Air combat decision-making method based on position weight speed updating particle swarm algorithm

Abstract

The invention provides an air combat decision-making method based on a position weight speed updating particle swarm algorithm, which comprises the following steps of: firstly, performing problem simplification on a complex air combat scene, establishing a success rate model of our-party and enemy aircraft attack and coping, and establishing and analyzing an index system in the air combat scene by adopting an analytic hierarchy process; the method comprises the following steps: firstly, obtaining an aircraft attack success influence factor and a corresponding weight, then obtaining a simplified objective function in combination with a multi-objective optimization model and a linear weighting method, and finally, introducing position weight information and an optimization speed updating strategy on the basis of a traditional PSO algorithm, and proposing a PW-PSO algorithm. And the air combat decision is solved based on the improved particle swarm optimization (PW-PSO). Through the scheme of the invention, the convergence speed of the algorithm can be effectively improved, the decision-making efficiency is improved, more accurate attack to enemy targets and effective avoidance of own damage are realized, and the reliability and the real-time performance of air combat decision making are remarkably improved.

Description

Technical Field

[0001] The invention relates to the field of intelligent optimization algorithm and aviation combat command technology, and in particular to an air combat decision-making method based on a position weighted speed updating particle swarm algorithm. Background Art

[0002] Air combat decision-making is a very challenging and complex field in modern warfare. With the rapid development of drones and intelligent weapons, the situation in air combat changes rapidly, involving multi-dimensional factors of enemy and friendly aircraft, such as position, speed, attack capability and defense capability. Traditional air combat decision-making methods mainly rely on rule-based systems or classic algorithms in operations research, such as dynamic programming and genetic algorithms. Although these methods are effective under specific conditions, they often face problems such as low computational efficiency, slow convergence speed and easy to fall into local optimal solutions in complex air combat environments with high dimensions, multiple constraints and multiple targets.

[0003] Particle Swarm Optimization (PSO), as an optimization algorithm based on swarm intelligence, has been widely used in various optimization problems due to its advantages such as simplicity, easy implementation and strong global search capability. However, the standard PSO algorithm still has the challenges of slow convergence and easy falling into local optimal solutions in multi-objective problems. Although some researchers have improved PSO by introducing inertia weights, adaptive learning strategies and mutation operations, when dealing with problems such as air combat decision-making, which has high real-time requirements and strong dynamics, there are still problems such as insufficient convergence speed, easy falling into local optimal solutions and low search efficiency. Therefore, there is an urgent need for an improved particle swarm optimization algorithm that can improve the convergence speed while avoiding the local optimal trap and adapt to the dynamic changes of the air combat environment, so as to provide a more efficient and reliable solution for air combat decision-making. Summary of the invention

[0004] In order to solve the above problems, the present invention proposes an air combat decision-making method based on a position weighted speed updating particle swarm algorithm. On the basis of retaining the high decision-making accuracy of the traditional PSO algorithm, the position weight information is introduced and the speed updating strategy is optimized to obtain a particle swarm optimization algorithm based on position weight (PW-PSO). This optimization strategy can intelligently adjust the search speed of particles according to the distance between the current decision and the global optimal solution, thereby accelerating the convergence speed of the algorithm, ensuring the accuracy of the decision results, and providing a more efficient and reliable solution for air combat decision-making.

[0005] The present invention provides an air combat decision-making method based on a position weighted velocity updating particle swarm algorithm, comprising the following steps:

[0006] Step S1: Establish an air combat scenario model, simplify the complex air combat scenario, and establish a success rate model for attack and response of our and enemy aircraft;

[0007] Step S2: Based on the analytic hierarchy process (AHP), the constraint conditions are established and the index system in the air combat scenario is established and analyzed using the analytic hierarchy process (AHP) to obtain the influencing factors of the success of the aircraft attack and the corresponding weights;

[0008] Step S3: Establishing an objective function, using the core idea of ​​maximizing the damage to the enemy and minimizing the damage to one's own side, calculating the success rate through the established success rate model, and simplifying the objective function through a linear weighted method;

[0009] Step S4: Optimizing the particle swarm algorithm PSO, by adding position weights to the velocity update strategy of PSO to improve the efficiency and accuracy of the algorithm;

[0010] Step S5: Based on the improved particle swarm algorithm PW-PSO, the decision problem of the air combat process is solved by using the particle swarm algorithm with the speed update strategy with position weight added to obtain the attack target of our aircraft.

[0011] Furthermore, the step S1 specifically includes:

[0012] Step S11, simplifying the decision-making problem, simplifying the three-dimensional air combat into two-dimensional simulation, using the AHP method to determine the factors affecting the combat results, and constructing a multi-level structural model, the model includes a target layer, a criterion layer and a solution layer, the influencing factors of the target layer include the air combat capability index, the influencing factors of the criterion layer include the combat environment, the aircraft flight performance, the weapon system and the aircraft hardware performance, and the influencing factors of the solution layer include the aircraft attack range D, the aircraft detection range R and the aircraft attack angle A;

[0013] Step S12, setting conditions, assuming that each side has N aircraft, where N is a positive integer, representing the initial number of aircraft on each side, and each aircraft can only launch one attack in the set scenario, and defining the damage success rate P o (i,j) and P e (i, j) represent the success rate of our i-th aircraft destroying the enemy j-th aircraft and the enemy i-th aircraft destroying our j-th aircraft, respectively. e (i,j) also represents the threat value of the enemy's i-th aircraft to our j-th aircraft. The decision matrix J is introduced o (i,j) and J e (i, j), respectively represents the decision results of our aircraft on the enemy aircraft and the decision results of the enemy aircraft on our aircraft. The values ​​of the matrix elements are 0 or 1, where 0 means no attack and 1 means attack;

[0014] Step S13, coordinated attack and escape probability, calculate the success probability of all our aircraft coordinated attack on a certain enemy aircraft as:

[0015]

[0016] Under enemy attack, the probability of our aircraft escaping successfully is:

[0017]

[0018] Step S14, parameter setting, randomly generating the coordinates and directions of 2N aircraft within the specified scene range according to the pseudo-random number generator.

[0019] Furthermore, the step S2 specifically includes:

[0020] Step S21, judgment matrix construction, compare each factor of the same level in the criterion layer and the solution layer in pairs, understand its importance relative to the criterion of the previous layer, use a 1-9 scale to quantitatively express the importance of factors, and construct a judgment matrix;

[0021] Step S22, relative weight calculation of influencing factors. For the influencing factors in the criterion layer and the scheme layer, the normalized relative importance vector associated with each factor and its upper layer factor is calculated using the square root method:

[0022]

[0023]

[0024] Among them, W i is the relative importance vector, a ij represents the data in the i-th row and j-th column of the judgment matrix, n represents the number of columns in the judgment matrix, W i 0 represents the normalized relative importance vector;

[0025] Step S23, consistency check, verifies the rationality and feasibility of the judgment matrix, uses two key indicators, consistency index CI and consistency ratio CR, calculates CI and CR respectively by the following formula, and verifies whether the CR value is less than 0.1:

[0026]

[0027] Among them, n represents the number of columns of the judgment matrix, λ max represents the maximum eigenvalue of the judgment matrix, and RI represents the average value of the consistency criteria of random judgment matrices of the same order;

[0028] Step S24, determining the attack success factors and weights, using the AHP method to determine the factors that affect the success of the aircraft attack and the corresponding weights, that is, whether the factors in the solution layer have an impact on the target layer, and the weight of the impact;

[0029] Step S25, attack condition setting, determines the attack condition according to the current position and attitude information of the aircraft and the attack constraints set by the weapon system, the attack conditions include: the distance of the enemy aircraft relative to our aircraft is less than the detection range R of our aircraft, and the angle of the enemy aircraft relative to our aircraft is within the detection range of our aircraft, and it is set that each aircraft can only attack one enemy aircraft in one attack, and the number of attacks launched by all our aircraft does not exceed N; when all these attack conditions are met, the aircraft launches an attack.

[0030] Furthermore, the step S3 specifically includes:

[0031] Step S31, determining the main task, removing the greatest threat to our side from the target, that is, ensuring that our side causes the greatest damage to the enemy while minimizing our own losses;

[0032] Step S32, establish an objective function to ensure that our mission of causing maximum damage to the enemy is:

[0033] maxP o =α1·(1-D / R)+α2·J o

[0034] Among them J o represents the decision matrix, α1 and α2 represent weights, which are set according to the actual situation;

[0035] Our tasks to ensure that we minimize our own losses are:

[0036] minP e =β1·(1-D / R)+β2·J e

[0037] Among them J e represents the decision matrix, β1 and β2 represent weights;

[0038] Step S33, determining the target of attack, and determining the enemy aircraft to be attacked according to the threat level of the enemy aircraft to us, that is, the probability of the enemy's successful attack on us;

[0039] Step S34, simplify the multi-objective function problem, set weight coefficients according to the importance of the two objective functions, use the linear weighted sum method to combine and simplify the two objective functions, and set the objective function P o The weight coefficient is 0.6, and the objective function P e The weight coefficient is 0.4:

[0040] Q=ω1·P o -ω2·P e .

[0041] Furthermore, the step S4 specifically includes:

[0042] Step S41, the PSO algorithm idea is: create a randomly distributed particle population, each particle represents a potential solution to the problem, initialize the speed and position of each particle, update the speed and position of each particle according to the following speed update formula and position update formula, calculate the fitness value of each particle, and update the individual optimal position P best and the global optimal position G best :

[0043] v i (t+1)=ωv i (t)+c1(P best (t)-x i (t))+c2(G best (t)-x i (t))

[0044] x i (t+1)=x i (t)+v i (t+1)

[0045] Among them, v i (t) represents the velocity of the ith particle at time t, x i (t) represents the position of the i-th particle at time t, c1 and c2 represent learning factors, and ω represents the weight;

[0046] Step S42, the PW-PSO speed update strategy is: introduce the position weight parameter into the speed update formula. When the particle's judgment is far from the optimal decision, increase the weight parameter to accelerate the particle to move in the optimal direction to quickly narrow the gap. When the decision is close to the optimal solution, reduce the weight and slow down the particle speed. According to the idea of ​​dynamically adjusting the speed and the distance from the optimal solution, the speed update formula is adjusted as follows:

[0047] v i (t+1)=ωv i (t)+(c1(P best (t)-x i (t))+c2(G best (t)-x i (t)))(G best (t)-x i (t)).

[0048] Furthermore, the step S5 specifically includes:

[0049] Step S51, air combat scene initialization, randomly generate a group of particles, each particle represents a potential solution, that is, a group of aircraft numbers that our aircraft want to attack, the position and speed are randomly initialized, the initial position and speed of each particle are randomly generated in the solution space, and the maximum number of iterations Gen is set. When the maximum number of iterations is reached, the iteration is stopped;

[0050] Step S52, fitness evaluation, calculating the fitness value of each particle according to the objective function Q of the problem;

[0051] Step S53, update individual and global optimal positions:

[0052] Individual optimal position: For each particle, if its current fitness value is better than its historical best fitness value, its individual optimal position is updated;

[0053] Global optimal position: Among the individual optimal positions of all particles, the position with the best fitness value is selected as the global optimal position;

[0054] Step S54, speed update, generate a random number r, if r is less than 0.6, use the speed update formula of PW-PSO, if r is greater than or equal to 0.6, use the speed update formula of the original particle swarm algorithm PSO;

[0055] Step S55, position update, updating the position of the particle according to the position update formula;

[0056] Step S56, iterative check, the current iteration number increases by 1, if the current iteration number does not reach the maximum iteration number Gen, then return to step S52 to continue iterating; otherwise, enter step S57;

[0057] Step S57, output the optimal solution. When the algorithm terminates, the individual with the highest fitness value is output, that is, the optimized attack decision.

[0058] The beneficial technical effects of the present invention are as follows: by introducing position weights into the traditional particle swarm optimization algorithm, the present invention can dynamically adjust the search speed of particles, avoid the algorithm from falling into the local optimum, and improve the global search capability; by simplifying the problem of complex air combat scenes, a success rate model of our and enemy aircraft attacks and responses is established; and the hierarchical analysis method is used to establish and analyze the indicator system in the air combat scene, and the influencing factors of the success of the aircraft attack and the corresponding weights are obtained, and then the simplified objective function is obtained by combining the multi-objective optimization model and the linear weighted method, so that while improving the calculation efficiency, high-precision optimization results are guaranteed. Through the above method, in a complex and changeable air combat environment, the convergence speed of the algorithm can be effectively improved, the decision-making efficiency is improved, and the high-precision optimization results are maintained at the same time, so as to achieve more accurate strikes on enemy targets and effective avoidance of damage to one's own side, and significantly improve the reliability and real-time performance of air combat decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 It is a flow chart of an air combat decision method of a particle swarm algorithm based on a velocity update strategy of position weight provided by an embodiment of the present invention;

[0061] Figure 2 It is a structural model for air combat performance evaluation provided by an embodiment of the present invention;

[0062] Figure 3 is a schematic diagram of aircraft parameters provided by an embodiment of the present invention;

[0063] Figure 4 is a schematic diagram of aircraft positions provided by an embodiment of the present invention;

[0064] Figure 5 It is a comparison diagram of iterative convergence curves of the PSO and PW-PSO algorithms provided by the embodiment of the present invention when solving the air combat decision-making problem. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] like Figure 1 As shown, an air combat decision method based on a velocity update strategy particle swarm algorithm of position weight comprises the following steps:

[0067] Step 1: Establish an air combat scenario model, simplify the complex air combat scenario, and establish a success rate model for our and enemy aircraft attacks and responses.

[0068] Step 1.1 Simplify the decision-making problem, simplify the three-dimensional air combat into two dimensions for simulation, use the hierarchical analysis method to determine the factors that affect the combat results, and build a multi-level structural model, such as Figure 2 As shown in Figure 1, the model includes the target layer, the criterion layer and the solution layer. The influencing factors of the target layer include the air combat capability index, the influencing factors of the criterion layer include the combat environment, aircraft flight performance, weapon system and aircraft hardware performance, and the influencing factors of the solution layer include the aircraft attack range D, aircraft detection range R and aircraft attack angle A. The aircraft parameter diagram is shown in Figure 1. Figure 3 shown.

[0069] Step 1.2 Set the conditions. Assume that each side has N aircraft, where N is a positive integer, representing the initial number of aircraft on each side. Each aircraft can only launch one attack in the set scenario. Define the damage success rate P o (i,j) and P e (i, j) represent the success rate of our i-th aircraft destroying the enemy j-th aircraft and the enemy i-th aircraft destroying our j-th aircraft, respectively. e (i,j) also represents the threat value of the enemy's i-th aircraft to our j-th aircraft. The decision matrix J is introduced o (i,j) and J e (i,j), where J o (i,j) and J e (i, j) are both 6×6 matrices, representing the decision results of our aircraft on the enemy aircraft and the decision results of the enemy aircraft on our aircraft respectively. The values ​​of the matrix elements are 0 or 1, where 0 means no attack and 1 means attack.

[0070] Step 1.3 Coordinated attack and escape probability, calculate the success probability of all aircraft coordinating to attack a certain enemy aircraft:

[0071]

[0072] Given the probability of our aircraft escaping successfully under enemy attack:

[0073]

[0074] Step 1.4 parameter setting, randomly generate the coordinates and directions of 2N aircraft within the specified scene range (for example, 1000x1000) according to the pseudo-random number generator, and the number and scene parameters can be flexibly set.

[0075] Step 2 establishes constraints based on the analytic hierarchy process.

[0076] Step 2.1: Construct the judgment matrix. Compare each factor at the same level in the criterion layer and the solution layer in pairs to understand its importance relative to the criterion of the previous layer. Use a 1-9 scale to quantitatively express the importance of factors and construct a judgment matrix.

[0077] Step 2.2 Calculation of relative weights of influencing factors: For the influencing factors in the criterion layer and the solution layer, use the square root method to calculate the normalized relative importance vector of each factor associated with its upper layer factor (e.g., the influencing factor of the solution layer, the aircraft attack range, is related to the influencing factors of the criterion layer, the combat environment, the aircraft flight performance, the weapon system, and the aircraft hardware performance. Then use the square root method to calculate the normalized relative importance vector of the aircraft attack range and the combat environment, the aircraft flight performance, the weapon system, and the aircraft hardware performance):

[0078]

[0079] Step 2.3 consistency test, verify the rationality and feasibility of the judgment matrix, use the two key indicators of consistency index CI and consistency ratio CR, calculate CI and CR respectively through the following formula, and verify whether the CR value is less than 0.1:.

[0080]

[0081] Among them, n represents the number of columns of the judgment matrix, λ max represents the maximum eigenvalue of the judgment matrix, RI represents the average value of the consistency standard of the random judgment matrix of the same order, and Table 1 shows the average random consistency index calculated 1000 times for matrices of order 1 to 14:

[0082] Table 1. Average random consistency index

[0083]

[0084] Step 2.4 Determination of attack success factors and weights: Use the AHP method to determine the factors that affect the success of the aircraft attack and the corresponding weights, that is, whether the factors in the solution layer have an impact on the target layer, and the weight of the impact.

[0085] Step 2.5 sets the attack conditions. According to the aircraft's current position and attitude information and the attack constraints set by the weapon system, the attack conditions are determined. The attack conditions include: the distance of the enemy aircraft relative to our aircraft is less than the detection range R of our aircraft, and the angle of the enemy aircraft relative to our aircraft is within the detection range of our aircraft. It is set that each aircraft can only attack one enemy aircraft in one attack, and the number of attacks launched by all our aircraft does not exceed N. When all these attack conditions are met, the aircraft launches an attack.

[0086] Step 3: Establish the objective function.

[0087] Step 3.1 Identify the primary mission: Remove the greatest threat to us from the target, i.e., ensure maximum damage to the enemy while minimizing our own losses.

[0088] Step 3.2 Establish the objective function to ensure that our task of causing the most damage to the enemy is

[0089] maxP o =α1·(1-D / R)+α2·J o (8)

[0090] Among them J o represents the decision matrix, α1 and α2 represent weights, which are set according to the actual situation;

[0091] Our tasks to ensure that we minimize our own losses are:

[0092] minP e =β1·(1-D / R)+β2·J e (9)

[0093] Among them J e represents the decision matrix, β1 and β2 represent weights, which are set according to actual conditions.

[0094] Step 3.3 Determine the target of attack: Determine the target of attack and decide the enemy aircraft to attack based on the threat level of the enemy aircraft to us, that is, the probability of the enemy's successful attack on us.

[0095] Step 3.4 Simplify the multi-objective function problem: Use the linear weighted sum method to simplify the two objective functions, set the weight coefficients according to the importance of the two objective functions, set the weight coefficient of objective function (8) to 0.6, and the weight coefficient of objective function (9) to 0.4, then Q = ω1·P o -ω2·P e (10).

[0096] Step 4: Particle swarm algorithm optimization.

[0097] Step 4.1 The basic idea of ​​the PSO algorithm is: create a randomly distributed particle population, each particle represents a potential solution to the problem, initialize the speed and position of each particle; update the speed and position of each particle according to formulas (11) and (12); calculate the fitness value of each particle, and update the individual optimal position P best and the global optimal position G best :

[0098] v i (t+1)=ωv i (t)+c1(P best (t)-x i (t))+c2(G best (t)-x i (t)) (11)

[0099] x i (t+1)=x i (t)+v i (t+1) (12)

[0100] Among them, v i (t) represents the velocity of the ith particle at time t, x i (t) represents the position of the i-th particle at time t, c1 and c2 represent learning factors, and ω represents the weight;

[0101] Step 4.2 The PW-PSO speed update strategy is: introduce the position weight parameter into the speed update formula to realize the intelligent adjustment of the particle speed; when the particle judgment is far from the optimal decision, increase the weight parameter to prompt the particle to accelerate to the optimal direction to quickly narrow the gap; when the decision is close to the optimal solution, reduce the weight and slow down the particle speed to make the search process more detailed and avoid missing the potential better solution near the optimal solution. According to the idea of ​​dynamically adjusting the speed and the distance from the optimal solution, the speed update formula is adjusted to formula (13):

[0102] v i (t+1)=ωv i (t)+(c1(P best (t)-x i (t))+c2(G best (t)-x i (t)))(G best (t)-x i (t))(13)

[0103] Step 5 is to solve the air combat decision based on the improved particle swarm algorithm.

[0104] Step 5.1 Initialize the air combat scene: randomly generate a group of particles, each particle represents a potential solution, that is, a group of numbers of aircraft that our aircraft want to attack; randomly initialize the position and velocity, randomly generate the initial position and velocity of each particle in the solution space, set the maximum number of iterations Gen, initialize the number of iterations to 0, and stop the iteration when the maximum number of iterations is reached. Figure 4 A schematic diagram showing the positions of aircraft during an air combat.

[0105] Step 5.2: Fitness evaluation: Calculate the fitness value of each particle according to the objective function Q of the problem.

[0106] Step 5.3 Update individual and global optimal positions:

[0107] Individual optimal position: For each particle, if its current fitness value is better than its historical best fitness value, its individual optimal position is updated;

[0108] Global optimal position: Among the individual optimal positions of all particles, the position with the best fitness value is selected as the global optimal position.

[0109] Step 5.4 Speed ​​update: Generate a random number r. If r is less than 0.6, use the speed update formula of PW-PSO. If r is greater than or equal to 0.6, use the speed update formula of the original particle swarm algorithm PSO.

[0110] Step 5.5 Position update: Update the position of the particle according to the position update formula.

[0111] Step 5.6 iterative check, the current iteration number increases by 1, if the current iteration number does not reach the maximum iteration number Gen, return to step 5.2 to continue iterating; otherwise, go to step 5.7.

[0112] Step 5.7 outputs the optimal solution: When the algorithm terminates, the individual with the highest fitness value is output, that is, the optimized attack decision.

[0113] Figure 5 It is a comparison diagram of iterative convergence curves when the air combat decision model is solved by the PSO algorithm and the PW-PSO algorithm proposed in the present invention. The results show that the convergence speed of the optimized PW-PSO algorithm is 56.34% higher than that of the traditional PSO algorithm. The algorithm described in the present invention can efficiently solve the model with fewer iterations, showing better convergence performance and decision-making efficiency.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An air combat decision-making method based on position weighted velocity updating particle swarm algorithm, characterized in that: The method comprises: Step S1: Establish an air combat scenario model, simplify the complex air combat scenario, and establish a success rate model for attack and response of our and enemy aircraft; Step S2: Based on the analytic hierarchy process (AHP), the constraint conditions are established and the index system in the air combat scenario is established and analyzed using the analytic hierarchy process (AHP) to obtain the influencing factors of the success of the aircraft attack and the corresponding weights; Step S3: Establishing an objective function, using the core idea of ​​maximizing the damage to the enemy and minimizing the damage to one's own side, calculating the success rate through the established success rate model, and simplifying the objective function through a linear weighted method; Step S4: Optimizing the particle swarm algorithm PSO, by adding position weights to the velocity update strategy of PSO to improve the efficiency and accuracy of the algorithm; Step S5: Based on the improved particle swarm algorithm PW-PSO, the decision problem of the air combat process is solved by using the particle swarm algorithm with the speed update strategy with position weight added to obtain the attack target of our aircraft.

2. The method according to claim 1, characterized in that: The step S1 further comprises: Step S11, simplifying the decision-making problem, simplifying the three-dimensional air combat into two-dimensional for simulation, using the AHP method to determine the factors affecting the combat results, and constructing a multi-level structural model, the model includes a target layer, a criterion layer and a solution layer, the influencing factors of the target layer include the air combat capability index, the influencing factors of the criterion layer include the combat environment, the aircraft flight performance, the weapon system and the aircraft hardware performance, and the influencing factors of the solution layer include the aircraft attack range D, the aircraft detection range R and the aircraft attack angle A; Step S12, setting conditions, assuming that each side has N aircraft, where N is a positive integer, representing the initial number of aircraft on each side, and each aircraft can only launch one attack in the set scenario, and defining the damage success rate P o (i,j) and P e (i, j) represent the success rate of our i-th aircraft destroying the enemy j-th aircraft and the enemy i-th aircraft destroying our j-th aircraft, respectively. e (i,j) also represents the threat value of the enemy's i-th aircraft to our j-th aircraft. The decision matrix J is introduced o (i,j) and J e (i, j), respectively represents the decision results of our aircraft on the enemy aircraft and the decision results of the enemy aircraft on our aircraft. The values ​​of the matrix elements are 0 or 1, where 0 means no attack and 1 means attack; Step S13, coordinated attack and escape probability, calculate the success probability of all our aircraft coordinated attack on a certain enemy aircraft as: Under enemy attack, the probability of our aircraft escaping successfully is: Step S14, parameter setting, randomly generating the coordinates and directions of 2N aircraft within the specified scene range according to the pseudo-random number generator.

3. The method according to claim 2, characterized in that The step S2 further comprises: Step S21, judgment matrix construction, compare each factor of the same level in the criterion layer and the solution layer in pairs, understand its importance relative to the criterion of the previous layer, use a 1-9 scale to quantitatively express the importance of factors, and construct a judgment matrix; Step S22, relative weight calculation of influencing factors. For the influencing factors in the criterion layer and the scheme layer, the normalized relative importance vector associated with each factor and its upper layer factor is calculated using the square root method: Among them, W i is the relative importance vector, a ij represents the data in the i-th row and j-th column of the judgment matrix, n represents the number of columns in the judgment matrix, W i 0 represents the normalized relative importance vector; Step S23, consistency check, verifies the rationality and feasibility of the judgment matrix, uses two key indicators, consistency index CI and consistency ratio CR, calculates CI and CR respectively by the following formula, and verifies whether the CR value is less than 0.1: Among them, n represents the number of columns of the judgment matrix, λ max represents the maximum eigenvalue of the judgment matrix, and RI represents the average value of the consistency criteria of random judgment matrices of the same order; Step S24, determining the attack success factors and weights, using the AHP method to determine the factors that affect the success of the aircraft attack and the corresponding weights, that is, whether the factors in the solution layer have an impact on the target layer, and the weight of the impact; Step S25, attack condition setting, determines the attack condition according to the current position and attitude information of the aircraft and the attack constraints set by the weapon system, the attack conditions include: the distance of the enemy aircraft relative to our aircraft is less than the detection range R of our aircraft, and the angle of the enemy aircraft relative to our aircraft is within the detection range of our aircraft, and it is set that each aircraft can only attack one enemy aircraft in one attack, and the number of attacks launched by all our aircraft does not exceed N; when all these attack conditions are met, the aircraft launches an attack.

4. The method according to claim 1, characterized in that: The step S3 further comprises: Step S31, determining the main task, removing the greatest threat to our side from the target, that is, ensuring that our side causes the greatest damage to the enemy while minimizing our own losses; Step S32, establish an objective function to ensure that our mission of causing maximum damage to the enemy is: maxP o =α1·(1-D / R)+α2·J o Among them J o represents the decision matrix, α1 and α2 represent weights, which are set according to the actual situation; Our tasks to ensure that we minimize our own losses are: minP e =β1·(1-D / R)+β2·J e Among them J e represents the decision matrix, β1 and β2 represent weights; Step S33, determining the target of attack, and determining the enemy aircraft to be attacked according to the threat level of the enemy aircraft to us, that is, the probability of the enemy's successful attack on us; Step S34, simplify the multi-objective function problem, set weight coefficients according to the importance of the two objective functions, use the linear weighted sum method to combine and simplify the two objective functions, and set the objective function P o The weight coefficient is 0.6, and the objective function P e The weight coefficient is 0.4: Q=ω1·P o -ω2·P e 。 5. The method according to claim 1, characterized in that The step S4 further comprises: Step S41, the PSO algorithm idea is: create a randomly distributed particle population, each particle represents a potential solution to the problem, initialize the speed and position of each particle, update the speed and position of each particle according to the following speed update formula and position update formula, calculate the fitness value of each particle, and update the individual optimal position P best and the global optimal position G best : v i (t+1)=ωv i (t)+c1(P best (t)-x i (t))+c2(G best (t)-x i (t)) x i (t+1)=x i (t)+v i (t+1) Among them, v i (t) represents the velocity of the ith particle at time t, x i (t) represents the position of the i-th particle at time t, c1 and c2 represent learning factors, and ω represents the weight; Step S42, the PW-PSO speed update strategy is: introduce the position weight parameter into the speed update formula. When the particle's judgment is far from the optimal decision, increase the weight parameter to accelerate the particle to move in the optimal direction to quickly narrow the gap. When the decision is close to the optimal solution, reduce the weight and slow down the particle speed. According to the idea of ​​dynamically adjusting the speed and the distance from the optimal solution, the speed update formula is adjusted as follows: v i (t+1)=ωv i (t)+(c1(P best (t)-x i (t))+c2(G best (t)-x i (t)))(G best (t)-x i (t))。 6. The method according to claim 5, characterized in that The step S5 further comprises: Step S51, air combat scene initialization, randomly generate a group of particles, each particle represents a potential solution, that is, a group of aircraft numbers that our aircraft want to attack, the position and speed are randomly initialized, the initial position and speed of each particle are randomly generated in the solution space, and the maximum number of iterations Gen is set. When the maximum number of iterations is reached, the iteration is stopped; Step S52, fitness evaluation, calculating the fitness value of each particle according to the objective function Q of the problem; Step S53, update individual and global optimal positions: Individual optimal position: For each particle, if its current fitness value is better than its historical best fitness value, its individual optimal position is updated; Global optimal position: Among the individual optimal positions of all particles, the position with the best fitness value is selected as the global optimal position; Step S54, speed update, generate a random number r, if r is less than 0.6, use the speed update formula of PW-PSO, if r is greater than or equal to 0.6, use the speed update formula of the original particle swarm algorithm PSO; Step S55, position update, updating the position of the particle according to the position update formula; Step S56, iterative check, the current iteration number increases by 1, if the current iteration number does not reach the maximum iteration number Gen, then return to step S52 to continue iterating; otherwise, enter step S57; Step S57, output the optimal solution. When the algorithm terminates, the individual with the highest fitness value is output, that is, the optimized attack decision.

Images