A simplified particle swarm weapon target assignment method with priority

Abstract

The application provides a simplified particle swarm weapon target distribution method with priority, and belongs to the field of resource distribution. First, weapon and target information of a combat unit is acquired, and priority of the combat unit to a strike target is calculated. Then, a distribution method is selected according to the priority, if the priority is less than a threshold value, a simplified particle swarm method is adopted, if the priority is greater than the threshold value, weapon distribution is directly performed according to threat degrees of the strike targets in sequence. The application determines the priority of the strike target, so that the strike target with higher threat is preferentially distributed with the weapon, and the threat degree of the combat unit is reduced. The application adopts the simplified particle swarm algorithm, and the convergence time of the algorithm can be shortened, which is of great significance for efficient strike of targets in cooperative air combat.

Description

Technical Field

[0001] This invention belongs to the field of resource allocation and relates to a simplified target allocation method for particle swarm weapons with priority. Background Technology

[0002] The weapon target allocation problem is to find a suitable weapon-target allocation strategy that maximizes the expected damage to enemy targets or minimizes the expected loss of friendly forces. From a time-factor perspective, the weapon target allocation problem has two variations: a static allocation problem, where all weapons are fired within the same time period, and the states of the weapons and targets are known and fixed; and a dynamic allocation problem, where weapons are fired at different times, and the states of the weapons and targets are variable.

[0003] In practical application models, factors such as newly arriving targets or damage to friendly combat unit weapon systems necessitate real-time updates to the original weapon target allocation strategy. Therefore, considering dynamic weapon target allocation—judging the threat level of enemy targets based on real-time situational awareness and then allocating weapon quantities and types accordingly—is crucial for improving weapon strike efficiency in actual combat. Allocating targets based on their azimuth, distance, and threat level, while considering different weapon types, to establish a weapon target allocation model that more closely resembles the air combat environment, and designing a weapon target allocation method for this model, is a challenging research undertaking.

[0004] Current research focuses on dynamic weapon target allocation, considering the impact of factors such as the destruction of aerial targets, damage to weapon systems, or the arrival of new aerial targets on the initial weapon allocation strategy. By maximizing the survival probability during the defense process, i.e., maximizing the stability of weapons during the defense process, the strategy allocation is updated, making the weapon target allocation model more closely resemble the real air combat environment. However, further analysis of the priority of attack targets has more practical application significance for improving the strike efficiency of combat units. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a simplified particle swarm weapon target allocation method with priority. The overall objective is to minimize the threat posed by targets and maximize the value of attacking them. A hard constraint is applied to the number of weapons allocated to each target, establishing a weapon target allocation model that more realistically approximates the air combat environment. The priority of air targets is determined based on their value and threat coefficient. Then, targets are prioritized based on priority. For targets with higher priority, weapons are directly allocated; for targets with lower priority, a simplified particle swarm method is used, thereby improving the efficiency of weapon allocation.

[0006] To achieve the objective, the present invention adopts the following technical solution:

[0007] A simplified target allocation method for particle swarm weapon with priority includes the following steps:

[0008] Step 1: Determine the priority of targets;

[0009] 1.1) First, calculate the probability of damage to target k by combat unit i:

[0010]

[0011] Where ω1,ω2∈[0,1] are the weighting factors of the angle between the combat unit and the target and the distance between the combat unit and the target, respectively; f(d) represents the expected probability of weapon r striking in combat unit i (a known term); ik f(θ) represents the distance influence parameter between combat unit i and target k; ik ) represents the angle influence parameter between combat unit i and target k.

[0012] f(d ik The calculation is as follows: Assume that the maximum strike range of weapon r in combat unit i is denoted as d. max (Given information) When the distance d between combat unit i and target k is... ik Within the weapon's firing range, the probability of weapon r damaging target k decreases as the distance between combat unit i and target k increases. When the distance between combat unit i and target k exceeds the maximum firing range, the probability of weapon r damaging target k is 0. Therefore, the distance between weapon r and target k in combat unit i affects the parameter f(d). ik )for:

[0013]

[0014] f(θ ik The calculation is as follows: Given that the optimal angle between the weapon and the target is 40°, the parameter f(θ) affecting the angle between weapon r and target k in combat unit i is... ik The curve satisfies a normal distribution with a mean of 40 and a standard deviation of 40.

[0015]

[0016] Where θ represents the actual angle between weapon r in combat unit i and target k.

[0017] 1.2) Then, calculate the strike priority of the current combat unit i and each strike target j = 1, 2...m, and sort the strike priorities by size, denoted as p. i={δ i1 ,δ i2 ,....δ im}, p i Priority δ between medium combat unit i and target k ik The calculation formula is:

[0018]

[0019] Where α1, α2 ∈ [0, 1] are weighting factors, v o The value coefficient for attacking target k. The threat coefficient for attacking target k.

[0020] The calculation is as follows:

[0021]

[0022] Where ω1ω2...ω s These are the factors influencing the threat level of the combat unit, namely, the angle of the target toward the friendly combat unit, the distance to the friendly combat unit, the relative altitude of the two sides, and the target's speed or other parameters. o The value of striking targets is obtained by assigning scores to each combat unit in the combat simulation environment.

[0023] Step 2: Set priority p i Categorize the values ​​in the data;

[0024] Given a priority threshold β, if p i There exists a priority δ iz If p is greater than β, add the target z to list L; if p i There exists a priority δ ig If the value is less than β, add the target g to list L′;

[0025] Step 3: For the targets selected in L, first, a threshold η is given for the number of weapons to be assigned to each target; the probability of each combat unit hitting and damaging the current target k is calculated sequentially and sorted in descending order of probability value. Then, the combat unit with the highest probability of hitting and damaging the target is selected and one of its weapons is randomly assigned; next, the combat unit with the second lowest probability of hitting and damaging the target is selected and one of its weapons is randomly assigned, and so on; when the number of weapons for the current target meets η, weapon allocation stops.

[0026] Step 4: For the targets selected in L′, a simplified particle swarm optimization algorithm is used to allocate weapons.

[0027] Step 4.1: Initialize parameters based on individual information, including the population size N and the individual's position;

[0028] Step 4.2: Calculate the fitness function of the individual. The individual fitness function H is:

[0029]

[0030] Where j = 1, ..., m represents the target. The threat coefficient of the target, i = 1, ... l represents the combat unit, p ij x represents the probability of damage to target j by combat unit i. ij It is a 0,1 Boolean variable, x ij =1 indicates that the weapon in combat unit i is assigned to target j, x ij =0 Weapons in combat unit i are not assigned to target j. v j Indicates the value of striking target j It is represented as the negative value of the total value of all objectives.

[0031] Step 4.3: Update the individual's local and global optima according to the fitness function H, and then update the individual's position according to formula (7):

[0032]

[0033] in, Let p be the position of individual q at time k+1; best Let g be the current optimal position of individual q. best ω represents the optimal position of the entire group; ω is a weighting factor, representing the degree of individual movement speed. A larger ω indicates a stronger global convergence ability of the algorithm, while a smaller ω indicates a stronger local convergence ability of the algorithm; r1, r2 ∈ (0, 1) are the direction parameters of individual movement to increase the diversity of individual movement directions; c1, c2 are the learning factors of the individual, which are the adjustment step sizes for the individual's optimal position and the global optimal solution.

[0034] Step 4.4: For formula (7) Determine the position value and record it. The current value has x-coordinate u and y-coordinate h. Determine... The range of the first digit after the decimal point is denoted as follows: Let num be the first digit after the decimal point of the value. If num belongs to [0, 5], randomly select a number τ from (0, h) and let u be the x-coordinate. , The position value with ordinate τ is Let the original x-coordinate be u and the y-coordinate be h. The value is 0; let M be the total number of targets hit. If num belongs to (5,9], randomly select a number χ from (h,M), and let the position value be u on the x-coordinate and χ on the y-coordinate. Let the original x-coordinate be u and the y-coordinate be h. The value is 0.

[0035] Step 4.5: Determine if the maximum number of iterations has been reached. If it has, proceed to step 4.6; otherwise, return to step 4.3.

[0036] Step 4.6: Return the allocation result.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] (1) By determining the priority of the strike targets, the present invention prioritizes the allocation of weapons to the strike targets with higher threats, thereby reducing the degree of threat to the combat units.

[0039] (2) The present invention adopts a simplified particle swarm algorithm, which shortens the convergence time of the algorithm, which is of great significance for efficient target strike in cooperative air combat. Attached Figure Description

[0040] Figure 1 The flowchart designed for this invention;

[0041] Figure 2 Matching weapon targets for m combat units and n targets;

[0042] Figure 3 The process of searching for the optimal solution for PS-PSO. Detailed Implementation

[0043] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0044] This invention implements a priority-based simplified particle swarm weapon target allocation method, constructed according to the system structure diagram shown in the figure. The method includes acquiring weapon and target information of the combat unit, calculating the priority of the targets, and then selecting an allocation method based on the priority. If the priority is less than a threshold β, the simplified particle swarm method is used; if the priority is greater than β, weapons are directly allocated according to the threat level of the targets. The process is as follows: Figure 1 As shown, it includes the following steps:

[0045] The first step is to determine the priority of the targets.

[0046] Assume the Red team's forces are organized as shown in the table below: 10 MiG-29 fighter jets (30 points), 4 Ka-2 early warning helicopters (6 points each); the Blue team's forces consist of 18 F-16 fighter jets (60 points each), 2 EC-130H electronic warfare aircraft (70 points each), and 1 E-2K early warning aircraft (160 points each). The experiment is set up with the Red team as friendly and the Blue team as the enemy.

[0047] The experiment was conducted using a set of parameter configurations in an air combat scenario. The friendly forces consisted of 7 combat units, denoted as M = {M1, M2, ..., M7}, and the enemy had 6 target sets, T = {t1, t2, ..., t6}. The 7 combat units carried two types of weapons, the performance parameters of which are shown in Table 2. The number of different weapon types was N = {[2,4],[2,4],[2,4],[2,4],[2,4],[2,4],[2,4]}. Weapons were allocated by default based on their type order.

[0048] In this example, all selected enemy aircraft have the same value, so they are all set to 1. The particle encoding method is matrix encoding, where weapon types are sorted by quantity from smallest to largest. The current weapon types and performance parameters are shown in Table 1.

[0049] Table 1 Weapon Types and Performance Parameters ↓

[0050] Tab.1Weapontype and performance parameters

[0051]

[0052] For the current target, parameters such as the target's orientation, distance, and relative altitude of both sides are obtained through electronic equipment such as radar, and the threat level of the target to the combat unit is calculated. As shown in Table 2:

[0053] Table 2 Threat coefficient of targets to combat units ↓

[0054] Tab.2The threat factor of the target to the combat unit

[0055]

[0056] Meanwhile, the probability of damage to the target by the current combat units M1, M2, ... M7 is determined as shown in Table 3:

[0057] Table 3 Probability of Combat Units Hitting Targets ↓

[0058] Tab.3The probability of hitting the target by the combat unit

[0059]

[0060] The probability of weapon r of combat unit i damaging target k is p. ij This allows us to determine the expected threat level of target j, based on the formula... α1 = 0.35, α2 = 0.25, and the priority P of each combat unit for the target is shown in Table 4:

[0061] Table 4. Priority of targets by combat units

[0062] Tab.4The probability of hitting the target by the combat unit

[0063]

[0064] As shown in Table 4, the priority of combat unit M1 for attacking targets is t1, t2, t3, t4, t5, t6; the priority of combat unit M2 for attacking targets is t2, t1, t3, t4, t6, t5; the priority of combat unit M3 for attacking targets is t4, t3, t5, t2, t1, t6; the priority of combat unit M4 for attacking targets is t5, t1, t6, t3, t2, t4; the priority of combat unit M5 for attacking targets is t5, t6, t4, t3, t2, t1; the priority of combat unit M6 for attacking targets is t3, t5, t4, t6, t1, t2; and the priority of combat unit M7 for attacking targets is t4, t5, t6, t3, t2, t1.

[0065] The second step is to classify the values ​​in priority P;

[0066] Assuming a priority threshold β = 92%, and since there are no targets in P with a priority greater than β, a simplified particle swarm optimization algorithm is directly adopted.

[0067] The third step involves using a simplified particle swarm optimization algorithm to match weapons with selected targets.

[0068] First, based on the individual's fitness function Calculate the individual fitness H, and update the individual's local optimum p according to the fitness function. best and the global optimal g best Then update the individual position according to the following formula:

[0069]

[0070] The experimental setup included r1 = 0.6, r2 = 0.3, ω = 0.8, a particle learning factor of 2, a maximum number of iterations n = 100, a population size of 70 particles, and a maximum of η = 6 weapons assigned to each target. The experiment was repeated 100 times. The optimal solution obtained by the algorithm is:

[0071]

[0072] The relationship between the fitness function obtained by the algorithm and the number of iterations is as follows: Figure 3 As shown.

[0073] The above-described embodiments are merely illustrative of the implementation methods of the present invention, but should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the protection scope of the present invention.

Claims

1. A simplified particle swarm weapon target assignment method with priority, characterized in that, Includes the following steps: Step 1: Determine the priority of targets; 1.1) Calculate combat units Against the target of attack Probability of damage of attack: （1）； in, These are the weighting factors for the angle between the combat unit and the target, and the distance between the combat unit and the target, respectively. For combat units Chinese weapons The expected probability of a hit; For combat units With the target Distance-affected parameters; For combat units With the target The included angle affects the parameter; 1.2) Calculate the current combat unit With each target The attack priority is determined and sorted by priority, denoted as follows: , Medium combat unit With the target Priority between The calculation formula is: （4）； in, As a weighting factor, To strike the target The value coefficient, To strike the target The threat level; Step 2: Categorize each value in the priority among the values; Given a priority threshold ,like Priority exists in Greater than to strike this target Add to list ;like Priority exists in Less than to strike this target Add to list ; Step 3: Targeting The selected targets are first given a threshold number of weapons to be assigned to each target. According to the target of each combat unit The probability of damage from each strike is calculated sequentially and sorted from highest to lowest probability value. Then, the combat unit with the highest probability of damage is selected and randomly assigned one of its weapons. Next, the combat unit with the second lowest probability of damage is selected and randomly assigned a weapon, and this process is repeated until the number of weapons available for the current target satisfies the condition... If so, then the distribution of weapons will cease; Step 4: Targeting The selected targets are then used to allocate weapons using a simplified particle swarm optimization algorithm. Step 4 is described in detail below: Step 4.1: Initialize parameters according to individual's information, number of population individuals , initialize individual's position; Step 4.2: Calculate the fitness function of the individual; wherein the fitness function of the individual is f(x) = 1 - (x - 1)2 （6）； in, Indicate the target of the attack. The threat level of the target. Indicates combat unit, Represents the calculation of combat units Targets The probability of damage from a hit, It is a 0,1 Boolean variable. Indicates combat unit Medium weapons were assigned to strike targets superior, combat unit Medium weapons are not assigned to strike targets superior; Indicate the target value This is represented as the negative value of the total value of all objectives; Step 4.3: Update the individual local optimum and global optimum, then update the individual position according to formula (7): Update the individual local optimum and global optimum, then update the individual position according to formula (7): （7）； in, For individuals exist The position at that moment; For individuals Current optimal position This is the optimal position for the entire group; This is a weighting factor, representing the degree of an individual's movement speed. A larger value indicates that the algorithm has a stronger global convergence ability. A smaller value indicates that the algorithm has a stronger local convergence ability; To increase the diversity of individual movement directions; , where is the learning factor for the individual, and is the adjustment step size for the individual's optimal position and the global optimal solution; Step 4.4: For formula (7) Determine the position value and record it. The x-coordinate of the current value is The vertical axis is ,judge The range of the first digit after the decimal point is denoted as follows: The first digit after the decimal point of the value is ,like belong ,exist Randomly select a number Let the x-coordinate be The vertical axis is Position value Let the original x-coordinate be . The vertical axis is of The value is 0; record For the total number of targets, if belong ,exist Randomly select a number Let the x-coordinate be The vertical axis is Position value Let the original x-coordinate be . The vertical axis is of The value is 0; Step 4.5: Determine if the maximum number of iterations has been reached. If it has, proceed to step 4.6; otherwise, return to step 4.

3. Step 4.6: Return the allocation result.

2. The simplified particle swarm weapon target assignment method with priority according to claim 1, characterized in that, In step 1.1) The calculation is as follows: Assuming the combat unit Weapons exist in China The maximum strike distance is denoted as (Given information) When the combat unit With the target distance Within the weapon's firing range, the weapon Targets The probability of damage varies with the combat unit With the target The distance increases and the size decreases, when the combat unit With the target When the distance is greater than the maximum attack range, the weapon... Targets The probability of damage is 0, therefore the combat unit Chinese weapons With the target Distance influence parameters for: （2）。 3. The simplified particle swarm weapon target assignment method with priority according to claim 1, characterized in that, In step 1.1), The calculation is as follows: Given that the optimal angle of attack between the weapon and the target is... Therefore, combat units Chinese weapons With the target The included angle affects the parameters A normal curve with a mean of 40 and a standard deviation of 40: （3）。 4. The simplified particle swarm weapon target assignment method with priority according to claim 1, characterized in that, In step 1.2) above, is calculated as follows: （5）； in, These are the factors influencing the threat level of the combat unit, namely, the angle of the target toward the friendly combat unit, the distance to the friendly combat unit, the relative altitude of the two sides, and the target speed or other parameters. The value of striking targets is obtained by assigning scores to each combat unit in the combat simulation environment.

Images