Intelligent optimization method for distributed unmanned aerial vehicle interference task distribution

Abstract

The invention discloses an intelligent optimization method for distributed unmanned aerial vehicle interference task distribution, and the method comprises the steps: setting a fitness function through calculating the total number of distributed unmanned aerial vehicles and the total interference suppression degree; a sparse rate controllable Boolean type particle swarm optimization algorithm and an actual value particle swarm optimization algorithm are combined to optimize a distributed unmanned aerial vehicle interference starting position and interference task distribution relation, a Pareto optimal theory is introduced to select an optimal solution, and comprehensive consideration of multiple factors is realized. According to the method, the multi-objective optimization problem is effectively solved, the distributed unmanned aerial vehicle interference tasks are reasonably distributed, a small distributed unmanned aerial vehicle total number and a high interference suppression total degree are obtained, and the method has important military significance and economic value.

Description

Technical Field

[0001] The present invention belongs to the field of electronic countermeasures, and in particular relates to an intelligent optimization method for distributed unmanned aerial vehicle interference task allocation. Background Art

[0002] With the continuous development of electronic countermeasure technology and scheduling methods, traditional "one-to-one" interference cannot achieve the expected effect, and the significance of coordinated interference becomes more and more prominent. Distributed UAVs have the characteristics of low cost and strong generation capability. The superior command post completes the allocation of distributed UAVs to radar interference tasks, finds a suitable allocation strategy for the target radar to achieve the best interference effect, maximizes the interference efficiency, and completes coordinated interference.

[0003] The distributed UAV jamming task allocation problem is a typical multi-objective optimization problem. Huang Jun et al. considered the jamming coverage and energy consumption, and solved the jamming task allocation problem based on a heuristic genetic algorithm. Ran Huaming et al. established a collaborative jamming task allocation model, and also used a genetic algorithm for optimization and solution. As an intelligent optimization algorithm, the particle swarm optimization algorithm has the characteristics of simple structure, few parameters involved, and fast convergence speed. It is widely used in solving nonlinear optimization problems. Li Jun et al. proposed an improved particle swarm optimization algorithm, which takes the total jamming utility as the objective function to optimize the jamming task allocation. In the existing research results on multi-jammer multi-objective jamming task allocation, most of them consider a single fitness function, do not comprehensively consider multiple optimization objectives, and only consider the task allocation relationship, without considering the number and location of distributed UAVs in the optimization problem. Hu Hongbo et al. used the sum of weighted coefficients to convert multiple objectives into a single fitness function, considered a variety of influencing factors, and used an immune genetic algorithm to solve the task allocation problem, but the optimization effect was greatly affected by the setting of the weighted coefficients.

[0004] From the above content, it can be seen that how to comprehensively consider the impact of multiple factors on the final solution and achieve comprehensive optimization of the relationship between the number of drones and task allocation is of great significance. Summary of the invention

[0005] The present invention proposes an intelligent optimization method for distributed UAV jamming task allocation, which solves the problem of effectively jamming the target radar network while keeping the number of distributed UAVs as small as possible, and realizes the allocation of distributed UAV jamming tasks.

[0006] The technical solution to realize the present invention is: an intelligent optimization method for distributed UAV jamming task allocation, the steps are as follows:

[0007] Step 1: Set the target radar parameters and the effective cross-sectional area of ​​the distributed UAVs, and calculate the initial maximum detection distance R of each target radarkmax , the target radar serial number k = 1, ..., K, and then calculate the detection coverage range Sq of the target radar network without interference max .

[0008] Step 2, establish a multi-objective particle swarm optimization algorithm, which consists of a sparse rate controllable Boolean particle swarm algorithm and a real-valued particle swarm algorithm; use the sparse rate controllable Boolean particle swarm algorithm to optimize the UAV interference startup position, and use the real-valued particle swarm optimization algorithm to optimize the UAV interference task allocation, that is, the particle position of the sparse rate controllable Boolean particle swarm algorithm is the position index matrix, and the particle position of the real-valued particle swarm optimization algorithm is the task allocation matrix; set the multi-objective particle swarm optimization algorithm parameters, and randomly initialize the position index matrix, task allocation matrix and corresponding particle speed.

[0009] Step 3: Determine the location where the distributed UAV interference is turned on according to the location index matrix, and determine the correspondence between the presence or absence of the distributed UAV and the target radar according to the task allocation matrix.

[0010] Step 4: Calculate the detection coverage range Sq of the target radar network after interference according to the maximum detection distance of the target radar after interference, and calculate two fitness functions for distributed UAVs: the total number of distributed UAVs and the total degree of interference suppression, and obtain the distributed UAV interference task allocation optimization model.

[0011] Step 5: The two fitness functions are weightedly summed into a single fitness function, and the local optimal solution of the particle is updated according to the single fitness function.

[0012] Step 6: Select the global optimal solution of particles based on Pareto optimality theory.

[0013] Step 7: Update the particle position and particle velocity based on the sparse rate controllable Boolean particle swarm algorithm and the real-valued particle swarm algorithm.

[0014] Step 8: Determine whether the end condition is met. If not, return to step 3; otherwise, terminate the optimization and obtain the optimal archive, that is, the optimal Pareto frontier.

[0015] Step 9: According to the selection criteria, the optimal particles are selected from the optimal archive to determine the UAV jammer startup position and allocate jammer tasks.

[0016] Compared with the prior art, the present invention has the following significant advantages:

[0017] (1) The sparse rate controllable Boolean particle swarm optimization algorithm and the real-valued particle swarm optimization algorithm are combined to simultaneously optimize the location of distributed UAV interference startup and the allocation of interference tasks.

[0018] (2) The Pareto optimality theory is introduced to effectively solve the multi-objective optimization problem and obtain fewer distributed drones and a greater degree of interference suppression. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of the target radar and distributed drone locations.

[0020] Figure 2 This is a flow chart of an intelligent optimization method for distributed UAV jamming task allocation proposed by the present invention. DETAILED DESCRIPTION

[0021] 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.

[0022] The following will further introduce the specific implementation method, as well as the technical difficulties and inventive points of this invention in combination with this design example.

[0023] The present invention provides an intelligent optimization method for distributed UAV interference task allocation, which takes the total number of distributed UAVs and the total degree of interference suppression as optimization targets, combines a real-valued particle swarm optimization algorithm and a sparse rate controllable Boolean particle swarm optimization algorithm with a multi-objective particle swarm optimization algorithm to optimize the distributed UAV interference startup position and the interference task allocation relationship.

[0024] The maximum number of distributed drones is L. On a rectangular plane with the coordinate origin as the reference point and containing M×N unit points, the spacing between adjacent units is d c , randomly place K target radars, K≤M×NL, to form a target radar network. The schematic diagram of the target radar and distributed UAV positions is shown in Figure 1 As shown in the flow chart of the method Figure 2 As shown, the specific steps are as follows:

[0025] Step 1: Calculate the initial maximum detection distance R of the target radar based on the target radar parameters and the effective cross-sectional area of ​​the distributed UAV kmax , k = 1, ..., K, according to the target radar detection distance, the detection coverage range Sq of the target radar network without interference can be calculated max .

[0026] Step 2: Establish a multi-objective particle swarm optimization algorithm, use the sparse rate controllable Boolean particle swarm algorithm to optimize the UAV interference startup position, and use the real-valued particle swarm optimization algorithm to optimize the UAV interference task allocation. That is, the particle position of the sparse rate controllable Boolean particle swarm algorithm is the position index matrix, and the particle position of the real-valued particle swarm optimization algorithm is the task allocation matrix. Set the multi-objective particle swarm optimization algorithm parameters, the number of particle groups is group, the number of iterations is G, the archive size is 1.5×group, the number of optimization targets is 2, and the number of grids in each dimension is s. Randomly generate a distributed UAV interference startup position index matrix And the corresponding particle speed in is a set of natural numbers, is a real number set, the element id in the mth row and nth column of the position index matrix ID m,n ∈{0,1}, set the unit point id of the kth target radar k = 0. Randomly generate interference task allocation matrix The corresponding particle velocity vector Maximum particle velocity v max =K+1, the value range of the elements in the task allocation matrix AI is [0, K+1].

[0027] Step 3: Determine the location of the distributed UAV interference startup according to the location index matrix ID, and determine the presence or absence of the distributed UAV and its corresponding relationship with the target radar according to the task allocation matrix AI, as follows:

[0028] Step 3.1: Use random mutation to control the total number of "1" in the position index matrix ID, that is, count the number of "1" in the position index matrix, denoted as N on If N on If the maximum number of drones L is greater than the specified number, the position index matrix N on -L elements that are "1" change from "1" to "0", so that the number of array elements N on is equal to the maximum number of drones specified. On the contrary, if N on If the number of drones is less than the specified maximum number, the LN on The elements that are "0" change from "0" to "1", making N on =L, thereby adjusting the position index matrix. The element with a value of "1" in the position index matrix W means that there is a distributed UAV interfering with the startup at that location;

[0029] Step 3.2: Assign the lth element ai in the matrix AI according to the task l The value range of ai determines the presence or absence of the lth distributed UAV and the corresponding target radar: if 0≤ai l<1, the lth distributed UAV does not work, that is, there is no distributed UAV; if 1≤ai l <2, then the lth distributed UAV interferes with the first target radar; if K≤ai l ≤K+1, the lth distributed UAV interferes with the Kth target radar.

[0030] Step 4: Calculate the detection coverage range Sq of the target radar network after interference according to the maximum detection distance of the target radar after interference, and calculate two fitness functions for distributed UAVs: the total number of distributed UAVs and the total degree of interference suppression, and obtain the distributed UAV interference task allocation optimization model, as follows:

[0031] Step 4.1: Calculate the distance x between the distributed UAV and its corresponding target radar lk , x lk The subscript of indicates that the lth UAV interferes with the kth target radar. The specific calculation formula is as follows:

[0032]

[0033] Among them (m l ,n l ) is the position of the lth UAV, (m k ,n k ) is the position of the kth target radar. According to the distance between the kth target radar and the UAV interfering with it and the relevant parameters of the distributed UAV, the maximum detection distance R of the kth target radar after interference is calculated. k , k = 1,…,K, and calculate the detection coverage range Sq after the target radar network is interfered.

[0034] Step 4.2: Calculate the total number of distributed drones TN and the total degree of interference suppression P all , the total number of distributed drones TN is the total number of values ​​not less than 1 in the task allocation matrix AI, and the total degree of interference suppression P all As shown below:

[0035]

[0036] Step 4.3: Consider the total number of distributed UAVs TN and the total degree of interference suppression P all To optimize the goal, the optimization model of distributed UAV jamming task allocation can be expressed as:

[0037]

[0038] Step 5: The two fitness functions are weighted and summed into a single fitness function. Let the single fitness function be f = α·TN + β·P all, α and β are weight parameters, and the particle local optimal solution AI is updated according to a single fitness function pb 、ID pb .

[0039] Step 6: Select the global optimal solution based on Pareto optimality theory, as follows:

[0040] Step 6.1: According to the fitness function TN and P all Compare the current particle with the particles in the archive, and select the Pareto optimal solution to add to the archive.

[0041] Step 6.2: Constrain the size of the archive to ensure that the number of particles in the archive does not exceed the limit value. Sort the particles in the archive from large to small according to the single fitness function f. If the total number of particles in the current archive exceeds the limit value by Q, delete the first Q particles with larger f to maintain a stable archive size.

[0042] Step 6.3: Based on the adaptive grid, the multi-objective optimization problem archive target space is divided into two parts according to the grid number s. 2 Divide it into sub-regions of equal size, select the region with the least congestion, that is, the sub-region with the least number of particles, and arbitrarily select a particle as the leader, which is the global optimal solution AI gb 、ID gb .

[0043] Step 7: Update the particle position and particle velocity based on the sparse rate controllable Boolean particle swarm algorithm and the real-valued particle swarm algorithm, as follows:

[0044] Step 7.1: Use the sparse rate controllable Boolean particle swarm optimization algorithm to update the position index matrix ID and its corresponding particle velocity matrix V id , the particle position and particle velocity update formula of sparse rate controllable Boolean particle swarm optimization algorithm is as follows:

[0045]

[0046] Where W 1 , and are Boolean values ​​between 0 and 1, and their length is equivalent to the particle dimension and is the learning factor, represents the influence on the particle approaching the individual optimal solution, It is expressed as the influence on the particle approaching the global optimal solution, W 1 is the inertia weight. is the position index matrix of the i-th particle in the t-th iteration, i.e., the particle position of the sparse rate controllable Boolean particle swarm optimization algorithm, is the particle velocity of the ith particle in the tth iteration, represents the local optimal solution of the tth iteration, represents the global optimal solution of the tth iteration. is the particle position of the ith particle in the t+1th iteration, that is, the updated particle position, It is the particle velocity of the i-th particle at the t+1-th iteration, that is, the updated particle velocity.

[0047] Step 7.2: Use the real-valued particle swarm optimization algorithm to update the interference task allocation matrix AI and its corresponding particle velocity matrix V ai , the real-valued particle swarm optimization algorithm particle position and particle velocity update formula is as follows

[0048]

[0049] Among them, W 2 , and are all real values, rand is a random number in the interval [0,1], usually learning factors and The values ​​in are all 2.05. 2 The value w in 2 As the number of iterations decreases from 0.9 to 0.4. is the task assignment matrix of the ith particle in the tth iteration, i.e., the particle position of the real-valued particle swarm optimization algorithm, is the particle velocity of the ith particle in the tth iteration, represents the local optimal solution of the tth iteration, represents the global optimal solution of the tth iteration. is the particle position of the ith particle in the t+1th iteration, that is, the updated particle position, It is the particle velocity of the i-th particle at the t+1-th iteration, that is, the updated particle velocity.

[0050] When the updated position of a particle is not within the constraint range, boundary processing is required to ensure that the particle position is within the constraint range. Common boundary processing methods include absorbing boundaries.

[0051] Step 8: Determine whether the end condition is met. If so, terminate the optimization and obtain the optimal archive, that is, the optimal Pareto frontier. If not, return to step 3.

[0052] Step 9: According to the selection criteria, the particle with the smallest single fitness function value f in the optimal archive is selected as the optimal particle, and the optimal solution ID of the position index matrix and the task allocation matrix is ​​obtained. opt and AI opt, thereby determining the distributed UAV jammer startup position and jammer task allocation relationship.

[0053] In summary, the present invention innovatively proposes an intelligent optimization method for distributed UAV interference task allocation. The method optimizes the distributed UAV interference startup position and interference task allocation based on a multi-objective particle swarm optimization algorithm, and combines the sparse rate controllable Boolean particle swarm optimization algorithm and the real-valued particle swarm optimization algorithm to optimize the position index and task allocation based on the Pareto optimal theory to obtain the optimal solution. The influence of multiple factors on the optimal solution is comprehensively considered to obtain a lower total number of distributed UAVs and a larger total degree of interference suppression. The total number of distributed UAVs decreases by more than 20% relative to the maximum number of distributed UAVs, and the total degree of interference suppression reaches more than 50%.

Claims

1. An intelligent optimization method for distributed UAV jamming task allocation, characterized in that: Here are the steps: Step 1: Set the target radar parameters and the effective cross-sectional area of ​​the distributed UAVs, and calculate the initial maximum detection distance R of each target radar kmax , the target radar serial number k = 1, ..., K, and then calculate the detection coverage range Sq of the target radar network without interference max ; Step 2, establish a multi-objective particle swarm optimization algorithm, which consists of a sparse rate controllable Boolean particle swarm algorithm and a real-valued particle swarm algorithm; use the sparse rate controllable Boolean particle swarm algorithm to optimize the UAV interference startup position, and use the real-valued particle swarm optimization algorithm to optimize the UAV interference task allocation, that is, the particle position of the sparse rate controllable Boolean particle swarm algorithm is the position index matrix, and the particle position of the real-valued particle swarm optimization algorithm is the task allocation matrix; set the parameters of the multi-objective particle swarm optimization algorithm, and randomly initialize the position index matrix, the task allocation matrix and the corresponding particle speed; Step 3: Determine the location of the distributed UAV interference startup according to the location index matrix, and determine the correspondence between the presence or absence of the distributed UAV and the target radar according to the task allocation matrix; Step 4: Calculate the detection coverage range Sq of the target radar network after being interfered according to the maximum detection distance of the target radar after being interfered, and calculate two fitness functions for distributed UAVs: the total number of distributed UAVs and the total degree of interference suppression, and obtain the distributed UAV interference task allocation optimization model; Step 5: The two fitness functions are weightedly summed into a single fitness function, and the local optimal solution of the particle is updated according to the single fitness function; Step 6: Select the global optimal solution of particles based on Pareto optimal theory; Step 7: Update the particle position and particle velocity based on the sparse rate controllable Boolean particle swarm algorithm and the real-valued particle swarm algorithm; Step 8: Determine whether the end condition is met. If not, return to step 3; otherwise, terminate the optimization and obtain the optimal archive, that is, the optimal Pareto frontier; Step 9: According to the selection criteria, the optimal particles are selected from the optimal archive to determine the UAV jammer startup position and allocate jammer tasks.

2. The intelligent optimization method for distributed UAV jamming task allocation according to claim 1 is characterized in that: Assume that M×N unit points are evenly divided in the rectangular plane, the reference point is the coordinate origin, the maximum number of distributed drones is L, and the spacing between adjacent units on the rectangular plane containing M×N unit points with the coordinate origin as the reference point is d c , randomly place K target radars, K≤M×NL, to form a target radar network.

3. The intelligent optimization method for distributed UAV jamming task allocation according to claim 2 is characterized in that: In step 2, a multi-objective particle swarm optimization algorithm is established, as follows: Set the parameters of the multi-objective particle swarm optimization algorithm, the number of particle groups is group, the number of iterations is G, the archive size is 1.5×group, the number of optimization targets is 2, and the number of grids in each dimension is s; randomly generate the distributed UAV interference startup position index matrix And the corresponding particle speed in is a set of natural numbers, is a real number set, the element id in the mth row and nth column of the position index matrix ID m,n ∈{0,1}, set the unit point id of the kth target radar k =0; Randomly generate interference task allocation matrix The corresponding particle velocity vector Maximum particle velocity v max =K+1, the value range of the elements in the task allocation matrix AI is [0, K+1].

4. The intelligent optimization method for distributed UAV jamming task allocation according to claim 3 is characterized in that: In step 3, the location of the distributed UAV interference startup is determined according to the location index matrix, and the correspondence between the presence or absence of the distributed UAV and the target radar is determined according to the task allocation matrix, as follows: Step 3.1: Use random mutation to control the total number of "1" in the position index matrix ID, that is, count the number of "1" in the position index matrix, denoted as N on ; if N on If the maximum number of drones L is greater than the specified number, the position index matrix N on -L elements that are "1" change from "1" to "0", so that the number of array elements N on Equal to the maximum number of drones specified; On the contrary, if N on If the number of drones is less than the specified maximum number, the LN on The elements that are "0" change from "0" to "1", making N on =L, thereby adjusting the position index matrix. The element with a value of "1" in the position index matrix W indicates that there is a distributed UAV interfering with the startup at that location; Step 3.2: Assign the lth element ai in the matrix AI according to the task l The value range of ai determines the presence or absence of the lth distributed UAV and the corresponding target radar: if 0≤ai l <1, the lth distributed UAV does not work, that is, there is no distributed UAV; if 1≤ai l <2, then the lth distributed UAV interferes with the first target radar; if K≤ai l ≤K+1, the lth distributed UAV interferes with the Kth target radar.

5. The intelligent optimization method for distributed UAV jamming task allocation according to claim 4 is characterized in that: In step 4, the detection coverage range Sq of the target radar network after interference is calculated according to the maximum detection distance of the target radar after interference, and two fitness functions for distributed UAVs are calculated: the total number of distributed UAVs and the total degree of interference suppression, and the distributed UAV interference task allocation optimization model is obtained, which is as follows: Step 4.1: Calculate the distance x between the distributed UAV and its corresponding target radar lk , the specific calculation formula is as follows: Among them (m l ,n l ) is the position of the lth UAV, (m k ,n k ) is the position of the kth target radar; According to the distance between the kth target radar and the UAV that interferes with it and the relevant parameters of the distributed UAV, the maximum detection distance R of the kth target radar after interference is calculated. k , k = 1, ..., K, and calculate the detection coverage range Sq of the target radar network after being interfered; Step 4.2: Calculate the total number of distributed drones TN and the total degree of interference suppression P all , the total number of distributed drones TN is the total number of values ​​not less than 1 in the task allocation matrix AI, and the total degree of interference suppression P all As shown below: Sq max Indicates the detection coverage of the target radar network without interference; Step 4.3: Consider the total number of distributed UAVs TN and the total degree of interference suppression P all To optimize the goal, the optimization model of distributed UAV jamming task allocation is expressed as:

6. The intelligent optimization method for distributed UAV jamming task allocation according to claim 5 is characterized in that: In step 5, the two fitness functions are weighted and summed into a single fitness function, and the local optimal solution of the particle is updated according to the single fitness function, as follows: Assume that the single fitness function is f = α·TN + β·P all , α and β are weight parameters, and the particle local optimal solution AI is updated according to a single fitness function pb 、ID pb .

7. The intelligent optimization method for distributed UAV jamming task allocation according to claim 6 is characterized in that: In step 6, the global optimal solution of particles is selected based on the Pareto optimality theory, as follows: Step 6.1: According to the fitness function TN and P all Compare the current particle with the particles in the archive, and select the Pareto optimal solution to add to the archive; Step 6.2: Constrain the size of the archive to ensure that the number of particles in the archive does not exceed the limit value. Sort the particles in the archive from large to small according to the single fitness function f. If the total number of particles in the current archive exceeds the limit value by Q, delete the first Q particles with larger f, so as to maintain a stable archive size. Step 6.3: Based on the adaptive grid, the multi-objective optimization problem archive target space is divided into two parts according to the grid number s. 2 Divide it into sub-regions of equal size, select the region with the least congestion, that is, the sub-region with the least number of particles, and arbitrarily select a particle as the leader, which is the global optimal solution AI gb 、ID gb .

8. The intelligent optimization method for distributed UAV jamming task allocation according to claim 7 is characterized in that: In step 7, the particle position and particle velocity are updated based on the sparse rate controllable Boolean particle swarm algorithm and the real-valued particle swarm algorithm, as follows: Step 7.1: Use the sparse rate controllable Boolean particle swarm optimization algorithm to update the position index matrix ID and its corresponding particle velocity matrix V id , the particle position and particle velocity update formula of sparse rate controllable Boolean particle swarm optimization algorithm is as follows: Among them, W 1 , and are Boolean values ​​of 0 and 1, and their length is equivalent to the particle dimension. and is the learning factor, represents the influence on the particle approaching the individual optimal solution, It is expressed as the influence on the particle approaching the global optimal solution, W 1 is the inertia weight; is the position index matrix of the i-th particle in the t-th iteration, i.e., the particle position of the sparse rate controllable Boolean particle swarm optimization algorithm, is the particle velocity of the ith particle in the tth iteration, represents the local optimal solution of the tth iteration, represents the global optimal solution of the tth iteration; is the particle position of the ith particle in the t+1th iteration, that is, the updated particle position, is the particle velocity of the i-th particle in the t+1-th iteration, that is, the updated particle velocity; Step 7.2: Use the real-valued particle swarm optimization algorithm to update the interference task allocation matrix AI and its corresponding particle velocity matrix V ai , the real-valued particle swarm optimization algorithm particle position and particle velocity update formula is as follows: Among them, W 2 , and are all real values, rand is a random number in the interval [0,1], usually learning factors and The values ​​in are all 2.05; the inertia weight W 2 The value w in 2 As the number of iterations decreases from 0.9 to 0.4; is the task assignment matrix of the ith particle in the tth iteration, i.e., the particle position of the real-valued particle swarm optimization algorithm, is the particle velocity of the ith particle in the tth iteration, represents the local optimal solution of the tth iteration, represents the global optimal solution of the tth iteration; is the particle position of the ith particle in the t+1th iteration, that is, the updated particle position, is the particle velocity of the i-th particle in the t+1-th iteration, that is, the updated particle velocity; When the updated position of a particle is not within the constraint range, boundary processing is required to ensure that the particle position is within the constraint range. Common boundary processing methods include absorbing boundaries.

9. The intelligent optimization method for distributed UAV jamming task allocation according to claim 8 is characterized in that: In step 9, according to the selection criteria, the optimal particles are selected from the optimal archive to determine the UAV jammer startup position and allocate jammer tasks, as follows: According to the selection criteria, the particle with the smallest single fitness function value f in the optimal archive is selected as the optimal particle, and the optimal solution ID of the position index matrix and the task allocation matrix is ​​obtained. opt and AI opt , thereby determining the distributed UAV jammer startup position and jammer task allocation relationship.

Images