Unmanned aerial vehicle cooperative attack rapid target allocation method based on case reasoning

Abstract

The invention provides an unmanned aerial vehicle cooperative attack rapid target allocation method based on case reasoning, relates to the technical field of self-explosion unmanned aerial vehicle cluster cooperative combat decision making, and designs a rapid target allocation algorithm model with a switchable mechanism. According to the algorithm, optimization algorithm model solution of a cooperative combat task can be realized, an algorithm mechanism is converted under different combat task requirements, the solution efficiency of the algorithm is improved, the optimality of understanding is guaranteed, and the algorithm has excellent adaptability to a target allocation task of large-scale high-intensity strike. The problem that the resolving time of a traditional distribution algorithm is long and the quality of the solution is continuously reduced along with the increase of the number of weapons of both combat parties can be solved to a certain extent.

Description

Technical Field

[0001] The invention relates to the technical field of self-destructing unmanned aerial vehicle (UAV) cluster coordinated combat decision-making technology, and in particular to a UAV coordinated attack rapid target allocation method based on case reasoning. Background Art

[0002] The coordinated attack strategy of multiple self-destructing drones is a new type of weapon combat style that has gradually developed under the complex and ever-changing needs of military operations. In essence, it is a multi-dimensional variable nonlinear combinatorial optimization problem with multiple complexities and time sensitivity. The cluster combat of self-destructing drones has the unique advantage of integrated reconnaissance and strike, which can meet the needs of future informationized operations and is currently a commonly used combat method. As a new type of intelligent weapon and ammunition equipment, the self-destructing drone combines advanced artificial intelligence control technology with drone technology and is mostly used in large-scale cluster combat. At the same time, the coordinated combat mode also makes up for the lack of damage capability of a single self-destructing drone, and can cause more lethal attacks on enemy targets. In the context of highly informationized military warfare, the coordinated strike strategy of multiple missiles plays a key role in military combat missions.

[0003] Compared with single-unit combat, cluster combat can perform more complex combat tasks, such as single-target convergence attack, single-target time-sequence attack, and multi-target coordinated attack. It can improve the success rate of task completion, has stronger fault tolerance and robustness, and can still complete the task in the case of single damage or failure. At present, the combat of self-destructing drone clusters also faces many challenges. In addition to the performance constraints of the single unit, it is also necessary to consider the coordination constraints between self-destructing drones, such as spatial coordination constraints and temporal coordination constraints. At the same time, it also increases the difficulty of the planning problem, greatly increasing the complexity of the algorithm and the cost of computing time. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a method for rapid target allocation of cooperative attacks of unmanned aerial vehicles based on case reasoning. The present invention optimizes and improves the cooperative combat target allocation algorithm of self-destructing unmanned aerial vehicles, and designs a rapid target allocation algorithm model with a switchable mechanism. The algorithm can realize the optimization algorithm model solution of cooperative combat tasks, convert the algorithm mechanism under different combat mission requirements, improve the algorithm solution efficiency, ensure the optimality of the solution, and have excellent adaptability to the target allocation task of large-scale high-intensity strikes. To a certain extent, it can overcome the problems of long solution time of traditional allocation algorithms and the continuous decline of solution quality as the number of weapons of the combating parties increases.

[0005] A method for rapid target assignment of UAV coordinated attack based on case-based reasoning includes the following steps:

[0006] Step 1: Based on the requirements of cluster combat strike missions and battlefield environment, a unified modeling of the battlefield space environment is performed;

[0007] Assume that m self-destructing drones form a drone cluster to jointly perform strike missions, attacking n targets scattered at different locations on the battlefield, and there are t threat radius sets distributed in the battlefield, R = {R1, R2, R3...R t}, the coordinates are {(x t1 ,y t1 ),(x t2 ,y t2 )...(x tt ,y tt )} threat source, and ensure that each self-destructing drone performs at most one mission and strikes one target; where the initialization defines the self-destructing drone cluster M = {M1,M2,...,M m}, ground target set N = {N1, N2, ..., N n}, set the target initial position to (x i ,y j ), target N j The value of Q j ; In order to smoothly implement the strike mission, during the execution of the coordinated strike mission, the self-destructing drone needs to avoid the threat sources appearing on the battlefield and finally reach the target mission point to carry out the strike. And according to the target type, the upper limit of the number of strikes for each target is set to D jmax That is, the number of drones assigned to each target should be less than or equal to D jmax indivual;

[0008] Step 2: Design a two-dimensional battlefield simulation model using the grid method to describe the battlefield space in detail;

[0009] In the two-dimensional battlefield simulation model, each grid represents a coordinate point, which is used to indicate the position of the drone, target, and obstacle. The planning space is represented by a set Ω, where (x, y) represents the horizontal and vertical coordinates of a point in the space Ω. max and max They represent the maximum distance positions specified by the horizontal and vertical coordinates, respectively, as shown in the following formula:

[0010] Ω={(x,y)|0≤x≤x max ,0≤y≤y max}

[0011] Step 3: Model the battlefield threat source, select radar threat, anti-aircraft artillery threat and terrain threat as threat source parameters and calculate the threat source threat degree;

[0012] Step 3.1, calculate the radar threat level within the battlefield;

[0013] The radar echo signal power is expressed as:

[0014]

[0015] Where P t is the transmitting power, G is the transmitting antenna gain, σ is the radar scattering area, R is the distance from the target to the radar, in meters, and λ is the wavelength of the electromagnetic wave; for a certain UAV, the radar scattering area is constant, and the above formula is simplified to:

[0016]

[0017] Where K is a constant. When the radar is not interfered with, the probability of detecting a target is related to the quality of the received signal. Assuming that the radar can perform omnidirectional detection, the maximum detection radius of the radar is R. max , then the detection probability of the radar to the UAV is expressed as:

[0018]

[0019] Step 3.2, calculate the threat level of anti-aircraft artillery;

[0020] When the enemy radar detects the drone, the air defense missile will attack according to the target information. The kill probability of the air defense missile is:

[0021] P m =T·Y·Z[1-(1-A·E·W) Q ]

[0022] Where T is the radar detection probability of the surface-to-air missile, Y is the radar interception guidance probability, Z is the information transmission conversion probability, A is the missile successful launch probability, E is the missile flight reliability probability, W is the missile kill probability, and Q is the number of missiles launched at one time.

[0023] Step 3.3, calculate the terrain threat level;

[0024] Calculate the height threat and size of the terrain:

[0025]

[0026] Where, T h Indicates the threat level of terrain altitude, h m Indicates the altitude of the current terrain, h u Indicates the upper limit of the drone's flight altitude. If the terrain altitude exceeds the upper limit of the drone's low-altitude flight, the drone's terrain threat is set to the maximum value;

[0027] When the real-time flight altitude of the UAV is h real When , the threat of obstacles at different altitudes to the drone is expressed as:

[0028]

[0029] Step 4: Establish a reasonable optimization target allocation model and define the target optimization function according to the optimization model constraints;

[0030] Step 4.1, calculate the voyage cost;

[0031] Step 4.1.1, calculate the track distance in each track segment;

[0032] l MS To represent the track length of the Mth UAV on the kth track segment, S is the number of track segments:

[0033]

[0034] In the formula, Represents the starting point coordinates of the Mth UAV on the kth track segment.

[0035] Step 4.1.2, calculate the range cost of each strike plan;

[0036] The attack plan is: if the first drone among m drones attacks the first target among n targets, it is one attack plan; if the first drone among m drones attacks the second target among n targets, it is another attack plan; if it attacks the nth target among n targets, it is the nth attack plan; each drone has the same number of attack plans, and in each attack plan, a drone only attacks one target.

[0037] In the process of trajectory planning, a trajectory is divided into S segments. The trajectory distance L of the Mth UAV attacking the Nth target MN The formula is:

[0038]

[0039] Step 4.1.3, normalize the voyage cost value;

[0040] The flight cost F after unification to the same order of magnitude l The formula is:

[0041]

[0042] Where, L MN It is represented as the track distance of the Mth UAV attacking the Nth target, L Nmax It represents the maximum flight distance of all drones when attacking. The ratio of the two can be used to calculate the range cost of each attack plan.

[0043] Step 4.2, calculate the threat cost in the battlefield;

[0044] The threat cost is inversely proportional to the distance from the threat source to the track. The entire track is divided into S track segments, and the threat cost value of each track segment is calculated. The overall threat cost is the sum of the threat costs of the S track segments.

[0045] Step 4.2.1, calculate the threat cost generated in each track segment;

[0046] Assume that there is an Mth drone attacking the Nth target in the battlefield. First, each track is divided into S track segments. Two adjacent track points of the drone are represented by L k and L k+1 , threat source coordinates are expressed as In order to obtain the threat cost of each track segment, each track segment is divided into C equal parts, and the threat cost is indirectly calculated by obtaining the distance from the threat source to the C track points, which is specifically expressed as:

[0047]

[0048] Where N Th is the number of threat sources, C is the number of equal parts of each track, is the coordinate of each track point, t=1,2,…,N Th , c=1,2,…,C;

[0049] Step 4.2.2, calculate the overall threat cost in each track;

[0050] Sum the threat costs of S track segments to get the total threat cost F in the track t for:

[0051]

[0052] Step 4.2.3: Normalize the threat cost and calculate it on a unified scale;

[0053] Normalize the threat cost to a unified scale, the threat cost F t The formula is:

[0054]

[0055] In the formula, F t MN F represents the threat cost of the Mth UAV attacking the Nth target. tmax It represents the track with the largest total threat cost among all tracks;

[0056] Step 4.3: Calculate the overall attack benefit;

[0057] Step 4.3.1, calculate the overall attack benefit in each attack plan;

[0058] The target residual value is used to evaluate the target attack benefit cost. The attack benefit generated by the Mth drone attacking the Nth target is expressed as:

[0059] F v =V N ·P MN ·x ij

[0060] Where V N is the target value of the Nth target, P MN is the probability of damage when the Mth UAV attacks the Nth target, x ij is the decision variable;

[0061] Step 4.3.2: Normalize the attack revenue;

[0062] Attack gain F v Unified to the same order of magnitude, expressed as:

[0063]

[0064] In the formula, represents the attack benefit of the Mth drone attacking the Nth target, F vmax Indicates the maximum attack benefit that can be obtained among all attack plans.

[0065] Step 4.4, calculate the enemy target value;

[0066] Combined with the different types of weapons and their attack ranges, different threat levels are set for the target from 0 to 1. typ , which is used to clearly evaluate and distinguish the degree of strike value of each target in the battlefield environment.

[0067] Step 4.5: Analyze the constraints generated in the target allocation;

[0068] The constraints are as follows:

[0069] (1) Range constraints

[0070] The distance traversed by each self-destructing drone should be less than its range, expressed as:

[0071]

[0072] In the formula, It is the maximum range of a self-destructing drone. is the actual crossing distance of the self-destructing drone,

[0073] N v It is the maximum flight altitude constraint.

[0074] (2) Decision variable constraints

[0075]

[0076] In the formula, the decision variable x ij is a 0-1 variable. When drone i acts on target j, x ij =1 means to carry out the attack, x ij = 0 means no strike was carried out. The decision variable is represented as x ij ∈(0,1),i=1,...,M,j=1,...,N, Indicates that a task is performed by at least one drone once, and the number of drones assigned does not exceed D jmax shelf; This means that a drone can only perform one mission at most;

[0077] Step 4.6, according to the target allocation optimization index, construct the objective function and describe each cost function;

[0078] The target allocation optimization index requires that the attack cost of the drone be minimized and the attack benefit be maximized.

[0079] The optimization objective function of multi-UAV collaborative target allocation is expressed as:

[0080] F=λ1F l +λ2F t -λ3F v

[0081] In the formula, λ = (λ1, λ2, λ3) is the weight vector in the optimization objective function, which indicates the different influence of each factor on the target allocation result, and satisfies λ i ∈(0,1); F l is the voyage cost, F t is the threat cost, F v For the attack benefit, when the range cost and threat cost are minimized and the attack benefit is maximized, the final optimal solution for target allocation is output.

[0082] Step 5: Use the improved genetic algorithm based on the adaptive penalty function idea and the greedy algorithm switching mechanism to perform target allocation, and perform algorithm switching operations on the two algorithms in different combat environments. The genetic algorithm based on the penalty function idea is improved on the traditional genetic algorithm to solve the objective function optimization problem with multiple sets of constraints;

[0083] Step 5.1, initialize the battlefield environment, set the number of self-destruct drones, attack targets and threat sources;

[0084] Step 5.2, set the initial population size, number of iterations, selection rate, crossover rate and mutation rate;

[0085] Step 5.3, setting the encoding operation mode for the initial population;

[0086] Step 5.3.1, use binary coding to perform encoding operations, with the correspondence between self-destructing drones and targets, that is, each possible allocation scheme represents a chromosome, and each gene in the chromosome represents whether a specific drone is assigned to a target;

[0087] Step 5.3.2, encode the m self-destructing drones and n targets in the battlefield environment into binary codes, using 1 to represent that the self-destructing drone attacks the target, and 0 to represent that it does not attack the target. Then each chromosome has m×n numbers, i.e., “genes”;

[0088] Step 5.3.3: Group every n genes into a group, and arrange each group of genes vertically to obtain the multi-self-destructing drone target allocation matrix C m×n ;

[0089] Step 5.3.4: The i-th row of the matrix represents the self-destructing drone, i = 1, 2, 3, ..., m, and the j-th column of the matrix represents the target, c ij is the attack situation of the self-destructing UAV on the target, if c ij =0 means self-destructive drone M i For target N j No attack, if c ij =1 means self-destructive drone M i Number to target N j to strike;

[0090] Step 5.4, initialize the population and generate N individuals;

[0091] Step 5.5, determining the constraint conditions of the self-destructing UAV according to step 4.5;

[0092] Step 5.6: Construct a penalty function. For individuals that violate the constraints, subtract the penalty function when calculating the fitness value to reduce the fitness value.

[0093] Step 5.6.1. Use the two constraints in step 4.5 as penalty terms in the penalty function, and select the penalty factor, that is, the weight coefficient; construct the penalty function p(x)=f(x)+h(x)G(x), where f(x) is the objective function value, h(x) is the penalty factor, and G(x) is the penalty term.

[0094] Step 5.6.2: Substitute the constraints into the penalty function, which is expressed as p(x)=f(x)+η(n+m). According to the cumulative penalty principle, the penalty for individuals who violate more overall constraints is greater; where 0<η≤1, and can be adaptively adjusted according to the number of feasible solutions.

[0095] Step 5.7: Construct the fitness function, the formula is as follows: Fitness(x) = C max -p(x)

[0096]

[0097] Among them, f(x) is the objective function value, C max is the maximum estimated value of the objective function, obtained by estimation.

[0098] Step 5.8: Select, crossover and mutation operations are performed on the population;

[0099] After constructing the fitness function and penalty function, the population is subjected to genetic operations such as selection, crossover, and mutation, and iterated sequentially.

[0100] Step 5.8.1: According to the principle of survival of the fittest, the fitness of the individuals to the environment is measured by the value of the individual fitness function, and the individuals in the initial population are sorted;

[0101] Step 5.8.2: Use the roulette wheel method to select chromosomes in the population that meet the fitness standard value, and then use the elite selection and retention strategy to select and retain excellent individuals in the sorting results;

[0102] Step 5.8.3: Perform crossover and mutation operations on the retained excellent individuals to generate offspring populations, and combine the two populations to form a new offspring population R of size 2N. t .

[0103] Step 5.9, generating an adaptive fitness function value;

[0104] Assume that the number of feasible solutions after each iteration is P1, the total number is N, and the penalty factor That is, the weight of the penalty item changes in direct proportion to the increase or decrease of feasible solutions. When there are too many feasible solutions, the penalty factor will also increase in the next iteration, otherwise it will decrease.

[0105] Step 6: When the genetic algorithm with the improved adaptive penalty function idea cannot meet the combat requirements, the allocation algorithm uses the greedy algorithm to continue solving the algorithm.

[0106] Step 6.1, parameter initialization: It is known that there are n targets as input for solving the problem, and the solution set J of the problem is set to empty; the parameters of the maximum allowed number of iterations of the greedy algorithm, the end mark threshold and the number of targets are initialized.

[0107] Step 6.2, define the objective function: According to the problem requirements, randomly generate task pairs of self-destructing drones and attack targets, and calculate the objective evaluation function. Select a metric and sort the n inputs according to this metric;

[0108] Step 6.3, traverse all the attack plans of m self-destructing drones and n targets, and calculate the corresponding cost value through the cost function formula;

[0109] Step 6.4: According to the absolute value of the objective function, establish the cost matrix, where A is defined as 11 ,A 12 ,A 13 ...A 1m The value of is the cost value of the attack plan consisting of the target 1 and M drones. The remaining row elements A mn They all represent the attack cost of the Mth drone attacking the Nth target, and finally form a cost matrix containing N×M cost value elements.

[0110]

[0111] Step 6.4.1. Use the comparison method to find the largest element in the cost matrix A and record its position.

[0112] Step 6.4.2: Assume that the largest element in the cost matrix A is A 00 , it means that after comprehensive consideration of the target cost and track cost, the attack benefit of our drone on the enemy's N0 target by M0 is the greatest, so we implement the M0-to-N0 attack plan;

[0113] Step 6.4.3: Since N0 has become the target of M0, 00 The row and column are set to ∞;

[0114] Step 6.4.4: A 00 After the row and column are set to ∞, A 00 No longer participate in subsequent assignment tasks, then return to step 6.3 to continue the loop iteration, looking for the next optimal solution to update the status;

[0115] Step 6.4.5, after completing all one-to-one assignments, restore the original cost matrix structure, set all assigned solutions to ∞ to avoid searching again, and re-search the many-to-one assignment results for the remaining elements in the matrix;

[0116] Step 6.5, output the optimal solution: when all n inputs have been searched, if the task conditions are met, the search stops, that is, the current solution set is the global optimal solution, and the search result is output;

[0117] Step 7: Set the main switching mechanism of the target allocation algorithm;

[0118] Step 7.1, initialize battlefield global variable information according to battlefield environment requirements;

[0119] Step 7.2: Define combat scenarios of different scales according to the different numbers and levels of combat weapons, and specify the completion time T for each target assignment task. c , where the number of our self-destruct drones participating in the battle is m, the number of enemy combat weapons is n, and it is stipulated that m>n for each battlefield scale. The following are the three defined combat scales, and subsequent actual battlefields are also classified according to this definition;

[0120] If the number of our drone targets m≤20 and the number of enemy targets n≤20, it is a small-scale conflict;

[0121] The number of our drone targets is 20<m≤50 and the number of enemy targets is 20<m≤50, which is a medium-scale operation;

[0122] The number of our drone targets is 50<m≤100 and the number of enemy targets is 50<m≤100, which is a large-scale battle;

[0123] Other situations are not taken into account.

[0124] Step 7.3: According to the nature of the task, the planning time and the requirements for optimizing performance, clarify the goals and conditions of the switching algorithm, and clarify the completion time T for each target assignment task. f .

[0125] Step 7.4, establish a mission planning model database. Conduct comparative tests of the two algorithms at different combat scales, and continuously increase the experimental scale to obtain multiple groups of calculation time and objective function change values ​​under different combat situations. Store the experimental results to form a database and establish a historical target allocation model;

[0126] Step 7.5: Enter actual battlefield combat parameters into the system.

[0127] Step 7.6: Identify the characteristic indicators of the actual battlefield situation and make similarity judgments, including the number of weapons in combat, the scale of operations, and the time T for the command to be issued. f As the main feature judgment indicator;

[0128] Step 7.7: Develop a switching strategy;

[0129] Step 7.7.1. Compare the case data including calculation time and target function change value in the existing historical database with the actual combat conditions, search in the database, and select the target allocation model with the highest similarity as the matching object.

[0130] Step 7.7.2: Allocate time T according to the target specified in the mission c As a benchmark, complete the similarity judgment through step 7.6 and select the case with the same scale as the actual battlefield and the highest similarity;

[0131] Step 7.7.3, extract the most similar case, compare the optimization performance of the genetic algorithm and the greedy algorithm within the benchmark time range, and compare the algorithm decision results. According to the comparison results, due to the high similarity, match the battlefield environment in the database with the actual combat scenario, and the algorithm decision in this case is fed back as the optimal decision result under the current situation.

[0132] Step 7.8, monitor and evaluate the results; if there are new changes in the battlefield situation, it is necessary to identify the characteristic indicators under the new situation again, return to step 7.6, and make new algorithm planning decision results to ensure the real-time and accuracy of target allocation.

[0133] The beneficial effects of adopting the above technical solution are:

[0134] The present invention provides a method for rapid target allocation of coordinated attacks of unmanned aerial vehicles based on case reasoning. Compared with other task planning algorithms, the optimization model algorithm provided by the present invention mainly solves the problem that the calculation time of target allocation in traditional combat mission scenarios is too long, and the result deviation increases with the complexity of the battlefield. At the same time, a large-scale combat mission scenario is set, and simulation is carried out in combination with a switchable control target allocation algorithm. The results show that the optimization algorithm model has excellent adaptability to the target allocation task of large-scale strike missions, and the optimality of the solution is also guaranteed to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0135] Figure 1 This is a flow chart of a method for rapid target allocation for cooperative attack by unmanned aerial vehicles in an embodiment of the present invention;

[0136] Figure 2 This is a schematic diagram of grid-based battlefield modeling in an embodiment of the present invention;

[0137] Figure 3 This is a flowchart of an improved genetic algorithm in an embodiment of the present invention;

[0138] Figure 4 This is a flow chart of a greedy algorithm in an embodiment of the present invention;

[0139] Figure 5This is a flow chart of the algorithm switching mechanism in an embodiment of the present invention;

[0140] Figure 6 This is a comparison chart of algorithm optimization performance in an embodiment of the present invention. DETAILED DESCRIPTION

[0141] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0142] A fast target assignment method for UAV coordinated attack based on case-based reasoning, such as Figure 1 As shown, the present invention mainly adopts a collaborative target allocation algorithm based on a switching mechanism to calculate a reasonable target allocation result, determines the optimization function through battlefield modeling, constraint condition analysis, cost function calculation and other links, and uses an improved adaptive penalty function genetic algorithm or a greedy algorithm to solve the solution in different combat scenarios. Specifically, it includes the following steps:

[0143] Step 1: Based on the requirements of cluster combat strike missions and battlefield environment, a unified modeling of the battlefield space environment is performed;

[0144] Assume that m self-destructing drones form a drone cluster to jointly perform strike missions, attacking n targets scattered at different locations on the battlefield, and there are t threat radius sets distributed in the battlefield, R = {R1, R2, R3...R t}, the coordinates are {(x t1 ,y t1 ),(x t2 ,y t2 )...(x tt ,y tt )}, and ensure that each self-destructing drone performs at most one mission and strikes one target; where the initialization defines the self-destructing drone cluster M = {M1,M2,...,M m}, ground target set N = {N1, N2, ..., N n}, set the target initial position to (x i ,y j ), target N j The value of Q j ; In order to smoothly implement the strike mission, during the execution of the coordinated strike mission, the self-destructing drone needs to avoid the threat sources appearing on the battlefield and finally reach the target mission point to carry out the strike. And according to the target type, the upper limit of the number of strikes for each target is set to D jmax That is, the number of drones assigned to each target should be less than or equal to D jmax indivual;

[0145] Step 2: Use the grid method to design a two-dimensional battlefield simulation model and describe the battlefield space in detail; Figure 2 The figure shows a schematic diagram of the grid battlefield modeling of the present invention. The two-dimensional simulated battlefield environment mainly includes the position of the UAV on the left, the position of the threat source in the battlefield and its threat range, which are represented by blank circles for simplicity, and key information such as the target position and attack range on the right.

[0146] Taking into account the actual needs of coordinated attack mission planning, the grid graph method, which is suitable for large-scale battlefield modeling and has good visualization performance, is used to model the battlefield space. In the two-dimensional battlefield simulation model, each grid represents a coordinate point, which is used to represent the position of drones, targets, and obstacles; the planning space is represented by a set Ω, (x, y) represents the horizontal and vertical coordinates of a position point in the space Ω, and x max and max They represent the maximum distance positions specified by the horizontal and vertical coordinates, respectively, as shown in the following formula:

[0147] Ω={(x,y)|0≤x≤x max ,0≤y≤y max}

[0148] Step 3: Model the battlefield threat source, select radar threat, anti-aircraft artillery threat and terrain threat as threat source parameters and calculate the threat source threat degree;

[0149] Step 3.1, calculate the radar threat level within the battlefield;

[0150] During the flight, the drone needs to take a series of measures to avoid the radar detection area, thereby reducing the risk of being detected by the radar. The radar echo signal power is expressed by the formula:

[0151]

[0152] Where P t is the transmitting power, G is the transmitting antenna gain, σ is the radar scattering area, R is the distance from the target to the radar, in meters, and λ is the wavelength of the electromagnetic wave; for a certain UAV, the radar scattering area is constant, and the above formula is simplified to:

[0153]

[0154] Where K is a constant. When the radar is not interfered with, the probability of detecting a target is related to the quality of the received signal. Assuming that the radar can perform omnidirectional detection, the maximum detection radius of the radar is R. max , then the detection probability of the radar to the UAV is expressed as:

[0155]

[0156] Step 3.2, calculate the threat level of anti-aircraft artillery;

[0157] When the enemy radar detects the drone, the air defense missile will attack according to the target information. It is an important firepower threat to low-altitude drones. The kill probability of the air defense missile is:

[0158] P m =T·Y·Z[1-(1-A·E·W) Q ]

[0159] Where T is the radar detection probability of the surface-to-air missile, Y is the radar interception guidance probability, Z is the information transmission conversion probability, A is the missile successful launch probability, E is the missile flight reliability probability, W is the missile kill probability, and Q is the number of missiles launched at one time.

[0160] Step 3.3, calculate the terrain threat level;

[0161] When the drone maintains a normal altitude, it will face the threat of obstacles such as mountains and slopes. When flying at a lower altitude, it will face the threat of high-altitude wires, chimneys, etc. There may be mountains of different heights in the planning area. During flight, the aircraft may collide with the ground due to insufficient flying altitude. The higher the altitude, the greater the threat to the drone. The height threat and size of the terrain are calculated:

[0162]

[0163] Where, T h Indicates the threat level of terrain altitude, h m Indicates the altitude of the current terrain, h u Indicates the upper limit of the drone's flight altitude. If the terrain altitude exceeds the upper limit of the drone's low-altitude flight, the drone's terrain threat is set to the maximum value;

[0164] When the real-time flight altitude of the UAV is h real When , the threat of obstacles at different altitudes to the drone is expressed as:

[0165]

[0166] Step 4: Based on the specific task requirements, a reasonable optimization target allocation model is established by conducting a detailed analysis of key factors such as cost or benefit indicators and collaborative constraints, and the target optimization function is defined according to the optimization model constraints;

[0167] Step 4.1, calculate the voyage cost;

[0168] Step 4.1.1, calculate the track distance in each track segment;

[0169] l MSTo represent the track length of the Mth UAV on the kth track segment, S is the number of track segments:

[0170]

[0171] In the formula, Represents the starting point coordinates of the Mth UAV on the kth track segment.

[0172] Step 4.1.2, calculate the range cost of each strike plan;

[0173] The attack plan is: if the first drone among m drones attacks the first target among n targets, it is one attack plan; if the first drone among m drones attacks the second target among n targets, it is another attack plan; if it attacks the nth target among n targets, it is the nth attack plan; each drone has the same number of attack plans, and in each attack plan, a drone only attacks one target.

[0174] In the process of trajectory planning, a trajectory is divided into S segments. The trajectory distance L of the Mth UAV attacking the Nth target MN The formula is:

[0175]

[0176] Step 4.1.3, normalize the voyage cost value;

[0177] The flight cost F after unification to the same order of magnitude l The formula is:

[0178]

[0179] Where, L MN It is represented as the track distance of the Mth UAV attacking the Nth target, L Nmax It represents the maximum flight distance of all drones when attacking. The ratio of the two can be used to calculate the range cost of each attack plan.

[0180] Step 4.2, calculate the threat cost in the battlefield;

[0181] For the convenience of calculation, since it is found that the threat level of the threat source is related to and inversely proportional to the distance when modeling the threat, it is considered that the threat cost is inversely proportional to the distance from the threat source to the track. The entire track is divided into S track segments, and the threat cost value of each track segment is calculated. The overall threat cost is the sum of the threat costs of the S track segments.

[0182] Step 4.2.1, calculate the threat cost generated in each track segment;

[0183] Assume that there is an Mth drone attacking the Nth target in the battlefield. First, each track is divided into S track segments. Two adjacent track points of the drone are represented by L k and L k+1 , threat source coordinates are expressed as In order to obtain the threat cost of each track segment, each track segment is divided into C equal parts, and the threat cost is indirectly calculated by obtaining the distance from the threat source to the C track points, which is specifically expressed as:

[0184]

[0185] Where N Th is the number of threat sources, C is the number of equal parts of each track, is the coordinate of each track point, t=1,2,…,N Th , c=1,2,…,C;

[0186] Step 4.2.2, calculate the overall threat cost in each track;

[0187] Sum the threat costs of S track segments to get the total threat cost F in the track t for:

[0188]

[0189] Step 4.2.3: Normalize the threat cost and calculate it on a unified scale;

[0190] Normalize the threat cost to a unified scale, the threat cost F t The formula is:

[0191]

[0192] In the formula, F t MN F represents the threat cost of the Mth UAV attacking the Nth target. tmax It represents the track with the largest total threat cost among all tracks;

[0193] Step 4.3: Calculate the overall attack benefit;

[0194] Step 4.3.1, calculate the overall attack benefit in each attack plan;

[0195] The target residual value is used to evaluate the target attack benefit cost. The attack benefit generated by the Mth drone attacking the Nth target is expressed as:

[0196] F v =V N ·P MN ·x ij

[0197] Where V N is the target value of the Nth target, P MN is the probability of damage when the Mth UAV attacks the Nth target, x ij is the decision variable;

[0198] Step 4.3.2: Normalize the attack revenue;

[0199] Similar to the normalization of the above indicators, the attack benefit F v Unified to the same order of magnitude, expressed as:

[0200]

[0201] In the formula, represents the attack benefit of the Mth drone attacking the Nth target, F vmax Indicates the maximum attack benefit that can be obtained among all attack plans.

[0202] Step 4.4, calculate the enemy target value;

[0203] Combined with the different types of weapons and their attack ranges, different threat levels are set for the target from 0 to 1. typ , which is used to clearly evaluate and distinguish the strike value of each target in the battlefield environment. In the process of executing combat missions, combat strike targets are usually divided into individual targets, ground targets, and air targets. Conventional combat targets can perform combat tasks such as battlefield reconnaissance and firepower strikes. Targets can communicate with each other and achieve combat coordination. Individual targets rely on flexible battlefield mobility characteristics and good concealment and camouflage capabilities; ground vehicle targets such as tanks, infantry fighting vehicles, and vehicle-mounted anti-tank missiles have different combat strike weapons and undertake different combat missions at the same time; air targets such as armed helicopters can quickly strike targets such as ground armored equipment, with strong mobility, good penetration effect, and good damage effect.

[0204] The specific value performance of each weapon in this embodiment is shown in Table 1.

[0205] Table 1 Classification of target types

[0206]

[0207] Step 4.5: Analyze the constraints generated in the target allocation;

[0208] Collaborative target allocation involves task matching between multiple UAVs and multiple targets. In this process, a series of constraints must be considered to ensure the effective execution of the task and the safety of the UAV. Considering the actual combat needs, the constraints considered in target allocation in this paper are as follows:

[0209] (1) Range constraints

[0210] Consider the distance that a suicide drone can fly before its fuel runs out. The distance that each suicide drone traverses should be less than its range, expressed as:

[0211]

[0212] In the formula, It is the maximum range of a self-destructing drone. is the actual crossing distance of the self-destructing drone,

[0213] N v It is the maximum flight altitude constraint.

[0214] (2) Decision variable constraints

[0215] Constraints are restrictions on decision variables, which limit the range of decision variables and ensure that the solution is feasible in practical situations. Constraints can be equalities or inequalities, depending on the nature of the problem. The constraints in this article are as follows:

[0216]

[0217] In the formula, the decision variable x ij is a 0-1 variable. When drone i acts on target j, x ij =1 means to carry out the attack, x ij = 0 means no strike was carried out. The decision variable is represented as x ij ∈(0,1),i=1,...,M,j=1,...,N, Indicates that a task is performed by at least one drone once, and the number of drones assigned does not exceed D jmax shelf; This means that a drone can only perform one mission at most;

[0218] Step 4.6, according to the target allocation optimization index, construct the objective function and describe each cost function;

[0219] The target allocation optimization index requires that the attack cost of the drone be minimized and the attack benefit be maximized.

[0220] The relative importance of each target is determined according to the commander's strategic planning or expert knowledge base, that is, different weights are assigned to them, and the linear weighted method is used to simplify the complex multi-objective problem into a single-objective optimization problem for solution. The optimization objective function of multi-UAV collaborative target allocation is expressed as:

[0221] F=λ1F l +λ2Ft -λ3F v

[0222] In the formula, λ = (λ1, λ2, λ3) is the weight vector in the optimization objective function, which indicates the different influence of each factor on the target allocation result, and satisfies λ i ∈(0,1); F l is the voyage cost, F t is the threat cost, F v For the attack benefit, when the range cost and threat cost are minimized and the attack benefit is maximized, the final optimal solution for target allocation is output.

[0223] Step 5: Use the improved genetic algorithm based on the adaptive penalty function idea and the greedy algorithm switching mechanism to perform target allocation, and perform algorithm switching operations on the two algorithms in different combat environments. The genetic algorithm based on the penalty function idea is improved on the traditional genetic algorithm to solve the objective function optimization problem with multiple sets of constraints; Figure 3 The figure shows a flow chart of the target allocation algorithm of the improved genetic algorithm of the present invention. The improved genetic algorithm is mainly used for the allocation mode under small-scale or non-urgent tasks in the target allocation algorithm based on the switching mechanism. In the improved genetic algorithm, an adaptive penalty function is mainly introduced to control the number of feasible solutions, thereby changing the quality of the optimal solution and improving the accuracy of the algorithm.

[0224] Step 5.1, initialize the battlefield environment, set the number of self-destruct drones, attack targets and threat sources;

[0225] Step 5.2, set the initial population size, number of iterations, selection rate, crossover rate and mutation rate;

[0226] Step 5.3, setting the encoding operation mode for the initial population;

[0227] Step 5.3.1, use binary coding to perform encoding operations, with the correspondence between self-destructing drones and targets, that is, each possible allocation scheme represents a chromosome, and each gene in the chromosome represents whether a specific drone is assigned to a target;

[0228] Step 5.3.2, encode the m self-destructing drones and n targets in the battlefield environment into binary codes, using 1 to represent that the self-destructing drone attacks the target, and 0 to represent that it does not attack the target. Then each chromosome has m×n numbers, i.e., “genes”;

[0229] Step 5.3.3: Group every n genes into a group, and arrange each group of genes vertically to obtain the multi-self-destructing drone target allocation matrix C m×n ;

[0230] Step 5.3.4: The i-th row of the matrix represents the self-destructing drone, i = 1, 2, 3, ..., m, and the j-th column of the matrix represents the target, c ij is the attack situation of the self-destructing UAV on the target, if c ij =0 means self-destructive drone M i For target N j No attack, if c ij =1 means self-destructive drone M i Number to target N j to strike;

[0231] Step 5.4, initialize the population and generate N individuals;

[0232] Step 5.5, determining the constraint conditions of the self-destructing UAV according to step 4.5;

[0233] Step 5.6: Construct a penalty function. For individuals that violate the constraints, subtract the penalty function when calculating the fitness value to reduce the fitness value.

[0234] Step 5.6.1. Use the two constraints in step 4.5 as penalty terms in the penalty function, and select the penalty factor, that is, the weight coefficient; construct the penalty function p(x)=f(x)+h(x)G(x), where f(x) is the objective function value, h(x) is the penalty factor, which is a function or constant, and G(x) is the penalty term.

[0235] Step 5.6.2: Substitute the constraints into the penalty function, which is expressed as p(x)=f(x)+η(n+m). According to the cumulative penalty principle, the penalty for individuals who violate more overall constraints is greater; where 0<η≤1, and can be adaptively adjusted according to the number of feasible solutions.

[0236] Step 5.7, construct fitness function;

[0237] In order to avoid the situation where the cooperative constraints between the self-destructing drones in step 5.6.1 are not met, a penalty term is added to the fitness function to affect the future evolution direction of the population, find the global optimal solution, and reduce the generation of infeasible solutions. Since the fitness function is a positive value or 0, the formula is constructed as follows: Fitness(x)=C max -p(x)

[0238]

[0239] Among them, f(x) is the objective function value, the fitness function is determined by the objective function, C max is the maximum estimated value of the objective function, obtained by estimation.

[0240] Step 5.8: Select, crossover and mutation operations are performed on the population;

[0241] After constructing the fitness function and penalty function, the population is subjected to genetic operations such as selection, crossover, and mutation, and iterated sequentially.

[0242] Step 5.8.1: According to the principle of survival of the fittest, the fitness of the individuals to the environment is measured by the value of the individual fitness function, and the individuals in the initial population are sorted;

[0243] Step 5.8.2: Use the roulette wheel method to select chromosomes in the population that meet the fitness standard value, and then use the elite selection and retention strategy to select and retain excellent individuals in the sorting results;

[0244] Step 5.8.3: Perform crossover and mutation operations on the retained excellent individuals to generate offspring populations, and combine the two populations to form a new offspring population R of size 2N. t .

[0245] Step 5.9, generating an adaptive fitness function value;

[0246] Assume that the number of feasible solutions after each iteration is P1, the total number is N, and the penalty factor That is, the weight of the penalty item changes in direct proportion to the increase or decrease of feasible solutions. When there are too many feasible solutions, the penalty factor will also increase in the next iteration, and vice versa. The penalty function in the genetic algorithm can adaptively adjust the proportion of feasible solutions output according to different population sizes.

[0247] Step 6: When the genetic algorithm based on the improved adaptive penalty function idea cannot meet the operational requirements, in order to ensure the quality and timeliness of the decision solution, the allocation algorithm uses the greedy algorithm to continue solving the algorithm. Figure 4 The figure shows a flow chart of the target allocation algorithm of the greedy algorithm of the present invention. The greedy algorithm is mainly used in the allocation mode under large-scale or urgent tasks in the target allocation algorithm based on the switching mechanism. The greedy algorithm can achieve fast allocation while ensuring the accuracy of the results.

[0248] Step 6.1, parameter initialization: It is known that there are n targets as input for solving the problem, and the solution set J of the problem is set to empty; the parameters of the maximum allowed number of iterations of the greedy algorithm, the end mark threshold and the number of targets are initialized.

[0249] Step 6.2, define the objective function: According to the problem requirements, randomly generate task pairs of self-destructing drones and attack targets, and calculate the objective evaluation function. Select a metric and sort the n inputs according to this metric;

[0250] Step 6.3, traverse all the attack plans of m self-destructing drones and n targets, and calculate the corresponding cost value through the cost function formula;

[0251] Step 6.4: According to the absolute value of the objective function, establish the cost matrix, where A is defined as 11 ,A 12 ,A 13 ...A 1m The value of is the cost value of the attack plan consisting of the target 1 and M drones. The remaining row elements A mn They all represent the attack cost of the Mth drone attacking the Nth target, and finally form a cost matrix containing N×M cost value elements.

[0252]

[0253] Step 6.4.1. Use the comparison method to find the largest element in the cost matrix A and record its position.

[0254] Step 6.4.2: Assume that the largest element in the cost matrix A is A 00 , it means that after comprehensive consideration of the target cost and track cost, the attack benefit of our drone on the enemy's N0 target by M0 is the greatest, so we implement the M0-to-N0 attack plan;

[0255] Step 6.4.3: Since N0 has become the target of M0, in order to avoid subsequent repeated allocation and over-concentrated attacks, A 00 The row and column are set to ∞;

[0256] Step 6.4.4: A 00 After the row and column are set to ∞, A 00 No longer participate in subsequent assignment tasks, then return to step 6.3 to continue the loop iteration, looking for the next optimal solution to update the status;

[0257] Step 6.4.5, after completing all one-to-one assignments, restore the original cost matrix structure, set all assigned solutions to ∞ to avoid searching again, and re-search the many-to-one assignment results for the remaining elements in the matrix;

[0258] Step 6.5, output the optimal solution: when all n inputs have been searched, if the task conditions are met, the search stops, that is, the current solution set is the global optimal solution, and the search result is output;

[0259] Step 7: Set the main switching mechanism of the target allocation algorithm;

[0260] For different problems of different nature and complexity, if the solution space of the problem is large and the complexity is high, a genetic algorithm needs to be used for global search; if the solution space of the problem is small and the complexity is low, it is only necessary to switch to the greedy algorithm to find a better solution; for the computational complexity and execution time of the algorithm, as well as the limited computing resources, if time and resources are limited, a greedy algorithm needs to be used to quickly obtain a solution. Figure 5 The flowchart of the algorithm switching mechanism of the present invention is shown. According to the battlefield environment requirements, the global battlefield environment variables are initialized, and the goals and conditions of the switching algorithm are clarified for the nature of the problem, time and resource constraints, and accuracy requirements. It is set to switch to the genetic algorithm when higher accuracy or global optimal solution is required; when the solution space of the problem is large and the computing resources are insufficient, switch to the greedy algorithm. Using the switching mechanism for target allocation can select the most suitable algorithm for decision-making under the current situation, enhance combat flexibility, and improve the success rate of the task.

[0261] Step 7.1, initialize battlefield global variable information according to battlefield environment requirements;

[0262] Step 7.2: Define combat scenarios of different scales according to the different numbers and levels of combat weapons, and specify the completion time T for each target assignment task. c , where the number of our self-destruct drones participating in the battle is m, the number of enemy combat weapons is n, and it is stipulated that m>n for each battlefield scale. The following are the three defined combat scales, and subsequent actual battlefields are also classified according to this definition;

[0263] If the number of our drone targets m≤20 and the number of enemy targets n≤20, it is a small-scale conflict;

[0264] The number of our drone targets is 20<m≤50 and the number of enemy targets is 20<m≤50, which is a medium-scale operation;

[0265] The number of our drone targets is 50<m≤100 and the number of enemy targets is 50<m≤100, which is a large-scale battle;

[0266] Other situations are not taken into account.

[0267] Step 7.3: According to the nature of the task, the planning time and the requirements for optimizing performance, clarify the goals and conditions of the switching algorithm, and clarify the completion time T for each target assignment task. f .

[0268] Step 7.4, establish a mission planning model database. Conduct comparative tests of the two algorithms at different combat scales, and continuously increase the experimental scale to obtain multiple groups of calculation time and objective function change values ​​under different combat situations. Store the experimental results to form a database and establish a historical target allocation model;

[0269] Step 7.5: Enter actual battlefield combat parameters into the system.

[0270] Step 7.6: Identify the characteristic indicators of the actual battlefield situation and make similarity judgments, including the number of weapons in combat, the scale of operations, and the time T for the command to be issued. f As the main feature judgment indicator;

[0271] Step 7.7: Develop a switching strategy;

[0272] Step 7.7.1. Compare the case data including calculation time and target function change value in the existing historical database with the actual combat conditions, search in the database, and select the target allocation model with the highest similarity as the matching object.

[0273] Step 7.7.2: Allocate time T according to the target specified in the mission c As a benchmark, complete the similarity judgment through step 7.6 and select the case with the same scale as the actual battlefield and the highest similarity;

[0274] Step 7.7.3, extract the most similar case, compare the optimization performance of the genetic algorithm and the greedy algorithm within the benchmark time range, and compare the algorithm decision results. According to the comparison results, due to the high similarity, match the battlefield environment in the database with the actual combat scenario, and the algorithm decision in this case is fed back as the optimal decision result under the current situation.

[0275] Step 7.8, monitor and evaluate the results; if there are new changes in the battlefield situation, it is necessary to identify the characteristic indicators under the new situation again, return to step 7.6, and make new algorithm planning decision results to ensure the real-time and accuracy of target allocation.

[0276] like Figure 6 The figure shows the optimization performance comparison of the algorithm of the present invention. Through experimental comparison, it can be concluded that the switching mechanism proposed in this paper is effective and feasible. First, the specified execution time of the task is used as a benchmark to compare the optimization of the algorithm within this time range. In non-urgent tasks, that is, when there is ample time for task planning, the optimality of the algorithm is considered, and the algorithm with the best optimization performance is selected for decision-making; in urgent tasks, that is, when the task planning time is extremely short, the optimization should be abandoned, and reasonable task planning should be carried out based on the planning time. Under the premise of meeting different combat requirements, the decision-making algorithm that best suits the requirements can be selected to the greatest extent, which increases the flexibility of battlefield decision-making and improves the overall combat effectiveness.

[0277] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) to form a technical solution.

Claims

1. A method for rapid target assignment of UAV coordinated attack based on case-based reasoning, characterized in that: The following steps are involved: Step 1: Based on the requirements of cluster combat strike missions and battlefield environment, a unified modeling of the battlefield space environment is performed; Assume that m self-destructing drones form a drone cluster to jointly perform strike missions, attacking n targets scattered at different locations on the battlefield, and there are t threat radius sets distributed in the battlefield, R = {R1, R2, R3...R t }, the coordinates are {(x t1 ,y t1 ),(x t2 ,y t2 )...(x tt ,y tt )} threat source, and ensure that each self-destructing drone performs at most one mission and strikes one target; where the initialization defines the self-destructing drone cluster M = {M1,M2,...,M m }, ground target set N = {N1, N2, ..., N n }, set the target initial position to (x i ,y j ), target N j The value of Q j ; In order to smoothly implement the strike mission, during the execution of the coordinated strike mission, the self-destructing drone needs to avoid the threat sources appearing on the battlefield and finally reach the target mission point to carry out the strike; and according to the target type, the upper limit of the number of strikes for each target is set to D jmax That is, the number of drones assigned to each target should be less than or equal to D jmax indivual; Step 2: Design a two-dimensional battlefield simulation model using the grid method to describe the battlefield space in detail; In the two-dimensional battlefield simulation model, each grid represents a coordinate point, which is used to indicate the position of the drone, target, and obstacle. The planning space is represented by a set Ω, where (x, y) represents the horizontal and vertical coordinates of a point in the space Ω. max and max They represent the maximum distance positions specified by the horizontal and vertical coordinates, respectively, as shown in the following formula: Ω={(x,y)|0≤x≤x max ,0≤y≤y max } Step 3: Model the battlefield threat source, select radar threat, anti-aircraft artillery threat and terrain threat as threat source parameters and calculate the threat source threat degree; Step 4: Establish a reasonable optimization target allocation model and define the target optimization function according to the optimization model constraints; Step 5: Use the improved genetic algorithm based on the adaptive penalty function idea and the greedy algorithm switching mechanism to perform target allocation, and perform algorithm switching operations on the two algorithms in different combat environments. The genetic algorithm based on the penalty function idea is improved on the traditional genetic algorithm to solve the objective function optimization problem with multiple sets of constraints; Step 6: When the genetic algorithm with the improved adaptive penalty function idea cannot meet the operational requirements, the allocation algorithm uses the greedy algorithm to continue solving the algorithm; Step 7: Set the main switching mechanism of the target allocation algorithm to achieve rapid target allocation for UAV coordinated attack.

2. According to the case-based reasoning method for rapid target allocation of UAV coordinated attack according to claim 1, it is characterized in that: The step 3 specifically comprises the following steps: Step 3.1, calculate the radar threat level within the battlefield; The radar echo signal power is expressed as: Where P t is the transmitting power, G is the transmitting antenna gain, σ is the radar scattering area, R is the distance from the target to the radar, in meters, and λ is the wavelength of the electromagnetic wave; for a certain UAV, the radar scattering area is constant, and the above formula is simplified to: Where K is a constant. When the radar is not interfered with, the probability of detecting a target is related to the quality of the received signal. Assuming that the radar can perform omnidirectional detection, the maximum detection radius of the radar is R. max , then the detection probability of the radar to the UAV is expressed as: Step 3.2, calculate the threat level of anti-aircraft artillery; When the enemy radar detects the drone, the air defense missile will attack according to the target information. The kill probability of the air defense missile is: P m =T·Y·Z[1-(1-A·E·W) Q ] Where, T is the radar detection probability of the surface-to-air missile, Y is the radar interception guidance probability, Z is the information transmission conversion probability, A is the missile successful launch probability, E is the missile flight reliability probability, W is the missile kill probability, and Q is the number of missiles launched at one time; Step 3.3, calculate the terrain threat level; Calculate the height threat and size of the terrain: Where, T h Indicates the threat level of terrain altitude, h m Indicates the altitude of the current terrain, h u Indicates the upper limit of the drone's flight altitude. If the terrain altitude exceeds the upper limit of the drone's low-altitude flight, the drone's terrain threat is set to the maximum value; When the real-time flight altitude of the UAV is h real When , the threat of obstacles at different altitudes to the drone is expressed as: 。 3. The method for rapid target allocation of UAV coordinated attack based on case reasoning according to claim 1 is characterized in that: The step 4 specifically comprises the following steps: Step 4.1, calculate the voyage cost; Step 4.1.1, calculate the track distance in each track segment; l MS To represent the track length of the Mth UAV on the kth track segment, S is the number of track segments: In the formula, Represents the starting point coordinates of the Mth UAV on the kth track segment; Step 4.1.2, calculate the range cost of each strike plan; The attack plan is: if the first drone among m drones attacks the first target among n targets, it is one attack plan; if the first drone among m drones attacks the second target among n targets, it is another attack plan; if it attacks the nth target among n targets, it is the nth attack plan; each drone has the same number of attack plans, and in each attack plan, a drone attacks only one target; In the process of trajectory planning, a trajectory is divided into S segments. The trajectory distance L of the Mth UAV attacking the Nth target MN The formula is: Step 4.1.3, normalize the voyage cost value; The flight cost F after unification to the same order of magnitude l The formula is: Where, L MN It is represented as the track distance of the Mth UAV attacking the Nth target, L Nmax It represents the maximum flight distance of all drones when attacking. The ratio of the two can be used to calculate the range cost of each attack plan. Step 4.2, calculate the threat cost in the battlefield; The threat cost is inversely proportional to the distance from the threat source to the track. The entire track is divided into S track segments, and the threat cost value of each track segment is calculated. The overall threat cost is the sum of the threat costs of the S track segments. Step 4.2.1, calculate the threat cost generated in each track segment; Assume that there is an Mth drone attacking the Nth target in the battlefield. First, each track is divided into S track segments. Two adjacent track points of the drone are represented by L k and L k+1 , threat source coordinates are expressed as In order to obtain the threat cost of each track segment, each track segment is divided into C equal parts, and the threat cost is indirectly calculated by obtaining the distance from the threat source to the C track points, which is specifically expressed as: Where N Th is the number of threat sources, C is the number of equal parts of each track, is the coordinate of each track point, t=1,2,…,N Th , c=1,2,…,C; Step 4.2.2, calculate the overall threat cost in each track; Sum the threat costs of S track segments to get the total threat cost F in the track t for: Step 4.2.3: Normalize the threat cost and calculate it on a unified scale; Normalize the threat cost to a unified scale, the threat cost F t The formula is: In the formula, F t MN F represents the threat cost of the Mth UAV attacking the Nth target. tmax It represents the track with the largest total threat cost among all tracks; Step 4.3: Calculate the overall attack benefit; Step 4.3.1, calculate the overall attack benefit in each attack plan; The target residual value is used to evaluate the target attack benefit cost. The attack benefit generated by the Mth drone attacking the Nth target is expressed as: F v =V N ·P MN ·x ij Where V N is the target value of the Nth target, P MN is the probability of damage when the Mth UAV attacks the Nth target, x ij is the decision variable; Step 4.3.2: Normalize the attack revenue; Attack gain F v Unified to the same order of magnitude, expressed as: In the formula, represents the attack benefit of the Mth drone attacking the Nth target, F vmax It represents the maximum attack benefit that can be obtained among all attack plans; Step 4.4, calculate the enemy target value; Combined with the different types of weapons and their attack ranges, different threat levels are set for the target from 0 to 1. typ , used to clearly assess and distinguish the strike value of each target in the battlefield environment; Step 4.5: Analyze the constraints generated in the target allocation; Step 4.6, according to the target allocation optimization index, construct the objective function and describe each cost function; The target allocation optimization index requires that the attack cost of the drone be minimized and the attack benefit be maximized; The optimization objective function of multi-UAV collaborative target allocation is expressed as: F=λ1F l +λ2F t -λ3F v In the formula, λ = (λ1, λ2, λ3) is the weight vector in the optimization objective function, which indicates the different influence of each factor on the target allocation result, and satisfies F l is the voyage cost, F t is the threat cost, F v For the attack benefit, when the range cost and threat cost are minimized and the attack benefit is maximized, the final optimal solution for target allocation is output.

4. The method for rapid target allocation of UAV coordinated attack based on case reasoning according to claim 1 is characterized in that: The step 5 specifically comprises the following steps: Step 5.1, initialize the battlefield environment, set the number of self-destruct drones, attack targets and threat sources; Step 5.2, set the initial population size, number of iterations, selection rate, crossover rate and mutation rate; Step 5.3, setting the encoding operation mode for the initial population; Step 5.3.1, use binary coding to perform encoding operations, with the correspondence between self-destructing drones and targets, that is, each possible allocation scheme represents a chromosome, and each gene in the chromosome represents whether a specific drone is assigned to a target; Step 5.3.2, encode the m self-destructing drones and n targets in the battlefield environment into binary codes, using 1 to represent that the self-destructing drone attacks the target, and 0 to represent that it does not attack the target. Then each chromosome has m×n numbers, i.e. "genes"; Step 5.3.3: Group every n genes into a group, and arrange each group of genes vertically to obtain the multi-self-destructing drone target allocation matrix C m×n ; Step 5.3.4: The i-th row of the matrix represents the self-destructing drone, i = 1, 2, 3, ..., m, and the j-th column of the matrix represents the target, c ij is the attack situation of the self-destructing UAV on the target, if c ij =0 means self-destructive drone M i For target N j No attack, if c ij =1 means self-destructive drone M i Number to target N j to strike; Step 5.4, initialize the population and generate N individuals; Step 5.5, determining the constraint conditions of the self-destructing UAV according to step 4.5; Step 5.6: Construct a penalty function. For individuals that violate the constraints, subtract the penalty function when calculating the fitness value to reduce the fitness value. Step 5.6.

1. Use the two constraints in step 4.5 as penalty terms in the penalty function, and select the penalty factor, i.e., the weight coefficient; construct the penalty function p(x)=f(x)+h(x)G(x), where f(x) is the objective function value, h(x) is the penalty factor, and G(x) is the penalty term; Step 5.6.2: Substitute the constraints into the penalty function, which is expressed as p(x)=f(x)+η(n+m). According to the cumulative penalty principle, the penalty for individuals who violate more overall constraints is greater; where 0<η≤1, and can be adaptively adjusted according to the number of feasible solutions. Step 5.7: Construct the fitness function, the formula is as follows: Fitness(x) = C max -p(x) Among them, f(x) is the objective function value, C max is the maximum estimated value of the objective function, obtained by estimation; Step 5.8: Select, crossover and mutation operations are performed on the population; After constructing the fitness function and penalty function, the population is subjected to genetic operations such as selection, crossover, and mutation, and iterated sequentially; Step 5.8.1: According to the principle of survival of the fittest, the fitness of the individuals to the environment is measured by the value of the individual fitness function, and the individuals in the initial population are sorted; Step 5.8.2: Use the roulette wheel method to select chromosomes in the population that meet the fitness standard value, and then use the elite selection and retention strategy to select and retain excellent individuals in the sorting results; Step 5.8.3: Perform crossover and mutation operations on the retained excellent individuals to generate offspring populations, and combine the two populations to form a new offspring population R of size 2N. t ; Step 5.9, generating an adaptive fitness function value; Assume that the number of feasible solutions after each iteration is P1, the total number is N, and the penalty factor That is, the weight of the penalty item changes in direct proportion to the increase or decrease of feasible solutions. When there are too many feasible solutions, the penalty factor will also increase in the next iteration, otherwise it will decrease.

5. The method for rapid target allocation of UAV coordinated attack based on case reasoning according to claim 1 is characterized in that: The step 6 specifically comprises the following steps: Step 6.1, parameter initialization: It is known that there are n targets as input for solving the problem, and the solution set J of the problem is set to be empty; the parameters of the maximum allowed number of iterations, the end mark threshold and the number of targets of the greedy algorithm are initialized; Step 6.2, define the objective function: according to the problem requirements, randomly generate task pairs of self-destructing drones and attack targets, and calculate the objective evaluation function; select a metric and sort the n inputs according to this metric; Step 6.3, traverse all the attack plans of m self-destructing drones and n targets, and calculate the corresponding cost value through the cost function formula; Step 6.4: According to the absolute value of the objective function, establish the cost matrix, where A is defined as 11 ,A 12 ,A 13 ...A 1m The value of is the cost value of the attack plan consisting of the target 1 and M drones. The remaining row elements A mn They all represent the attack cost of the attack plan of the Mth UAV attacking the Nth target, and finally form a cost matrix containing N×M cost value elements; Step 6.4.

1. Use the comparison method to find the largest element in the cost matrix A and record its position. Step 6.4.2: Assume that the largest element in the cost matrix A is A 00 , it means that after comprehensive consideration of the target cost and track cost, the attack benefit of our drone on the enemy's N0 target by M0 is the greatest, so we implement the M0-to-N0 attack plan; Step 6.4.3: Since N0 has become the target of M0, 00 The row and column are set to ∞; Step 6.4.4: A 00 After the row and column are set to ∞, A 00 No longer participate in subsequent assignment tasks, then return to step 6.3 to continue the loop iteration, looking for the next optimal solution to update the status; Step 6.4.5, after completing all one-to-one assignments, restore the original cost matrix structure, set all assigned solutions to ∞ to avoid searching again, and re-search the many-to-one assignment results for the remaining elements in the matrix; Step 6.5: Output the optimal solution: When all n inputs have been searched, if the task conditions are met, the search stops, that is, the current solution set is the global optimal solution, and the search results are output.

6. The method for rapid target allocation of UAV coordinated attack based on case reasoning according to claim 1 is characterized in that: The step 7 specifically includes the following steps: Step 7.1, initialize battlefield global variable information according to battlefield environment requirements; Step 7.2: Define combat scenarios of different scales according to the different numbers and levels of combat weapons, and specify the completion time T for each target assignment task. c , where the number of our self-destruct drones participating in the battle is m, the number of enemy combat weapons is n, and it is stipulated that m>n for each battlefield scale. The following are the three defined combat scales, and subsequent actual battlefields are also classified according to this definition; If the number of our drone targets m≤20 and the number of enemy targets n≤20, it is a small-scale conflict; The number of our drone targets is 20<m≤50 and the number of enemy targets is 20<m≤50, which is a medium-scale operation; The number of our drone targets is 50<m≤100 and the number of enemy targets is 50<m≤100, which is a large-scale battle; Other circumstances are not taken into account; Step 7.3: According to the nature of the task, the planning time and the requirements for optimizing performance, clarify the goals and conditions of the switching algorithm, and clarify the completion time T for each target assignment task. f ; Step 7.4, establish a task planning model database; conduct comparative tests on the two algorithms at different combat scales, and continuously increase the experimental scale to obtain multiple groups of calculation time and objective function change values ​​under different combat situations; store the experimental results to form a database, and establish a historical target allocation model; Step 7.5: Input actual battlefield combat parameters into the system; Step 7.6: Identify the characteristic indicators of the actual battlefield situation and make similarity judgments, including the number of weapons in combat, the scale of operations, and the time T for the command to be issued. f As the main feature judgment indicator; Step 7.7: Develop a switching strategy; Step 7.7.1, compare the case data including calculation time and target function change value in the existing historical database with the actual combat conditions, search in the database, and select the target allocation model with the highest similarity as the matching object; Step 7.7.2: Allocate time T according to the target specified in the mission c As a benchmark, complete the similarity judgment through step 7.6 and select the case with the same scale as the actual battlefield and the highest similarity; Step 7.7.3, extract the most similar case, compare the optimization performance of the genetic algorithm and the greedy algorithm within the benchmark time range, and compare the algorithm decision results. According to the comparison results, due to the high similarity, match the battlefield environment in the database with the actual combat scenario, and the algorithm decision in this case is fed back as the optimal decision result under the current situation; Step 7.8, monitor and evaluate the results; if there are new changes in the battlefield situation, it is necessary to identify the characteristic indicators under the new situation again, return to step 7.6, and make new algorithm planning decision results to ensure the real-time and accuracy of target allocation.

7. The method for rapid target allocation of UAV coordinated attack based on case reasoning according to claim 3 is characterized in that: The constraints described in step 4.5 are as follows: (1) Range constraints The distance traversed by each self-destructing drone should be less than its range, expressed as: In the formula, It is the maximum range of a self-destructing drone. is the actual crossing distance of the self-destructing drone, N v It is the maximum flight altitude constraint; (2) Decision variable constraints In the formula, the decision variable x ij is a 0-1 variable. When drone i acts on target j, x ij =1 means to carry out the attack, x ij = 0 means no strike was carried out; the decision variable is represented by x ij ∈(0,1),i=1,...,M,j=1,...,N, Indicates that a task is performed by at least one drone once, and the number of drones assigned does not exceed D jmax shelf; Indicates that a drone can only perform one mission at most.

Images