocation scheme that takes into account the influence of terrain. The fire allocation decision variables are the corresponding allocation relationship between weapon platforms and strike targets.

3. The method for optimizing the allocation of unmanned field firepower based on three-dimensional strike effectiveness according to claim 2, characterized in that, The objective function expression for the 3D terrain-adaptive joint fire strike model is: In the formula, V’ ( x () represents the attack value gain after terrain correction; C ( x This indicates the overall cost of the strike; d 3D ( i , j () indicates a three-dimensional terrain range constraint; R’ i This indicates the dynamically adjusted range of the weapon platform. x ij Indicates the decision variables for firepower allocation; LOS ij Indicates visibility factor; The number of weapon platforms allocated to the target; For the purpose of striking the target; m i For the first i The current number of weapon-like platforms; m Indicates the weapon platform category; i Indicates the first i Weapon-like platform; j Indicates the first j One objective; st represents constraints; in: ； ； In the formula, v j Indicates damage to the first The expected value of each goal; This indicates the corrected hit probability; c i Indicates the first The cost of a single strike by a weapon-type platform, where i = 1, 2, ..., m , m Indicates the weapon platform category.

4. The method for optimizing the allocation of unmanned field firepower based on three-dimensional strike effectiveness according to claim 1, characterized in that, The set of topographic influence factors includes the vegetation shading coefficient, which is calculated using the following formula: ； In the formula, V j Indicates the vegetation shading coefficient; G Indicates the blade projection index; LAI j Indicates leaf area index; Indicates the angle between the line of sight and the vertical direction; i Indicates the number of weapon platform categories; j Indicates the number of targets; where, LAI j The calculation formula is: ； In the formula, k represents the extinction coefficient; q Indicates the leaf tilt angle distribution parameters; NDVI j Represents the normalized vegetation index at the j-th target location; NDVI soil Indicates the NDVI baseline value for bare soil; NDVI veg This represents the baseline NDVI value for pure vegetation.

5. The method for optimizing the allocation of unmanned field firepower based on three-dimensional strike effectiveness according to claim 1, characterized in that, The set of terrain influence factors includes terrain correction factors and slope range correction coefficients, wherein: The formula for calculating the terrain correction factor is as follows: ； In the formula, TF ij Indicates the terrain correction factor; LOS ij Indicates visibility factor; α V represents the vegetation shading weighting coefficient; j Indicates the vegetation shading coefficient; The formula for calculating the slope range correction coefficient is as follows: ； In the formula, η ij This represents the slope range correction factor; γ Indicates the uphill range attenuation coefficient; β θ represents the downhill range gain coefficient; crit Indicates the critical slope angle; η max Indicates the maximum range gain limit; θ ij θ represents the slope angle between the i-th weapon platform and the j-th target; where θ ij The calculation formula is: ； In the formula, ( m xi , m yi , E i ) represents the three-dimensional coordinates of the i-th type of weapon platform; (t) xj , t yj , E j () represents the three-dimensional coordinates of the j-th target; Digital elevation values ​​of weapon deployment points extracted from a regular grid digital elevation model; This refers to the digital elevation values ​​of the target points extracted from the regular grid digital elevation model.

6. The method for optimizing the allocation of unmanned field firepower based on three-dimensional strike effectiveness according to claim 2, characterized in that, In the step of iteratively optimizing the fire allocation decision variables using the NSGA-II multi-objective optimization algorithm based on the aforementioned three-dimensional terrain-adaptive joint fire strike model to output an unmanned position fire allocation scheme that takes into account the influence of terrain, the iterative optimization method includes: Initialize the parameters of the NSGA-II multi-objective optimization algorithm and randomly generate an initial population representing the fire allocation scheme; use the modified objective function in the three-dimensional terrain adaptive joint fire strike model as the fitness function to evaluate the strike effectiveness and strike cost of each individual in the population. Perform fast nondominated sorting, crowding calculation, tournament selection, simulated binary crossover, and polynomial mutation operations to produce offspring populations; An elite-preservation strategy is used to generate a new generation of population. The process is iterated until convergence, and a set of Pareto optimal solutions is output. Based on the Pareto solution set, a joint fire allocation scheme that takes into account the influence of terrain is output.

7. The method for optimizing the allocation of unmanned field firepower based on three-dimensional strike effectiveness according to claim 2, characterized in that, The steps for establishing a three-dimensional terrain-adaptive joint fire strike model that considers terrain influence by correcting the strike effectiveness parameters of the joint fire allocation model based on the set of terrain influence factors include: Calculate the three-dimensional Euclidean distance between the weapon platform and the target; The hit probability is adjusted based on the terrain correction factor to obtain the corrected hit probability; The range is corrected according to the slope range correction factor to obtain the corrected range. Based on the three-dimensional Euclidean distance, the corrected hit probability, and the obtained corrected range, a three-dimensional terrain-adaptive joint fire strike model considering the influence of terrain is established.

8. A firepower allocation optimization system for unmanned positions based on three-dimensional strike effectiveness, characterized in that, The system includes: The data acquisition module is used to acquire the preset resolution digital elevation value and discrete vegetation cover data of the area where the unmanned site is located through a geospatial information platform that integrates multi-source remote sensing data, and to calculate the slope value based on the digital elevation value. The spatial fitting module is used to perform spatial fitting on the digital elevation value, vegetation cover index and slope value based on the discrete data using a Gaussian process machine learning model to obtain high-precision data, and generate a high-precision regular grid geographic parameter model based on the high-precision data. The fire allocation module is used to construct an initial joint fire allocation model and calculate a set of terrain influence factors based on the high-precision regular grid geographic parameter model; it corrects the strike effectiveness parameters of the joint fire allocation model according to the set of terrain influence factors to establish a three-dimensional terrain-adaptive joint fire strike model that considers terrain influence and iterates it to output a fire allocation scheme for unmanned positions that considers terrain influence.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the unmanned position firepower allocation optimization method based on three-dimensional strike effectiveness as described in any one of claims 1-7.

10. A data processing apparatus, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the unmanned position firepower allocation optimization method based on three-dimensional strike effectiveness as described in any one of claims 1-7.

Images