Urban confrontation-oriented task allocation and path planning method and system

Abstract

The invention discloses a task allocation and path planning method and system for urban confrontation. The task allocation and path planning method for urban confrontation comprises the following steps: acquiring real-time comprehensive situation information of an urban confrontation scene containing positions of all units of two confrontation parties; obtaining a shapefile image of the current confrontation area from GIS software; analyzing the shapefile image and extracting path key points in the urban road network; calculating passable paths between the units of the two confrontation parties according to the positions of the units of the two confrontation parties; establishing a maneuvering efficiency constraint model, a maneuvering threat constraint model and a marshalling firepower constraint model; constructing a target function; and solving a task allocation and path planning scheme. The invention aims to realize task allocation and path planning for urban confrontation, and improves the applicability and accuracy of task allocation and path planning in different urban confrontation scenes.

Description

Technical Field

[0001] The present invention belongs to the technical field of task allocation and path planning, and in particular relates to a method and system for task allocation and path planning for urban confrontation. Background Art

[0002] In urban confrontation, reasonable task allocation and path planning for our units based on confrontation goals, confrontation plans and real-time comprehensive confrontation situation are of great significance for reducing personnel and weapon equipment losses and improving confrontation effectiveness. Existing task allocation and path planning methods rarely consider the complex characteristics of urban confrontation (for example, key situation information such as road network structure and building distribution in different confrontation scenarios often have significant differences). Therefore, they are difficult to efficiently extract key points of paths and calculate traversable paths in different urban confrontation scenarios.

[0003] In addition, when analyzing and predicting the mobility and formation strategies of multiple units of our side, most of the existing task allocation and path planning methods lack comprehensive consideration of key confrontation factors such as unit type, basic unit attributes, day and night environment, and path conditions. They also do not take into account the impact of mobility efficiency, mobility threats, and formation firepower on task allocation and path planning solutions, resulting in the obtained solutions being insufficiently practical and difficult to adapt to real urban confrontation tasks. Summary of the invention

[0004] Technical problem to be solved by the present invention: In view of the above-mentioned problems in the prior art, a method and system for task allocation and path planning for urban confrontation are provided. The present invention aims to realize task allocation and path planning for urban confrontation, and to improve the applicability and accuracy of task allocation and path planning in different urban confrontation scenarios.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: A task allocation and path planning method for urban confrontation includes the following steps: S1, obtaining real-time comprehensive situation information of the urban confrontation scene including the positions of each unit of the confrontation parties; S2, obtain the shapefile image of the current confrontation area from the GIS software; S3, analyzes the shapefile image and extracts key points of paths in the urban road network; S4, calculating the traversable paths between the units of the two opposing sides according to the positions of the units of the two opposing sides; S5, combining the real-time comprehensive situation information of the urban confrontation scene and the traversable paths between the units of the two opposing sides, calculating all the key confrontation elements about the traversable paths and establishing a mobility efficiency constraint model, a mobility threat constraint model and a formation fire constraint model, wherein the key confrontation elements include unit type, unit basic attributes, day and night environment and path conditions; S6, constructing the objective function based on the mobility efficiency constraint model, mobility threat constraint model and group fire constraint model; S7, solve the task allocation and path planning solutions based on the constructed objective function.

[0006] Optionally, step S3 includes: S3.1, binarize the shapefile image to obtain a binary image; S3.2, extract polygon contours in the binary image to obtain the city’s road network and building information; S3.3, simplify the polygonal contour line and extract the vertices of the polygonal contour line as the path key points in the urban road network.

[0007] Optionally, step S4 includes: S4.1, use GPU or CPU parallel computing to analyze the connectivity between key points on the path: for any two key points on the path, calculate the difference between the ordinate and the abscissa, and take the maximum value of the two as the number of sampling points ; Uniformly sample the line segment connecting two path key points points, and round the coordinates of each point, and finally determine the sampled Whether there is a black point among the points, if there is a black point, it is determined that the two key points on the path are not connected; otherwise, it is determined that the two key points on the path are connected; S4.2, judging whether the position coordinates of each unit of the opposing parties are in the building according to the pixel values ​​of the positions of each unit of the opposing parties in the shapefile image, if the pixel value of the position coordinates in the shapefile image is black, it is determined that the unit is in the building, otherwise, it is determined that the unit is not in the building; S4.3, for the unit in the building, the position coordinate point of the unit is considered to be connected with the key point of the path corresponding to the building; for the unit not in the building, the connectivity of the position coordinate point of the unit with each key point of the path is determined one by one; finally, the connectivity of the units of the two adversaries with all the key points of the path is obtained; S4.4, based on the connectivity between the units of the two adversaries and all key points of the path, use the specified path planning algorithm to calculate the traversable path between the units of the two adversaries.

[0008] Optionally, the function expression of the maneuvering efficiency constraint model established in step S5 is: , , in, We will organize our team to go to The maneuver time of the opponent's unit; Boolean variable Indicates our Whether the unit decides from Passable paths to attack the opponent units, if so, then ,otherwise ,in , Indicates the number of our units, , represents the number of opponent units, , Indicates our Units reach the opponent's The total number of traversable paths per unit; Indicates our Units from Passable paths to attack the opponent The time required for each unit; Indicates the unit that attacks the opponent among all our units. The number of support units per unit. The more support units there are, the higher the mobility efficiency. It represents the gain of the support unit on maneuver efficiency; The function expression of the maneuver threat constraint model established in step S5 is: , In the above formula, For our unit to go to The mobile threat of an opponent's unit, is the importance value of our unit; Indicates our Units attack the opponent Is there a feasible path when there are units? If there is a feasible path, then ,otherwise, ; Indicates our Units from Passable paths to attack the opponent When there are units, the distance that can be seen by the opponent's unit; Indicates our Units from Passable paths to attack the opponent The comprehensive hazard level of the path when the unit is For our The defensive capability value of each unit; The function expression of the group firepower constraint model established in step S5 is: , , in, Indicates that our team's firepower is in conflict with the opponent's team's The minimum ratio of firepower per unit; Indicates that our team's firepower is in conflict with the opponent's team's The maximum ratio of the firepower of each unit, For the opponent The offensive firepower value of a unit; Boolean variable Indicates our Whether the unit decides from Passable paths to attack the opponent units, if so, then ,otherwise ; For our The offensive firepower value of each unit; The remaining firepower value for all our formations, It indicates the increase in the offensive firepower of our units from the logistics units. Indicates the unit that attacks the opponent among all our units. The number of logistic units per unit, Indicates the gain of the command unit's offensive firepower to our unit, a Boolean variable Indicates attacking the opponent Can our units maintain communication with the command unit? If so, ,otherwise ; Indicates the unit that attacks the opponent among all our units. The number of command units of a unit, For the opponent The offensive firepower value of each unit; The function expression of the objective function constructed in step S6 is: , in, They represent the weights of mobility efficiency, residual firepower and mobility threat respectively.

[0009] Optionally, the Units from Passable paths to attack the opponent The calculation function expression of the comprehensive danger degree of the path when the unit is: , in, Indicates our Units from Passable paths to attack the opponent The comprehensive hazard level of the path when the unit is ~ Represent the weights of different factors. represents the bias, Indicates our Units from Passable paths to attack the opponent The number of traffic intersections or crossroads on the path when the unit is Indicates our Units from Passable paths to attack the opponent The number of bridges on the path when the unit is Indicates our Units from Passable paths to attack the opponent The number of tunnels on the path when the unit is Indicates our Units from Passable paths to attack the opponent The ratio of the route length covered by the opponent's firepower to the total route length when the opponent fires a unit.

[0010] Optionally, the Units from Passable paths to attack the opponent The calculation function expression of the time required for each unit is: , in, Indicates our Units from Passable paths to attack the opponent The time required for each unit; Indicates our Units from Passable paths to attack the opponent The distance of the path when the unit is Indicates our Units attack the opponent Is there a feasible path when there are units? If there is a feasible path, then ,otherwise, ; Indicates our Units from Passable paths to attack the opponent The congestion level of the path when the unit is For our The speed of the unit, when the confrontation takes place during the day, our The speed of the unit is ,in, Indicates our The maximum maneuvering speed of a unit during the day; when the confrontation takes place at night, our The speed of the unit is ,in Indicates our The maximum maneuvering speed of a unit at night.

[0011] Optionally, in step S7, solving the task allocation and path planning scheme for the constructed objective function is to use an improved polygene adaptive genetic-simulated annealing algorithm to solve the task allocation and path planning scheme for the constructed objective function, including: S7.1, a multi-layer polygene coding strategy is used to solve the problem of task allocation and path planning. The multi-layer polygene coding strategy includes a task layer, a unit layer and an action route layer, wherein the task layer is mainly composed of task numbers and an integer coding strategy is adopted; the unit layer is composed of the numbers of our units, and a vacant gene mechanism is introduced to randomly set some genes to "-1" to indicate that the task is not assigned to any unit; the action route layer adopts an integer coding strategy; S7.2, generating an initial population based on the constraints in the mobility efficiency constraint model, the mobility threat constraint model, and the group fire constraint model; S7.3, use the objective function to calculate the fitness value of each individual in the current population: S7.4, sort the individuals in the current population according to the fitness value, select the individuals with the specified proportion to form the elite population, and the rest of the individuals are the ordinary population; S7.5, the elite population is subjected to adaptive simulated annealing to generate new individuals, wherein the adaptive simulated annealing to generate new individuals refers to randomly selecting new individuals in the neighborhood of the current individual and accepting them according to a specified probability; the individuals in the ordinary population are subjected to adaptive probability. Select individuals from the elite population for crossover or self-crossover, where N iter Indicates the current iteration number, Nepoch Represents the total number of iterations; the new individuals obtained by crossover or self-crossover between the common population and the individuals in the elite population are selected according to the adaptive mutation probability Implement mutation operation, and when implementing mutation operation, the number of mutation points is random, so that different individuals in the same population have different mutation points; only the unit layer and the action route layer participate in mutation; select a specified number of viable offspring from the current population and new individuals as the new current population; S7.6, determine whether the number of iterations reaches the preset threshold. If it has not reached the preset threshold, jump to step S7.4 to continue iteration; otherwise, use the objective function to calculate the fitness value of the new individuals in the new current population, and select the individual with the best fitness value as the final optimal task allocation and path planning solution.

[0012] In addition, the present invention also provides a task allocation and path planning system for urban confrontation, including a microprocessor and a memory connected to each other, and the microprocessor is programmed or configured to execute the task allocation and path planning method for urban confrontation.

[0013] In addition, the present invention also provides a computer-readable storage medium, which stores a computer program or instruction, and the computer program or instruction is programmed or configured to execute the task allocation and path planning method for urban confrontation through a processor.

[0014] In addition, the present invention also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the task allocation and path planning method for urban confrontation through a processor.

[0015] Compared with the prior art, the present invention has the following advantages: The task allocation and path planning method for urban confrontation in the present invention extracts the road network distribution, building information and path key points in the urban confrontation scene to obtain multiple passable paths between the units of the two opposing sides. On this basis, the method comprehensively considers key confrontation factors such as unit type, unit basic attributes, day and night environment, path conditions, etc., constructs a maneuver efficiency constraint model, a maneuver threat constraint model and a group firepower constraint model, and uses an improved multi-gene adaptive genetic-simulated annealing algorithm to solve a reasonable task allocation and path planning scheme. First, the present invention realizes a universal mechanism for extracting urban road network features and path key points, which can be applied to different urban confrontation scenarios. Secondly, in the process of generating task allocation and path planning schemes, the present invention uses a maneuver efficiency constraint model, a maneuver threat constraint model and a group firepower constraint model to uniformly optimize the scheme from three aspects: maneuver efficiency, maneuver threat and group firepower. The present invention can extract key points of paths from different confrontation areas and obtain multiple passable paths between any two points, and can efficiently generate task allocation and path planning solutions based on real-time confrontation situation information, thereby obtaining the maneuvering strategy and formation strategy of our units, thereby realizing task allocation and path planning for urban confrontation and improving the applicability and accuracy of task allocation and path planning in different urban confrontation scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of the basic flow of the method of the embodiment of the present invention.

[0017] Figure 2 Schematic diagram of the connectivity judgment principle between key points of a path in an embodiment of the present invention.

[0018] Figure 3 Schematic diagram of the principle of parallel processing in an embodiment of the present invention.

[0019] Figure 4 The figure is a flow chart of the improved multi-gene adaptive genetic-simulated annealing algorithm in the embodiment of the present invention.

[0020] Figure 5 Schematic diagram of the multi-layer multi-gene encoding method used in the embodiment of the present invention.

[0021] Figure 6 1 is the original Shapefile image of city 1 and the image binarization result in the embodiment of the present invention.

[0022] Figure 7 This is the result of extracting the polygonal contour lines of the image of city 1 in the embodiment of the present invention.

[0023] Figure 8This is the path key point extraction result of city 1 in the embodiment of the present invention.

[0024] Fig. 9 This is a search result for a drivable path between any two points in city 1 in an embodiment of the present invention.

[0025] Fig.10 is the weight of the objective function of the city in the embodiment of the present invention Task allocation and path planning results at this time.

[0026] Fig.11 is the weight of the objective function of the city in the embodiment of the present invention Task allocation and path planning results at this time.

[0027] Fig.12 is the firepower weight of a city group in the embodiment of the present invention Task allocation and path planning results at this time.

[0028] Fig.13 is the firepower weight of a city group in the embodiment of the present invention Task allocation and path planning results at this time.

[0029] Fig.14 It is a top view of City 2 based on Unreal Engine in an embodiment of the present invention.

[0030] Fig.15 is the weight of the objective function of city 2 in the embodiment of the present invention Task allocation and path planning results at this time.

[0031] Fig.16 is the weight of the objective function of city 2 in the embodiment of the present invention Task allocation and path planning results at this time.

[0032] Fig.17 is the weight of the objective function of city 2 in the embodiment of the present invention Task allocation and path planning results at this time.

[0033] Fig.18 is the weight of the objective function of city 2 in the embodiment of the present invention Task allocation and path planning results at this time.

[0034] Fig.19 is the firepower weight of the second city group in the embodiment of the present invention Task allocation and path planning results at this time.

[0035] Fig. 20 is the firepower weight of the second city group in the embodiment of the present invention Task allocation and path planning results at this time.

[0036] Fig.21 is the firepower weight of the second city group in the embodiment of the present invention Task allocation and path planning results at this time. DETAILED DESCRIPTION

[0037] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0038] like Figure 1 As shown, the task allocation and path planning method for urban confrontation in this embodiment includes the following steps: S1, obtaining real-time comprehensive situation information of the urban confrontation scene including the positions of each unit of the confrontation parties; S2, obtain the shapefile image of the current confrontation area from the GIS software; S3, analyzes the shapefile image and extracts key points of paths in the urban road network; S4, calculating the traversable paths between the units of the two opposing sides according to the positions of the units of the two opposing sides; S5, combining the real-time comprehensive situation information of the urban confrontation scene and the traversable paths between the units of the two opposing sides, calculating all the key confrontation elements about the traversable paths and establishing a mobility efficiency constraint model, a mobility threat constraint model and a formation fire constraint model, wherein the key confrontation elements include unit type, unit basic attributes, day and night environment and path conditions; S6, constructing the objective function based on the mobility efficiency constraint model, mobility threat constraint model and group fire constraint model; S7, solve the task allocation and path planning solutions based on the constructed objective function.

[0039] In step S1, the real-time comprehensive situation information of the urban confrontation scene including the positions of each unit of the confrontation parties is obtained. The key is the position information of each unit of each confrontation party. In addition, some other parameter definitions are also included. The real-time comprehensive situation information in this embodiment is obtained from a real game confrontation scene. The real-time comprehensive situation information of the confrontation specifically includes: Our The location of the unit ,in, Indicates our The horizontal coordinate of the unit, Indicates our The vertical coordinate of the unit, , Indicates the total number of our units. The location of the unit ,in, Indicates the opponent The horizontal coordinate of the unit, Indicates the opponent The vertical coordinate of the unit, , Represents the total number of counterparty units.

[0040] The types of our units include command units, logistics units, armored units, and support units, and a single unit can only belong to one of them. Indicates our Is the unit a command unit? If so, ,otherwise . Similarly, a Boolean variable Indicates our Is the unit a logistics unit? If so, ,otherwise . Boolean variables Indicates our Is the unit an armored unit? If so, ,otherwise . Boolean variables Indicates our Is the unit a guarantee unit? If so, ,otherwise .

[0041] Our Offensive firepower of a unit It is closely related to the type of unit and specifically meets the following conditions: , Among them, the offensive firepower value of the command unit is expressed as , the offensive firepower value of the logistics unit is expressed as , the offensive firepower value of the armored unit is expressed as , the offensive firepower value of the support unit is expressed as .

[0042] Opponent The offensive firepower of a unit is .

[0043] Our The defensive capability of each unit It is also closely related to the type of unit, and specifically meets the following conditions: , Among them, the defense capability value of the command unit is expressed as , the defense capability value of the logistics unit is expressed as The defensive capability value of the armored unit is expressed as , the defense capability value of the security unit is expressed as .

[0044] Our The importance value of each unit Specifically meet the following conditions: , Among them, Boolean variables Used to indicate whether the confrontation took place during the day or at night. If the confrontation took place during the day, then ; If the confrontation takes place at night, then The importance of the command unit during the day is expressed as , the importance value at night is expressed as , the importance value of the logistics unit is expressed as The importance of armored units is expressed as The importance of the security unit is expressed as .when The confrontation took place during the day. The speed of the unit is ,in, Indicates our The maximum maneuvering speed of a unit during the day; The confrontation took place at night. The speed of the unit is ,in, Indicates our The maximum maneuvering speed of a unit at night.

[0045] In this embodiment, step S2 obtains the shapefile image of the current confrontation area from the GIS software; Shapefile is a common Geographic Information System (GIS) data format used to store geographic spatial data, such as geometric shapes of points, lines, and polygons. In this embodiment, the Shapefile image data of the current urban confrontation area is directly obtained from the relevant GIS software.

[0046] Step S3 is used to analyze the Shapefile image data and extract key points of the path in the urban road network. In this embodiment, step S3 includes: S3.1, binarize the shapefile image to obtain a binary image; As an optional implementation, in this embodiment, the cv2.cvtColor function in the computer vision and image processing open source library OpenCV is used to convert the Shapefile image data in step 2 into a grayscale image, and the cv2.adaptiveThreshold function is used to perform adaptive thresholding to convert the grayscale image into a binary image.

[0047] S3.2, extract polygon contours in the binary image to obtain the city’s road network and building information; As an optional implementation, in this embodiment, the cv2.findContours function is used to extract polygon contours in the binary image to obtain the road network and building information of the city; S3.3, simplify the polygonal contour and extract the vertices of the polygonal contour as the key points of the path in the urban road network. As an optional implementation, in this embodiment, the shapely.geometry.Polygon function in the spatial geometry computing open source library Shapely is used to simplify the existing polygonal contour, and then the polygon.exterior.coords function is used to extract the vertices of the polygonal contour, which are the key points of the path.

[0048] Step S4 is used to analyze the connectivity between key points of the path, and calculate the passable path between the units of the two opposing parties according to the positions of the units of the two opposing parties. Specifically, step S4 in this embodiment includes: S4.1, using GPU or CPU parallel computing to analyze the connectivity between key points on the path: The connectivity determination method between key points on the path is as follows: Figure 2 As shown in the figure, for any two path key points, the vertical coordinate difference and horizontal coordinate difference between the two path key points are calculated, and the maximum value of the two is taken as the number of sampling points. ; Uniformly sample the line segment connecting two path key points points, and round the coordinates of each point, and finally determine the sampled Whether there is a black point among the points, if there is a black point, it is determined that the two key points on the path are not connected; otherwise, it is determined that the two key points on the path are connected; S4.2, judging whether the position coordinates of each unit of the opposing parties are in the building according to the pixel values ​​of the positions of each unit of the opposing parties in the shapefile image, if the pixel value of the position coordinates in the shapefile image is black, it is determined that the unit is in the building, otherwise, it is determined that the unit is not in the building; S4.3, for the unit in the building, the position coordinate point of the unit is considered to be connected with the key point of the path corresponding to the building; for the unit not in the building, the connectivity of the position coordinate point of the unit with each key point of the path is determined one by one; finally, the connectivity of the units of the two adversaries with all the key points of the path is obtained; S4.4, according to the connectivity between the units of the two opposing parties and all the key points of the path, a specified path planning algorithm is used to calculate the passable path between the units of the two opposing parties. For example, as an optional implementation, in this embodiment, the A-star algorithm is used to calculate the passable path between the units of the two opposing parties, wherein the A-star algorithm is a well-known path search algorithm.

[0049] In step S4.1, the connectivity between the key points of the analysis path is analyzed by using GPU or CPU in parallel, aiming to improve the efficiency of computing the connectivity between the key points of the analysis path by using parallel processing. The basic idea of ​​parallel processing is as follows Figure 3 As shown, assuming that the parallel processing is 3 threads (thread 1 to thread 3), the first parallel processing can process 3 tasks (data 1 to data 3) at the same time. For thread 1, the second processing data is 4, and the number of steps of processing data for each thread is the number of threads. This embodiment uses the CUDA programming module in the Numba package to implement GPU parallel computing. First, use the cuda.grid(2) method in the CUDA programming module to obtain the unique index of the cuda thread, which can be represented as a two-dimensional coordinate Then, we calculate by cuda.gridDim.x×cuda.blockDim.x and cuda.gridDim.y×cuda.blockDim.y respectively. Direction and The iteration step in the direction. cuda.gridDim represents the number of blocks in the grid, and cuda.blockDim represents the number of threads in the block. The multiplication of the two is the total number of threads in the grid. Since all threads process data at the same time, the number of steps forward in each iteration is equal to the total number of threads. For any two points and ,if and Unicom, then and To avoid double counting, Points in the direction, starting from the starting point, for Direction point, from The traversal starts from the current point of the traversal. In addition, the idea of ​​using CPU to implement parallel computing is similar to that of using GPU. First, a list of point pairs to be processed by each CPU is obtained according to the number of cores, and then the map method of the Pool module of the Multiprocessing package is used to implement parallel computing.

[0050] Based on the real-time comprehensive situation information of the confrontation in step S1 and the passable path between the confrontation units in step S4, all key confrontation elements about the passable path are calculated in step S5 of this embodiment, including: Indicates our Units from Passable paths to attack the opponent When there are units, the distance that the opponent's units can see through. The specific calculation steps are as follows: First, in our Units attack the opponent Unit Uniform sampling on traversable paths Then, obtain the visible relationship between each path point and all opponent units. For each path point, if there is at least one opponent unit that can see through the path point, then mark the path point as visible. If two adjacent path points are both visible, then the distance between the two path points is marked as the visible distance. After finding the path points, sum up all the visible distances to find the total distance that the opponent's unit can see on this path. .

[0051] Indicates our Units from Passable paths to attack the opponent The congestion level of the path when 1 unit is reached.

[0052] Indicates our Units from Passable paths to attack the opponent The number of traffic intersections or crossroads on the path when the unit is Indicates our Units from Passable paths to attack the opponent The number of bridges on the path when the unit is Indicates our Units from Passable paths to attack the opponent The number of tunnels on the path when the unit is Indicates our Units from Passable paths to attack the opponent The ratio of the route length covered by the opponent's firepower to the total route length when the opponent fires a unit. Indicates our Units from Passable paths to attack the opponent When the comprehensive danger level of the path is Units from Passable paths to attack the opponent The calculation function expression of the comprehensive danger degree of the path when the unit is: , in, Indicates our Units attack the opponent's unit from the accessible path The comprehensive hazard level of the path when the unit is ~ Represent the weights of different factors, represent the bias, Indicates our Units from The number of traffic intersections or crossroads on the path when attacking the opponent's unit through a passable path; Indicates our Units from Passable paths to attack the opponent The number of bridges on the path when the unit is Indicates our Units from Passable paths to attack the opponent The number of tunnels on the path when the unit is Indicates our Units from Passable paths to attack the opponent The ratio of the route length covered by the opponent's firepower to the total route length when the opponent fires a unit.

[0053] Indicates our Units from Passable paths to attack the opponent The distance of the path when the unit

[0054] Indicates our Units from Passable paths to attack the opponent The time required for each unit is Units from Passable paths to attack the opponent The calculation function expression of the time required for each unit is: , in, Indicates our Units from Passable paths to attack the opponent The time required for each unit; Indicates our Units from Passable paths to attack the opponent The distance of the path when the unit is Indicates our Units attack the opponent Is there a feasible path when there are units? If there is a feasible path, then ,otherwise, ; Indicates our Units from Passable paths to attack the opponent The congestion level of the path when the unit is For our The speed of the unit, when the confrontation takes place during the day, our The speed of the unit is ,in, Indicates our The maximum maneuvering speed of a unit during the day; when the confrontation takes place at night, our The speed of the unit is ,in Indicates our The maximum maneuvering speed of a unit at night.

[0055] In this embodiment, when establishing the maneuver efficiency constraint model in step S5, a single unit of our side is constrained not to attack multiple units of the opponent side at the same time, and can only select one path from the available paths to launch an attack. , The following constraints are met: , The mobility efficiency of our units needs to take into account the organization of our units, the path selection and the number of support units. and , the function expression of the maneuvering efficiency constraint model established in step S5 of this embodiment is: , , in, We will organize our team to go to The maneuver time of the opponent's unit; Boolean variable Indicates our Whether the unit decides from Passable paths to attack the opponent units, if so, then ,otherwise ,in , Indicates the number of our units, , represents the number of opponent units, , Indicates our Units reach the opponent's The total number of traversable paths per unit; Indicates our Units from Passable paths to attack the opponent The time required for each unit; Indicates the unit that attacks the opponent among all our units. The number of support units per unit. The more support units there are, the higher the mobility efficiency. represents the gain of the support unit on the mobility efficiency, and: , Our team went to Maneuver time of opponent units The larger it is, the lower the maneuvering efficiency is.

[0056] The mobile threat of our units needs to comprehensively consider the composition of our units, their importance, their defensive capabilities, the passability of the paths, the line of sight of the paths, and the overall danger level of the paths. and , the function expression of the mobile threat constraint model established in step S5 of this embodiment is: , In the above formula, For our unit to go to The mobile threat of an opponent's unit, For our The importance value of each unit; Indicates our Units attack the opponent Is there a feasible path when there are units? If there is a feasible path, then ,otherwise, ; Indicates our Units from Passable paths to attack the opponent When there are units, the distance that can be seen by the opponent's unit; Indicates our Units from Passable paths to attack the opponent The comprehensive hazard level of the path when the unit is For our The defensive capability value of each unit; The grouping firepower of our units needs to take into account the grouping of our units, offensive capabilities and other factors. In order to limit the grouping firepower of our units within a reasonable range, the following constraints can be established: , in, Indicates that our team's firepower is in conflict with the opponent's team's The minimum ratio of firepower per unit; Indicates that our team's firepower is in conflict with the opponent's team's The maximum ratio of the firepower of each unit, For the opponent The offensive firepower value of each unit; the functional expression of the group firepower constraint model established in step S5 is: , , in, Indicates that our team's firepower is in conflict with the opponent's team's The minimum ratio of firepower per unit; Indicates that our team's firepower is in conflict with the opponent's team's The maximum ratio of the firepower of each unit, For the opponent The offensive firepower value of a unit; Boolean variable Indicates our Whether the unit decides from Passable paths to attack the opponent units, if so, then ,otherwise ; For our The offensive firepower value of each unit; The remaining firepower value for all our formations, It indicates the increase in the offensive firepower of our units from the logistics units. Indicates the unit that attacks the opponent among all our units. The number of logistic units per unit, Indicates the gain of the command unit's offensive firepower to our unit, a Boolean variable Indicates attacking the opponent Can our units maintain communication with the command unit? If so, ,otherwise ; Indicates the unit that attacks the opponent among all our units. The number of command units of a unit, For the opponent The offensive firepower value of a unit.

[0057] According to the mobility efficiency constraint model, the mobility threat constraint model and the group firepower constraint model, it is necessary to minimize the mobility threat of our unit and maximize the mobility efficiency and the remaining firepower. The functional expression of the objective function constructed in step S6 of this embodiment is: , in, They represent the weights of mobility efficiency, residual firepower and mobility threat respectively.

[0058] In step S7 of this embodiment, the task allocation and path planning scheme for solving the constructed objective function is solved by using an improved multi-gene adaptive genetic-simulated annealing algorithm. Figure 4 The flowchart of the improved polygenic adaptive genetic-simulated annealing algorithm in this embodiment is shown in the figure. This embodiment adopts the improved polygenic adaptive genetic-simulated annealing algorithm to solve the objective function, and then generate a reasonable task allocation and path planning scheme. The process of solving the constraint model by this algorithm is as follows: first, according to the confrontation situation information such as the unit type and basic attributes of the units of the two opposing parties, a multi-layer polygenic encoding that meets the constraint conditions is designed, and the initial population size is reasonably selected according to the number of units of the two opposing parties. Secondly, the objective function is mapped to the fitness value function of each chromosome, and some dominant individuals are selected in each round of iteration to form an elite population. At the same time, according to the size of the units of the two opposing parties, appropriate adaptive crossover factor α, mutation factor β, and simulated annealing factor γ are selected, and a variety of crossover and mutation strategies are cited. Finally, through multiple iterations of this framework, a better feasible solution is finally found. Specifically, Figure 4 As shown, in step S7 of this embodiment, solving the task allocation and path planning scheme for the constructed objective function includes: S7.1, a multi-layer polygene encoding strategy is used to perform polygene encoding for solving the problem of task allocation and path planning solutions, such as Figure 5As shown in the figure, the multi-layer multi-gene coding strategy includes the task layer, the unit layer and the action route layer, wherein the task layer is mainly composed of the task number and adopts the integer coding strategy; the unit layer is composed of the numbers of our units, and the vacant gene mechanism is introduced to randomly set some genes to "-1" to indicate that the task is not assigned to any unit; the action route layer adopts the integer coding strategy; S7.2, generating an initial population based on the constraints in the mobility efficiency constraint model, the mobility threat constraint model, and the group fire constraint model; S7.3, use the objective function to calculate the fitness value of each individual in the current population: S7.4, sort the individuals in the current population according to the fitness value, select the individuals with the specified proportion to form the elite population, and the rest of the individuals are the ordinary population; S7.5, the elite population is subjected to adaptive simulated annealing to generate new individuals, wherein the adaptive simulated annealing to generate new individuals refers to randomly selecting new individuals in the neighborhood of the current individual and accepting them according to a specified probability; the individuals in the ordinary population are subjected to adaptive probability. Select individuals from the elite population for crossover or self-crossover, where N iter Indicates the current iteration number, N epoch Represents the total number of iterations; the new individuals obtained by crossover or self-crossover between the common population and the individuals in the elite population are selected according to the adaptive mutation probability Implement mutation operation, and when implementing mutation operation, the number of mutation points is random, so that different individuals in the same population have different mutation points; only the unit layer and the action route layer participate in mutation; select a specified number of viable offspring from the current population and new individuals as the new current population; S7.6, determine whether the number of iterations reaches the preset threshold. If it has not reached the preset threshold, jump to step S7.4 to continue iteration; otherwise, use the objective function to calculate the fitness value of the new individuals in the new current population, and select the individual with the best fitness value as the final optimal task allocation and path planning solution.

[0059] See also Figure 5, this embodiment adopts a polygene encoding method, and the designed multi-layer polygene encoding strategy mainly includes a task layer, a unit layer, and an action route layer. The task layer is mainly composed of task numbers, and an integer encoding strategy is adopted, which can be expanded according to the actual number of tasks; the unit layer is composed of the numbers of our units, and a vacant gene mechanism is introduced, that is, some genes are randomly set to "-1", indicating that the task is not assigned to any unit; the action route layer also adopts an integer encoding strategy, which can be expanded according to the number of selectable action routes. After completing the unit gene encoding, the population generated in each round of iteration is sorted according to the fitness value, and the top 1 / 5 of the individuals are selected to form an elite population, and the remaining individuals are ordinary populations. First, a crossover operation is performed on the population. This embodiment uses an adaptive crossover operation. The individuals in the ordinary population choose to cross or self-cross with the individuals in the elite population according to the adaptive probability α, and the calculation method is: , in, N iter Indicates the current iteration number of the algorithm. N epoch Represents the total number of iterations of the algorithm. The law of change of this function is consistent with the law of elite population participation in crossover. In the early stage of algorithm iteration, in order to prevent the algorithm from converging too quickly, the probability of elite individuals participating in crossover is small, which can ensure that the population evolves in a better direction while maintaining the diversity of population structure. In the later stage of algorithm iteration, the population structure is relatively stable, and the probability of crossover operation producing a better solution is low. At this time, increasing the probability of elite individuals participating in crossover can accelerate the convergence of the algorithm while searching the solution space, saving computing resources. On this basis, mutation operations are performed on some individuals in the population. The mutation operation mainly follows the following two principles: (1) The number of mutation points is random, that is, different individuals in the same population have different mutation points. (2) Only the unit layer and action route layer participate in mutation. The reason is that if the target layer participates in mutation, some target genes will be lost, resulting in a large number of invalid offspring individuals.

[0060] In order to maintain a balance between enriching population diversity and avoiding excessive population differences, the EA-GA algorithm designs an adaptive mutation probability calculation method, as shown in the following formula: , in δ represents the probability of individual variation, N iter Indicates the current iteration number of the algorithm. N epochIndicates the total number of iterations of the algorithm. The calculation process is as follows: In the early stages of the algorithm iteration, in order to ensure the diversity of the offspring population and avoid premature convergence of the algorithm, the initial mutation probability of the parent population is set to 40%, and decreases as the algorithm iterates. In the later stages of the algorithm iteration, the probability of being able to search for a better solution decreases, and at this time, the mutation probability is gradually reduced to 14.7%, which speeds up convergence while meeting the optimization needs of the offspring population.

[0061] The following describes the specific implementation of the method proposed in this embodiment in conjunction with two specific city scenarios (represented as City 1 and City 2, respectively). First, the following can be directly obtained from the relevant GIS software: Figure 6 The original Shapefile image of City 1 is shown on the left. Next, the cv2.cvtColor function of the OpenCV library is used to convert the Shapefile image into a grayscale image. Then, the cv2.adaptiveThreshold function is used to perform an adaptive thresholding operation on the grayscale image to convert the grayscale image into the following Figure 6 The binary image shown on the right. Figure 7 After obtaining the binary image of City 1 shown on the left, the cv2.findContours function is further used to extract the polygon contour lines in the binary image to obtain the road network and building information of the city. The polygon contour line extraction result of City 1 is shown in the figure below. Figure 7 As shown on the right. Figure 8 After extracting the polygonal contour of the image of City 1 shown on the left, the shapely.geometry.Polygon function of the Shapely library is used to simplify the existing polygonal contour, and then the polygon.exterior.coords function is used to extract the vertices of the polygonal contour. These vertices are the key points of the path. The extraction results of the polygonal vertices (path key points) of City 1 are shown in Figure 1. Figure 8 As shown on the right. Figure 8 The key points of the path of City 1 shown on the right are combined with Figure 2 The connectivity judgment method between the key points of the path and Figure 3 The basic idea of ​​parallel computing is to use GPU and CPU to perform parallel computing on the connectivity between all key points of the path in City 1. Based on the obtained connectivity judgment results, the A-star algorithm is used to solve the traversable path between any two points in City 1. The solution example is as follows Fig. 9 As shown. Fig.10As shown, according to the confrontation assumption, the position distribution of our 8 units in City 1 is represented by red and green circular signs, among which the 1st, 2nd, 4th and 7th red signs are armored units, the 5th and 6th red signs are command units, and the 0th and 3rd green signs are logistics units and support units, respectively. The position distribution of the opponent's 4 units in City 1 is represented by blue circular signs, and the 0th, 1st, 2nd and 3rd blue signs are all armored units. Through the task decomposition and path planning method proposed in this embodiment, grouping strategies and maneuvering strategies are generated for our 8 units to suppress the opponent's 4 units with firepower. In addition, according to the maneuvering efficiency constraint model, the maneuvering threat constraint model and the grouping firepower constraint model, it is necessary to minimize the maneuvering threat of our units and maximize the maneuvering efficiency and remaining firepower. The objective function constructed in this embodiment is: , in, Represent the weights of mobility efficiency, residual firepower and mobility threat respectively. When using Figure 4 The improved multi-gene adaptive genetic-simulated annealing algorithm is used to solve the objective function, and the generated grouping and maneuvering strategies are as follows Fig.10 As shown in the figure. Our first unit forms a separate group and attacks the opponent's 0th unit along the planned route; our 7th unit forms a separate group and attacks the opponent's 3rd unit along the planned route; our 0th, 2nd, 3rd, and 6th units form a group and attack the opponent's 1st unit along their respective planned routes; our 5th unit forms a group and attacks the opponent's 2nd unit along the planned route. The results in the figure show that when When our units are in a state of emergency, they will focus on the advantages of group firepower during the attack, pay less attention to mobility efficiency and mobility threat factors, and emphasize the use of troops several times larger than the enemy in sufficient time to steadily eliminate the enemy. When , the generated formation and maneuver strategy is as follows Fig.11 As shown. Our first unit forms a separate group and attacks the opponent's third unit along the planned route; our seventh unit forms a separate group and attacks the opponent's zero unit along the planned route; our 0th, 2nd, 3rd, and 6th units form a group and attack the opponent's first unit along their respective planned routes; our fifth unit forms a separate group and attacks the opponent's second unit along the planned route. The results in the figure show that our units no longer consider the factors of mobile threats during the offensive process. While focusing on the advantages of grouped firepower, they also emphasize the unit's mobile efficiency, emphasizing that at all costs of threats, the opponent should be steadily eliminated within a limited time using troops that are several times larger than the opponent. On the other hand, the grouped firepower of our units needs to take into account factors such as the grouping situation and offensive capabilities of our units. In order to limit the grouped firepower of our units within a reasonable range, the following constraints need to be considered: , When the weight of group firepower constraints When , the generated formation and maneuver strategy is as follows Fig.12 As shown in the figure. Our 6th unit forms a separate group and attacks the opponent's 2nd unit along the planned route; our 5th unit forms a separate group and attacks the opponent's 3rd unit along the planned route; our 3rd and 7th units form a group and attack the opponent's 1st unit along their respective planned routes; our 0th unit forms a separate group and attacks the opponent's 0th unit along the planned route, while our 1st, 2nd, and 4th units do not participate in this attack. The results in the figure show that the ratio (constraint) relationship between our group firepower and the opponent's firepower can significantly affect the results of the formation and maneuver strategy. The smaller the ratio, it means that we hope to use as few group firepower resources as possible to achieve the goal of destroying the opponent's units.

[0062] When the weight of group firepower constraints When , the generated formation and maneuver strategy is as follows Fig.13 As shown in the figure. Our 5th unit forms a separate group and attacks the opponent's 3rd unit along the planned route; our 6th unit forms a separate group and attacks the opponent's 2nd unit along the planned route; our 0th and 3rd units form a group and attack the opponent's 0th unit along their respective planned routes; our 7th unit forms a separate group and attacks the opponent's 1st unit along the planned route. The results in the figure show that the higher the lower limit of the ratio between our group firepower and the opponent's firepower, the more it means that we hope to use sufficient group firepower resources to achieve the goal of destroying the opponent's units.

[0063] This embodiment is also Fig.14 The Unreal Engine-based city 2 shown in the figure conducted a task allocation and path planning experiment, and similar conclusions were drawn as in city 1, further verifying the effectiveness and rationality of the method proposed in this embodiment. Fig.15 , Fig.16 , Fig.17 and Fig.18The influence of the weights of the maneuver efficiency, remaining firepower and maneuver threat of the objective function proposed in this embodiment on the task allocation and path planning results of our units is further verified. The results in the figure show that by controlling the weights between the first, second and third optimization objectives in the objective function, the method proposed in this embodiment can generate different grouping and maneuver strategies. Among them, the first optimization objective is related to the maneuver efficiency. The greater the weight, the more the obtained grouping and maneuver strategy focuses on the rapidity and suddenness of the maneuver, that is, to quickly eliminate the opponent with limited manpower; the second optimization objective is related to the remaining grouping firepower. The greater the weight, the more the obtained grouping and maneuver strategy focuses on the advantage of offensive firepower, emphasizing the use of several times the opponent's manpower to steadily eliminate the opponent in sufficient time; the third optimization objective is related to the maneuver threat. The greater the weight, the more the obtained grouping and maneuver strategy focuses on reducing the possible threats in the maneuver process, thereby reducing the loss of our units. Fig.19 , Fig. 20 and Fig.21The influence of the weight of the grouping firepower constraint proposed in this embodiment on the task allocation and path planning results of our units is further verified. The results in the figure show that by controlling the upper and lower limits of the ratio between our grouping firepower and the opponent's firepower, the method proposed in this embodiment can generate different grouping and maneuvering strategies. Among them, the larger the upper limit of the ratio between our grouping firepower and the opponent's firepower, the more the grouping and maneuvering strategies obtained focus on using sufficient firepower to eliminate the opponent, and less consideration of resource consumption; the smaller the lower limit of the ratio between our grouping firepower and the opponent's firepower, the more the grouping and maneuvering strategies obtained focus on using less firepower resources to eliminate the opponent, and minimize resource consumption. It can be seen that in urban confrontation, it is very important to reasonably and efficiently allocate tasks and plan paths for our units according to the confrontation goals, confrontation plans and real-time comprehensive situation of confrontation. The method of this embodiment proposes a task allocation and path planning method for urban confrontation, which is used to analyze and predict the grouping strategy and maneuvering strategy of multiple units of our side. First, the Shapefile image of a specific confrontation area is analyzed using the computer vision and image processing open source library OpenCV and the spatial geometry calculation open source library Shapely to obtain urban road network information and path key points. Secondly, based on the real-time comprehensive confrontation situation of the confrontation between the units of the two sides, such as unit position, unit type, unit basic attributes and path key points, the A-star algorithm is used to search for multiple passable paths between the units of the two sides, and the distance, congestion degree, threat degree and other key confrontation elements of all passable paths are calculated respectively. Then, a maneuver efficiency constraint model, a maneuver threat constraint model and a group fire constraint model are constructed, and the objective function is designed to optimize the grouping strategy and maneuver strategy. Finally, the improved multi-gene adaptive genetic-simulated annealing algorithm is used to solve the objective function and generate a reasonable task allocation and path planning scheme. The method of this embodiment realizes task allocation and path planning for urban confrontation, and improves the applicability and accuracy of task allocation and path planning in different urban confrontation scenarios.

[0064] This embodiment also provides a task allocation and path planning system for urban confrontation, including a microprocessor and a memory connected to each other, and the microprocessor is programmed or configured to execute the task allocation and path planning method for urban confrontation. This embodiment also provides a computer-readable storage medium, which stores a computer program or instruction, and the computer program or instruction is programmed or configured to execute the task allocation and path planning method for urban confrontation through a processor. This embodiment also provides a computer program product, including a computer program or instruction, which is programmed or configured to execute the task allocation and path planning method for urban confrontation through a processor.

[0065] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A task allocation and path planning method for urban confrontation, characterized in that: The steps include: S1, obtaining real-time comprehensive situation information of the urban confrontation scene including the positions of each unit of the confrontation parties; S2, obtain the shapefile image of the current confrontation area from the GIS software; S3, analyzes the shapefile image and extracts key points of paths in the urban road network; S4, calculating the traversable paths between the units of the two opposing sides according to the positions of the units of the two opposing sides; S5, combining the real-time comprehensive situation information of the urban confrontation scene and the traversable paths between the units of the two opposing sides, calculating all the key confrontation elements about the traversable paths and establishing a mobility efficiency constraint model, a mobility threat constraint model and a formation fire constraint model, wherein the key confrontation elements include unit type, unit basic attributes, day and night environment and path conditions; S6, constructing the objective function based on the mobility efficiency constraint model, mobility threat constraint model and group fire constraint model; S7, solve the task allocation and path planning solutions based on the constructed objective function.

2. The method for task allocation and path planning for urban confrontation according to claim 1 is characterized in that: Step S3 includes: S3.1, binarize the shapefile image to obtain a binary image; S3.2, extract polygon contours in the binary image to obtain the city’s road network and building information; S3.3, simplify the polygonal contour line and extract the vertices of the polygonal contour line as the path key points in the urban road network.

3. The task allocation and path planning method for urban confrontation according to claim 1 is characterized in that: Step S4 includes: S4.1, use GPU or CPU parallel computing to analyze the connectivity between key points on the path: for any two key points on the path, calculate the difference between the ordinate and the abscissa, and take the maximum value of the two as the number of sampling points ; Uniformly sample the line segment connecting two path key points points, and round the coordinates of each point, and finally determine the sampled Whether there is a black point among the points, if there is a black point, it is determined that the two key points on the path are not connected; otherwise, it is determined that the two key points on the path are connected; S4.2, judging whether the position coordinates of each unit of the opposing parties are in the building according to the pixel values ​​of the positions of each unit of the opposing parties in the shapefile image, if the pixel value of the position coordinates in the shapefile image is black, it is determined that the unit is in the building, otherwise, it is determined that the unit is not in the building; S4.3, for the unit in the building, the position coordinate point of the unit is considered to be connected with the key point of the path corresponding to the building; for the unit not in the building, the connectivity of the position coordinate point of the unit with each key point of the path is determined one by one; finally, the connectivity of the units of the two adversaries with all the key points of the path is obtained; S4.4, based on the connectivity between the units of the two adversaries and all key points of the path, use the specified path planning algorithm to calculate the traversable path between the units of the two adversaries.

4. The method for task allocation and path planning for urban confrontation according to claim 1 is characterized in that: The function expression of the maneuvering efficiency constraint model established in step S5 is: ， ， in, We will organize our team to go to The maneuver time of the opponent's unit; Boolean variable Indicates our Whether the unit decides from Passable paths to attack the opponent units, if so, then ,otherwise ,in , Indicates the number of our units, , represents the number of opponent units, , Indicates our Units reach the opponent's The total number of traversable paths per unit; Indicates our Units from Passable paths to attack the opponent The time required for each unit; Indicates the unit that attacks the opponent among all our units. The number of support units per unit. The more support units there are, the higher the mobility efficiency. It represents the gain of the support unit on maneuver efficiency; The function expression of the maneuver threat constraint model established in step S5 is: ， In the above formula, For our unit to go to The mobile threat of an opponent's unit, For our The importance value of each unit; Indicates our Units attack the opponent Is there a feasible path when there are units? If there is a feasible path, then ,otherwise, ; Indicates our Units from Passable paths to attack the opponent When there are units, the distance that can be seen by the opponent's unit; Indicates our Units from Passable paths to attack the opponent The comprehensive hazard level of the path when the unit is For our The defensive capability value of each unit; The function expression of the group firepower constraint model established in step S5 is: ， ， in, Indicates that our team's firepower is in conflict with the opponent's team's The minimum ratio of firepower per unit; Indicates that our team's firepower is in conflict with the opponent's team's The maximum ratio of the firepower of each unit, For the opponent The offensive firepower value of a unit; Boolean variable Indicates our Whether the unit decides from Passable paths to attack the opponent units, if so, then ,otherwise ; For our The offensive firepower value of each unit; The remaining firepower value for all our formations, It indicates the increase in the offensive firepower of our units from the logistics units. Indicates the unit that attacks the opponent among all our units. The number of logistic units per unit, Indicates the gain of the command unit's offensive firepower to our unit, a Boolean variable Indicates attacking the opponent Can our units maintain communication with the command unit? If so, ,otherwise ; Indicates the unit that attacks the opponent among all our units. The number of command units of a unit, For the opponent The offensive firepower value of each unit; The function expression of the objective function constructed in step S6 is: ， in, They represent the weights of mobility efficiency, residual firepower and mobility threat respectively.

5. The method for task allocation and path planning for urban confrontation according to claim 4 is characterized in that: Our Units from Passable paths to attack the opponent The calculation function expression of the comprehensive danger degree of the path when the unit is: ， in, Indicates our Units from Passable paths to attack the opponent The comprehensive hazard level of the path when the unit is ~ Represent the weights of different factors. Indicates bias, Indicates our Units from Passable paths to attack the opponent The number of traffic intersections or crossroads on the path when the unit is Indicates our Units from Passable paths to attack the opponent The number of bridges on the path when the unit is Indicates our Units from Passable paths to attack the opponent The number of tunnels on the path when the unit is Indicates our Units from Passable paths to attack the opponent The ratio of the route length covered by the opponent's firepower to the total route length when the opponent fires a unit.

6. The method for task allocation and path planning for urban confrontation according to claim 4 is characterized in that: Our Units from Passable paths to attack the opponent The calculation function expression of the time required for each unit is: ， in, Indicates our Units from Passable paths to attack the opponent The time required for each unit; Indicates that our unit is from Passable paths to attack the opponent The distance of the path when the unit is Indicates our Units attack the opponent Is there a feasible path when there are units? If there is a feasible path, then ,otherwise, ; Indicates our Units from Passable paths to attack the opponent The congestion level of the path when the unit is For our The speed of the unit, when the confrontation takes place during the day, our The speed of the unit is ,in, Indicates our The maximum maneuvering speed of a unit during the day; when the confrontation takes place at night, our The speed of the unit is ,in Indicates our The maximum maneuvering speed of a unit at night.

7. The method for task allocation and path planning for urban confrontation according to claim 1 is characterized in that: In step S7, the task allocation and path planning scheme for solving the constructed objective function is to use an improved multi-gene adaptive genetic-simulated annealing algorithm to solve the task allocation and path planning scheme for the constructed objective function, including: S7.1, a multi-layer polygene coding strategy is used to solve the problem of task allocation and path planning. The multi-layer polygene coding strategy includes a task layer, a unit layer and an action route layer. The task layer is mainly composed of task numbers and adopts an integer coding strategy; the unit layer is composed of the numbers of our units, and a vacant gene mechanism is introduced to randomly set some genes to "-1" to indicate that the task is not assigned to any unit; the action route layer adopts an integer coding strategy; S7.2, generating an initial population based on the constraints in the mobility efficiency constraint model, the mobility threat constraint model, and the group fire constraint model; S7.3, use the objective function to calculate the fitness value of each individual in the current population: S7.4, sort the individuals in the current population according to the fitness value, select the individuals with the specified proportion to form the elite population, and the rest of the individuals are the ordinary population; S7.5, the elite population is subjected to adaptive simulated annealing to generate new individuals, wherein the adaptive simulated annealing to generate new individuals refers to randomly selecting new individuals in the neighborhood of the current individual and accepting them according to a specified probability; the individuals in the ordinary population are subjected to adaptive probability. Select individuals from the elite population for crossover or self-crossover, where N iter Indicates the current iteration number, N epoch Represents the total number of iterations; the new individuals obtained by crossover or self-crossover between the common population and the individuals in the elite population are selected according to the adaptive mutation probability Implement mutation operation, and when implementing mutation operation, the number of mutation points is random, so that different individuals in the same population have different mutation points; only the unit layer and the action route layer participate in mutation; select a specified number of viable offspring from the current population and new individuals as the new current population; S7.6, determine whether the number of iterations reaches the preset threshold. If it has not reached the preset threshold, jump to step S7.4 to continue iteration; otherwise, use the objective function to calculate the fitness value of the new individuals in the new current population, and select the individual with the best fitness value as the final optimal task allocation and path planning solution.

8. A task allocation and path planning system for urban confrontation, comprising a microprocessor and a memory connected to each other, characterized in that: The microprocessor is programmed or configured to execute the task allocation and path planning method for urban confrontation as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program or instruction stored therein, characterized in that: The computer program or instruction is programmed or configured to execute the task allocation and path planning method for urban confrontation described in any one of claims 1 to 7 through a processor.

10. A computer program product comprising a computer program or instructions, characterized in that The computer program or instruction is programmed or configured to execute the task allocation and path planning method for urban confrontation described in any one of claims 1 to 7 through a processor.

Images