A trajectory data analysis method for land warfare game simulation

Abstract

The present invention relates to a trajectory data analysis method applied to land warfare war game simulation, and belongs to the field of war game simulation technology. The method includes: obtaining the trajectory information of combat units, which include infantry, chariots, tanks, and artillery; clustering the trajectory of each combat unit through a clustering algorithm to obtain the trajectory clustering information of each combat unit; determining the hotspot area of ​​each combat unit according to the trajectory clustering information of each combat unit; grouping the combat units to obtain a number of grouping units; weighted superposition of the hotspot areas of each combat unit in the grouping unit to obtain the hotspot area of ​​the grouping unit; and deriving the combat intent according to the hotspot area analysis of the grouping unit. The present invention classifies and groups the combat units, and then makes a more refined and granular intuitive analysis of the combat schematic.

Description

Technical Field

[0001] The invention relates to a trajectory data analysis method applied to land warfare war game simulation, belonging to the technical field of war game simulation. Background Art

[0002] Wargames are an effective means of conducting combat simulations. Their systems mainly include maps, deduction operators, and deduction rules. The deduction parties make a series of decisions and confrontations through analysis and understanding of historical knowledge, environment, and war experience, forming a simulation analysis of operations and wars. Wargames can be used to improve the combat deployment, combat experience, and understanding of operations of relevant personnel, and can play a positive role in the study of military theories, simulated operations, and combat evaluations.

[0003] The trajectory analysis process of war game simulation in the prior art is as follows:

[0004] 1) Simplify the entity's combat trajectory;

[0005] 2) Segment the simplified combat trajectory;

[0006] 3) Clustering the segmented trajectories using a density-based clustering method;

[0007] 4) Extract the statistical features of each cluster and detect the successor relationship of each cluster;

[0008] 5) Form a successor relationship chain based on all successor relationships;

[0009] 6) Draw a visual combat situation diagram based on the successor relationship chain.

[0010] It can be seen from the above existing technologies that the current research on the analysis of trajectory data of land warfare war games is mostly not refined enough, and it is difficult to reflect the data analysis rules of single or grouped combat units. Therefore, it is necessary to propose a trajectory data analysis method for war games that finely divides combat units. Summary of the invention

[0011] The purpose of this application is to provide a trajectory data analysis method applied to land warfare war game simulation, and to propose an effective technical solution for the trajectory data analysis method of war game simulation with fine division of combat units.

[0012] To achieve the above objectives, the present application proposes a technical solution for a trajectory data analysis method applied to land warfare game simulation, comprising the following steps:

[0013] 1) Obtain trajectory information of combat units, including infantry, combat vehicles, tanks, and artillery;

[0014] 2) Clustering the trajectory of each combat unit through a clustering algorithm to obtain the trajectory clustering information of each combat unit;

[0015] 3) Determine the hotspot area of ​​each combat unit based on the trajectory clustering information of each combat unit;

[0016] 4) Grouping the combat units to obtain a number of grouping units;

[0017] 5) The hot spots of each combat unit in the marshaling unit are weighted and superimposed to obtain the hot spots of the marshaling unit; the weights in the weighted superposition process are determined according to the influence and attack capability of the combat unit on the battlefield;

[0018] 6) Determine combat intent based on hot spot analysis of the organizational unit.

[0019] The beneficial effects of the technical solution of the trajectory data analysis method applied to land war game simulation of the present invention are as follows: the present invention classifies combat entities based on the diversity of combat entities, clusters the trajectories of different combat entities through a clustering algorithm, and then groups combat units, obtains the hot spots and spatial distribution rules of the grouping units based on the hot spots of the combat units in the grouping units, and finally obtains an intuitive combat intent based on the grouping units. The present invention classifies and groups combat units, and then makes a more refined and granular intuitive analysis of combat intent.

[0020] Furthermore, before clustering the trajectory of each combat unit, the method further includes a step of simplifying the trajectory information of the combat unit, and clustering the simplified trajectories of the combat units.

[0021] Furthermore, the trajectory information of the combat unit is simplified by replacing the trajectory points that fall on the same hexagonal grid with the center point of the hexagonal grid.

[0022] Furthermore, the clustering algorithm is a DBSCAN algorithm.

[0023] Furthermore, when the organizational units include infantry, combat vehicles and tanks, the weights of the infantry, combat vehicles and tank combat units are 0.1, 0.4 and 0.5 respectively.

[0024] Furthermore, the ε neighborhood of the infantry in the DBSCAN algorithm is 1 hexagonal grid unit, and the threshold MinPoints is 5. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of a trajectory data analysis method of the present invention applied to land warfare war game simulation;

[0026] Figure 2 It is a data classification diagram of the trajectory information deduced by the present invention;

[0027] Figure 3 It is a schematic diagram of the spatial distribution law of the single combat unit to the grouping unit of the present invention;

[0028] Figure 4 The present invention obtains a schematic diagram of the combat intention of the combat party based on the combat units and the formation units. DETAILED DESCRIPTION

[0029] Example of trajectory data analysis method applied to land warfare game simulation:

[0030] The main idea of ​​the present invention is that multiple types of data are included in the land warfare war game simulation process, and the trajectory data of the combat units, in addition to the movement information of the combat units themselves, can also include the associated behavior patterns between the combat units. The combat unit / entity trajectory data in the land warfare war game simulation is a position sequence with time information, which has spatiotemporal attributes. The analysis of its trajectory data often needs to consider the time and space attributes. Based on the trajectory information of multiple types of combat units, the present invention provides a land warfare war game simulation data analysis method that takes into account multiple types of combat units and grouped combat units, and analyzes the hot spots, distribution patterns and intuitive combat intentions of multiple types of combat units and grouped combat units in a more refined and granular manner.

[0031] Specifically, the trajectory data analysis method used in land warfare game simulation is as follows: Figure 1 As shown, the following steps are included:

[0032] 1) Obtain the land warfare war game trajectory information data to be analyzed, and classify the land warfare war game trajectory information data according to combat units.

[0033] In this step, if Figure 1 As shown, the combat units include infantry, combat vehicles, tanks, artillery, and other combat units, so the land warfare game trajectory information data includes infantry trajectory information, combat vehicle trajectory information, tank trajectory information, artillery trajectory information, and other combat unit trajectory information.

[0034] 2) Preprocess the trajectory information of each combat unit to obtain simplified combat unit trajectory information.

[0035] Let the trajectory information of the i-th combat unit be R i , which is expressed as follows

[0036]

[0037] in, is the time information of the jth trajectory point in the trajectory of the i-th combat unit; are the hexagonal grid horizontal and vertical coordinates of the jth trajectory point in the trajectory of the i-th combat unit; I is the total number of combat units; m is the total number of trajectory points.

[0038] For trajectory R i Preprocessing is performed, the main purpose of which is to simplify the trajectory R i , reducing the amount of calculation. The trajectory preprocessing of land war game simulation mainly simplifies the trajectory points that fall on the same hexagonal grid in continuous time by replacing them with the center points of the hexagonal grid.

[0039] After trajectory preprocessing, the simplified trajectory information of the ith combat unit is R i ', which is expressed as follows:

[0040]

[0041] in, is the time information of the jth trajectory point in the simplified trajectory of the i-th combat unit; The jth trajectory point in the simplified trajectory of the i-th combat unit is the horizontal and vertical coordinates of the center point of the hexagonal grid, n is the total number of trajectory points after simplification, and the number of trajectory points is simplified from m to n.

[0042] 3) The simplified trajectory information of the combat unit is clustered by a clustering algorithm to obtain a trajectory clustering result. Let i=1, that is, clustering is performed on the trajectory of the first combat unit.

[0043] In this step, the clustering algorithm uses the DBSCAN algorithm, and the clustering process is as follows:

[0044] a. Establish a knowledge base, which includes different combat units and their corresponding parameters. The parameters include adaptive selection of ε neighborhood and threshold MinPoints. ε neighborhood is the area within the radius ε of a given object, which is called the neighborhood of the object. Threshold MinPoints is the minimum number of neighborhood points for a given point to become a core object in the ε neighborhood. The ε neighborhood is calculated based on the hexagonal grid map of the war game. For example, the ε neighborhood of the infantry combat unit is 1 hexagonal grid unit, and the threshold MinPoints is set to 5, which means that if there are more than 5 infantrymen within a hexagonal grid distance, it is considered to be an infantry cluster.

[0045] b. After the simplification of the combat unit trajectory information R i', and find all the trajectory points whose distance to this trajectory point is less than or equal to the ε neighborhood (the distance information is calculated through the hexagonal grid of the chess map); then get the total number of all trajectory points. If the total number is less than the threshold MinPoints, the trajectory point is regarded as a noise point; if the total number is greater than or equal to the threshold MinPoints, the trajectory point is regarded as a core sample point and a new label cluster is assigned.

[0046] c. If the trajectory point is a core sample point, then use the same method as step b to find other core sample points in the ε neighborhood of the core sample point and expand the cluster.

[0047] d. Select other trajectory points and repeat steps b and c to complete the clustering process of the first combat unit.

[0048] 4) Determine the hotspot area of ​​the first combat unit based on the trajectory clustering results: In the clustering results, the area with higher density is the hotspot area. The area with higher density is the area with larger clusters in the clustering results. The clustering analysis results of a combat unit can get several hotspot areas, which reflects that this area covers more sample points and can be classified into the same cluster.

[0049] The spatial distribution law of the first combat unit is obtained based on the hot spot area of ​​the first combat unit and the environmental map.

[0050] The clustering results can reflect the hot spots of combat units. By mapping the hot spots to specific environmental maps, the spatial distribution characteristics of combat units in the map can be obtained. The spatial distribution characteristics are the distribution laws of combat units.

[0051] 5) Let i = 2, repeat steps 3) to 4) to obtain the hotspot area and spatial distribution law of the second combat unit, until i ≥ I, to obtain the hotspot area and spatial distribution law of all combat units.

[0052] 6) Group the combat units to obtain several grouping units.

[0053] The marshaling unit is obtained by combining combat units, for example: Figure 2 The infantry + combat vehicle + tank shown is a grouping unit. Of course, the specific grouping rules of the grouping unit can be determined by the joint effect of different combinations of combat units, or can be set as needed, or the grouping unit can be determined according to the object to be studied and analyzed. The grouping unit is similar to the force organization. For example, a synthetic battalion includes tanks, infantry, artillery, and infantry. Therefore, the study of the synthetic battalion grouping unit requires the joint consideration of tanks, infantry, artillery, and infantry combat units. Of course, if there is only one combat unit in the force organization, then there can also be only one combat unit in the grouping unit.

[0054] 7) Determine the hot spot area of ​​the marshaling unit according to the hot spot area of ​​each combat unit in the marshaling unit, and then determine the spatial distribution law of the marshaling unit. Let t = 1, that is, determine the hot spot area and spatial distribution law of the first marshaling unit.

[0055] The process of determining the hotspot area and spatial distribution law of the marshaling unit is as follows:

[0056] First, determine the combat units of the formation unit;

[0057] Secondly, the hot spots of the combat units are weighted and superimposed to obtain the hot spots of the formation units;

[0058] Finally, the hot spots of the marshalling units are mapped to the specific environmental map to obtain the spatial distribution pattern of the marshalling units.

[0059] The weights of different combat units are not the same. The size of the weight is determined by the combat unit's influence on the battlefield and its attack capability. For example: Figure 3 As shown in the figure, the formation unit includes infantry, combat vehicles and tanks. First, the hot spots of infantry, combat vehicles and tanks are obtained by clustering algorithm, and then the hot spots of the formation unit are obtained by weighted superposition of the hot spots of infantry, combat vehicles and tanks. Since the influence of infantry on the battlefield is far less than that of combat vehicles and tanks, the weights of infantry, combat vehicles and tank combat units are 0.1, 0.4 and 0.5 respectively.

[0060] 8) Let t=2, repeat step 7) to determine the hotspot area and spatial distribution law of the second marshaling unit, until t is greater than or equal to T, and the hotspot area and spatial distribution law of all marshaling units are determined.

[0061] 9) Analyze combat intent based on the hot spots and spatial distribution patterns of the marshaling units and provide intuitive combat intent.

[0062] Hotspots are important analytical locations for realizing troop deployment, material supply, personnel convergence, combat analysis, etc. Spatial distribution patterns are important analytical means for realizing and studying troop deployment, combat assessment, etc.

[0063] In this step, according to the rules in the previously established knowledge base, as well as the hot spots and spatial distribution laws of the formation units, the dovetail arrows can be used to intuitively display the combat intent of the combat party. The combat intent is the trend direction of attack or retreat. According to the different targets to be attacked, the weights can be set by yourself for acquisition. For example, according to the distribution law of the red formation unit 1 (including infantry and chariots) and the distribution law of the formation unit 2 (including tanks and cruise missiles), the final red combat distribution law is obtained by weighted fusion (formation unit 1 is set to 0.3, and formation unit 2 is set to 0.7). By observing the final distribution law and combining the actual trajectory direction, the combat intent is plotted on the map using dovetail arrows, which is the obtained combat intent.

[0064] like Figure 4 As shown, according to the distribution patterns generated by single combat units and formation units, the influence deployment of the red and blue forces is formed respectively. Combined with the war game map environment, the combat intentions of the combat parties are obtained through the fusion and observation of multiple data information.

[0065] In the above embodiment, the clustering algorithm adopts the DBSCAN algorithm. As other implementations, other clustering algorithms such as the DENCLUE algorithm may also be adopted, and the present invention does not limit this.

[0066] The present invention can analyze the data information of the army war game from multiple angles. Compared with the traditional method, it can analyze the data information of the war game better and more comprehensively, and provide new solutions for studying troop deployment, combat evaluation, and combat effectiveness. The present invention can also be used in other fields such as diversified traffic data analysis, providing a reliable foundation for further improving the analysis of various types of data.

Claims

1. A trajectory data analysis method applied to land warfare game simulation, characterized in that: The following steps are involved: 1) Obtain trajectory information of combat units, including infantry, combat vehicles, tanks, and artillery; 2) Clustering the trajectory of each combat unit through a clustering algorithm to obtain the trajectory clustering information of each combat unit; 3) Determine the hotspot area of ​​each combat unit based on the trajectory clustering information of each combat unit; 4) Grouping the combat units to obtain a number of grouping units; 5) The hot spots of each combat unit in the marshaling unit are weighted and superimposed to obtain the hot spots of the marshaling unit; the weights in the weighted superposition process are determined according to the influence and attack capability of the combat unit on the battlefield; 6) Determine combat intent based on hot spot analysis of the organizational unit.

2. The trajectory data analysis method for land warfare game simulation according to claim 1 is characterized in that: Before clustering the trajectory of each combat unit, the method further includes simplifying the trajectory information of the combat unit, and clustering the simplified trajectory of the combat unit.

3. The trajectory data analysis method for land warfare game simulation according to claim 2 is characterized in that: The method of simplifying the trajectory information of the combat unit is to simplify the trajectory points that fall on the same hexagonal grid by replacing them with the center point of the hexagonal grid.

4. The trajectory data analysis method for land warfare game simulation according to claim 1 is characterized in that: The clustering algorithm is the DBSCAN algorithm.

5. The trajectory data analysis method for land warfare game simulation according to claim 1 is characterized in that: When the organizational units include infantry, combat vehicles and tanks, the weights of infantry, combat vehicles and tank combat units are 0.1, 0.4 and 0.5 respectively.

6. The trajectory data analysis method for land warfare game simulation according to claim 4 is characterized in that: In the DBSCAN algorithm, the ε neighborhood of the infantry is 1 hexagonal grid unit, and the threshold MinPoints is 5.

Images