Optimization method for site deployment based on star-space joint positioning

By optimizing the deployment of UAVs using a star-space joint positioning method, combined with geometric accuracy factor and genetic particle swarm optimization algorithm, the position distribution of UAVs is optimized, solving the problems of low positioning accuracy, slow speed and complex structure in existing technologies, and achieving efficient real-time positioning of dynamic targets.

CN117433527BActive Publication Date: 2026-11-10XIDIAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311208686.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-19
Publication Date
2026-11-10
Estimated Expiration
2043-09-19

AI Technical Summary

Technical Problem

Existing positioning technologies suffer from limitations in multi-satellite cooperative positioning and UAV swarm cooperative positioning, including limitations in common-view conditions, weak reconnaissance signal energy, slow positioning speed, complex structure, and unsuitability for dynamic scenarios.

Method used

A site deployment optimization method based on star-space joint positioning is adopted, which combines geometric accuracy factor analysis, particle swarm optimization, and genetic algorithm to optimize the location distribution of UAVs. The UAV site deployment optimization is achieved through the genetic particle swarm optimization algorithm.

Benefits of technology

It improves the real-time positioning accuracy of dynamic targets, shortens the running time, reduces input parameters, simplifies the algorithm structure, expands the optimization range, and overcomes the limitations of existing technologies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117433527B_ABST
    Figure CN117433527B_ABST
Patent Text Reader

Abstract

This invention discloses a site deployment optimization method based on joint satellite-space positioning, mainly addressing the problems of slow positioning speed, complex structure, numerous input parameters, and inability to achieve real-time target positioning in existing technologies. The implementation scheme is as follows: A joint satellite-space positioning site deployment scenario is set up, along with primary and secondary UAV observation stations; the geometric accuracy factor of the UAV's initial position is calculated based on the deployment scenario and the primary UAV observation station; constraints are set according to the deployment scenario; an objective function model is constructed based on the deployment scenario and the geometric accuracy factor; and a genetic particle swarm optimization algorithm is used to move the UAV detection stations in real-time to find the optimal deployment position based on the constraints and the objective function model. This invention shortens the running time, reduces input parameter and collaborative positioning errors, and improves positioning speed and collaborative positioning accuracy. Simulation results show that the geometric accuracy factor of the optimized UAV detection station position is 45% higher than that of existing technologies, and it can be used for the position distribution of UAV reconnaissance platforms.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of radar technology, and further relates to a site deployment optimization method, which can be used for the location distribution of UAV reconnaissance platforms to improve the cooperative positioning accuracy between UAVs. Background Technology

[0002] Currently, common positioning methods include multi-satellite cooperative positioning and UAV swarm cooperative positioning, but both methods have certain drawbacks. For example, multi-satellite cooperative positioning requires common-line-of-sight conditions to achieve target positioning, and the energy of reconnaissance signals is often weak, which is not conducive to the measurement of time difference parameters. On the other hand, UAV swarm cooperative positioning has difficulty capturing targets with rapidly changing antenna pointing, resulting in incomplete reconnaissance information from the swarm.

[0003] Zeng Hui et al. disclosed a three-station time difference positioning station layout optimization algorithm in their paper number 1001-893X(2010)05-0018-05. It is used to solve for the optimal location. By using the horizontal positioning accuracy factor as the measurement criterion for positioning accuracy, the optimal station layout location is solved by traversing the established multi-objective programming mathematical model. However, this invention has the problem of long running time.

[0004] Liu Hongjian et al. disclosed a method for optimizing the deployment of measurement and control equipment based on an adaptive genetic algorithm in their paper number 1673-6338(2020)03-0325-06. The method is used to improve the crossover probability based on the genetic algorithm and optimize the best deployment location based on the error on the three-dimensional coordinate axis. However, this invention has the problem of slow positioning speed.

[0005] In his master's thesis, Xia Wei disclosed a method for deploying a multi-station passive time difference positioning system. Based on the shortcomings of the gray wolf optimization algorithm, he proposed an improved hybrid gray wolf optimization algorithm and analyzed its process for solving the optimal deployment and obtaining the optimal deployment location. However, this method has the problems of complex structure and a large number of input parameters.

[0006] In their paper number 1671-4598(2018)01-0300-04, Tang Yu et al. disclosed a centralized multi-sensor optimization deployment method based on the geometric dilution of target positioning accuracy. The method combines several typical multi-sensor deployment methods to calculate and analyze the geometric dilution of target positioning accuracy under different fusion estimation methods and different deployment methods, and determines the optimal deployment method through simulation. However, this method is only applicable to static scenarios and not to dynamic scenarios.

[0007] Grigorios G. Anagnostopoulos et al. disclosed a multi-objective optimization method for tuning indoor positioning systems in their paper with DOI number 10.1109 / IPIN.2017.8115908. This method is used for training and learning with real data recorded on site, but it is not suitable for scenarios requiring real-time target localization. Summary of the Invention

[0008] The purpose of this invention is to address the shortcomings of the existing technologies mentioned above by proposing a satellite-based joint positioning optimization method to shorten the operation time, reduce input parameters, improve positioning speed, and achieve real-time positioning of targets in motion scenarios.

[0009] The technical solution of this invention is as follows: setting up a deployment scenario; calculating the geometrical precision factor (GDOP) of the initial position of the UAV; setting algorithm constraints; constructing an objective function model; using a particle swarm optimization algorithm to deploy the UAV; and using a genetic particle swarm optimization algorithm to deploy the UAV. The implementation steps include the following:

[0010] (1) Set up a space-air joint optimization deployment scenario that includes multiple UAV observation stations, reconnaissance satellites and phased array radar targets;

[0011] (2) In the scenario of station deployment, one UAV is the main observation station and the rest are secondary observation stations. The position information of the phased array radar target is forwarded to the main UAV observation station through the reconnaissance satellite. The main UAV observation station calculates the geometric accuracy factor G of the UAV at the initial position based on the position information.

[0012] (3) Set constraints on the location distribution of UAVs according to the deployment scenario;

[0013] (4) Construct an objective function model for the UAV position distribution based on the deployment scenario and geometric accuracy factor G;

[0014] (5) Based on the constraints of the UAV location distribution and the objective function model, the genetic algorithm and the particle swarm algorithm are combined to form the genetic particle swarm algorithm. The genetic particle swarm algorithm is used to optimize the UAV deployment, that is, to find the optimal deployment location based on the real-time mobile UAV.

[0015] Compared with the prior art, the present invention has the following advantages:

[0016] First, by using geometric precision factors for error analysis, this invention improves the accuracy of station deployment and solves the problem of real-time positioning of dynamic targets.

[0017] Secondly, by using the particle swarm optimization algorithm, this invention reduces the number of input parameters, simplifies the algorithm structure, and solves the problems of numerous input parameters and complex structures in existing algorithms.

[0018] Third, by using a genetic algorithm, this invention expands the search range and solves the problem that existing algorithms are easily limited to local optima. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the implementation of the present invention;

[0020] Figure 2 This is a diagram of the star-sky joint station deployment scenario constructed in this invention;

[0021] Figure 3 This is a schematic diagram of the weight allocation of the objective function model in this invention;

[0022] Figure 4 The simulation graphs show the fitness evolution curves of the present invention and existing particle swarm optimization algorithms.

[0023] Figure 5 This is a diagram showing the geometric accuracy factor before and after station placement for the present invention and existing particle swarm algorithms.

[0024] Figure 6 This is a comparison chart of the fitness results of the present invention and existing particle swarm optimization algorithms. Detailed Implementation

[0025] The embodiments and effects of the present invention will be described in further detail below with reference to the accompanying drawings.

[0026] Reference Figure 1 The implementation steps of the present invention include the following:

[0027] Step 1: Set up the site deployment scenario.

[0028] Reference Figure 2 The space-air collaborative optimization deployment scenario includes a swarm of drones, a reconnaissance satellite, and a target phased array radar. The drone swarm contains multiple drone observation stations. An XYZ coordinate system is established for the optimized deployment scenario to set the coordinate positions of the drone observation stations and the target phased array radar. After the reconnaissance satellite calculates the coordinate position of the target phased array radar, it transmits the coordinate position information to the drone observation stations, which then complete the deployment of the drone observation stations at the current position of the target phased array radar based on the position information.

[0029] Step 2: Calculate the geometric accuracy factor G of the initial position of the UAV.

[0030] 2.1) Based on the optimized station deployment scenario, assume that the UAV swarm contains three UAV observation stations, one of which is the main UAV observation station with coordinates u1 = [x1, y1, z1]. T The other two are auxiliary observation stations for unmanned aerial vehicles (UAVs), with coordinates u2 = [x2, y2, z2]. T and u3=[x3,y3,z3] TThe coordinates of the target phased array radar are s = [x, y, z]. T ;

[0031] 2.2) Based on the optimized deployment scenario, calculate the distances R1, R2, and R3 from the three UAV observation stations to the target phased array radar:

[0032]

[0033] Where x1 is the length of the first UAV observation station on the x-axis, x2 is the length of the second UAV observation station on the x-axis, x3 is the length of the third UAV observation station on the x-axis, y1 is the length of the first UAV observation station on the y-axis, y2 is the length of the second UAV observation station on the y-axis, and y3 is the length of the third UAV observation station on the y-axis.

[0034] 2.3) Calculate the i-th distance R based on the distances R1, R2, and R3 from the three UAV observation stations to the target phased array radar. i The lengths D on the x-axis ix And the length D on the y-axis iy The measurement error S of the i-th UAV observation station i :

[0035]

[0036]

[0037] S i =D ix dx i +D iy dy i

[0038] Where, x i Let y be the length of the i-th UAV observation station on the x-axis. i Let be the length of the i-th UAV observation station on the y-axis, where i = 1, 2, 3;

[0039] 2.4) Based on the i-th distance R i The lengths D on the x-axis ix And the length D on the y-axis iy The measurement error S of the i-th UAV observation station i The total differential formula dR is obtained by calculation. i :

[0040] dR i =D ix dx+D iy dy+S i

[0041] 2.5) Apply the total differential formula dR i Rewritten as the total differential formula for distance difference dΔR i1 :

[0042] dΔR i1 =(D ix -D 1x )dx+(D iy -D 1y )dy+(S i -S1)

[0043] Among them, D 1x Let D be the length of the first distance R1 on the x-axis. 1y Let R1 be the length of the first distance on the y-axis, and S1 be the measurement error of the first UAV observation station;

[0044] 2.6) According to the total differential formula for distance difference dΔR i1 Calculate the matrix D between the three distances R1, R2, and R3, and the measurement error matrix dS between the three UAV observation stations:

[0045]

[0046] dS = [(S2-S1), (S3-S1)] T

[0047] Among them, D 2x D is the length of the second distance R2 on the x-axis. 3x Let D be the length of the third distance R3 on the x-axis. 2y D is the length of the second distance R2 on the y-axis. 3y S2 is the measurement error of the second UAV observation station, and S3 is the measurement error of the third UAV observation station.

[0048] 2.7) Based on the matrix D between the three distances and the measurement error matrix dS between the three UAV observation stations, the total differential formula dΔR is obtained. i1 The matrix form dΔR:

[0049] dΔR=DdU+dS

[0050] in,

[0051] 2.8) The target position estimation error dU is calculated from the matrix form dΔR:

[0052] dU=(D T D) -1 D T (dΔR-dS)

[0053] Among them, D T As the transpose of matrix D, the target's position estimation error dU is related to the matrix form dΔR and the position measurement error dS of each observation station itself;

[0054] 2.9) Calculate the standard deviation matrix E[dR×dR] of the measurement errors between observation stations from the target position estimation error dU. T The standard deviation matrix E[dS×dS] between the measurement errors of the observation station on the x-axis, y-axis, and z-axis and the standard deviation matrix of the observation station on the x-axis, y-axis, and z-axis. T ]:

[0055]

[0056]

[0057] Where η is the correlation coefficient between measurement errors, σ ΔR Let σ be the standard deviation of measurement errors among all observation stations, and let σ be the station placement errors in the x, y, and z directions, which are independent and identically distributed. s 2 =σ x 2 =σ y 2 =σ z 2 The standard deviation of the measurement error of each observation station on the x-axis, y-axis, and z-axis;

[0058] 2.10) The standard deviation matrix between measurement errors of the observation stations is E[dR×dR]. T The standard deviation matrix E[dS×dS] between the measurement errors of the observation station on the x-axis, y-axis, and z-axis and the standard deviation matrix of the observation station on the x-axis, y-axis, and z-axis. T The covariance matrix P of the radiation target positioning error was calculated. dR :

[0059]

[0060] Where G = (D T D) -1 D T ;

[0061] 2.11) From the covariance matrix P dR The estimated values ​​of the radiation target localization error σ on the x-axis and y-axis are calculated. x 2 and σ y 2 :

[0062]

[0063] Among them, P dR (1,1) is the covariance matrix P when x=1, y=1.dR The value of P dR (2,2) is the covariance matrix P when x=2, y=2. dR The value;

[0064] 2.12) Estimated value of the radiation target positioning error σ x 2 and σ y 2 The geometric accuracy factor G of the initial position of the UAV observation station was calculated as follows:

[0065]

[0066] Where tr(·) is the trace of the matrix. From the geometric precision factor expression, the main factors affecting positioning accuracy are: the position measurement error of each observation station itself, the time difference of arrival measurement error, and the layout of the observation stations.

[0067] Step 3: Based on the deployment scenario, set constraints on the location distribution of UAV detection stations.

[0068] In the deployment scenario, to reduce mutual interference between the three UAV observation stations, the distance between any two stations must be greater than or equal to 2 km; to improve the cooperative positioning capability among the UAV observation stations, the distance between any two stations must be less than or equal to 15 km; furthermore, to locate the phased array radar position of a target in any direction, the three UAV observation stations cannot be in a straight line. Based on these three requirements, the constraints are set as follows:

[0069]

[0070] in, For it to exist, R i and R j Let x1 be the distance from two different UAV observation stations to the target phased array radar, x2 be the length of the first UAV observation station on the x-axis, x3 be the length of the third UAV observation station on the x-axis, y1 be the length of the first UAV observation station on the y-axis, y2 be the length of the second UAV observation station on the y-axis, and y3 be the length of the third UAV observation station on the y-axis.

[0071] Step 4: Based on the station deployment scenario and the geometric accuracy factor G of the initial position of the UAV observation station, construct an objective function model for the UAV position distribution.

[0072] 4.1) Take N target position points centered on the target phased array radar, substitute the coordinates of these target position points into the geometric accuracy factor G formula for the initial position of the UAV observation station in step 2, and calculate the geometric accuracy factor G of the k-th target position point near the target phased array radar. k k = 1, 2, ..., N;

[0073] 4.2) Based on the distance between the k-th target location and the target phased array radar, set the weight ω of the k-th target location. k That is, when the distance between the target location and the target phased array radar is less than 1km, a weight ω is set. k The range is (0.6, 0.95); when the distance between the target location and the target phased array radar is greater than 1km, a weight ω is set. k The range is (0.35, 0.6), such as Figure 3 As shown;

[0074] 4.3) The geometric precision factor G of the k-th target location point k The weight ω of the k-th target location point k Construct an objective function model f for the location distribution of UAVs:

[0075]

[0076] When the objective function f is minimized, the corresponding location of the UAV observation station is the optimal location for UAV deployment.

[0077] Step 5: Based on the constraints of the UAV location distribution and the objective function model, the genetic algorithm and particle swarm algorithm are combined to form a genetic particle swarm algorithm, which is then used to optimize the UAV deployment.

[0078] 5.1) Set the relevant parameters for the genetic particle swarm optimization algorithm:

[0079] The location of the UAV observation station is limited to the constraints in step 3, and the average particle fitness value is set to T = f;

[0080] Let the population size be P and the maximum number of iterations be N. max The particle's position x(0) and velocity v(0);

[0081] Let P be the crossover probability factor in the genetic algorithm. c and the probability factor of variation P m ;

[0082] 5.2) Based on the initialized particle velocity v(0) and position x(0), the optimal value p of each individual particle is obtained by using the particle swarm optimization algorithm within the maximum number of iterations. i ' dThe particle position corresponding to the optimal value of each individual is (x i ,y i ), i = 1, 2, ..., P;

[0083] 5.3) Determine the difference, T, between the current particle's fitness value and the average fitness value of all particles:

[0084] If the current particle fitness value is greater than the average particle fitness value T, and the current particle random value is greater than the mutation probability factor P. m If so, the mutation operation of the genetic algorithm is executed to increase the particle placement range;

[0085] If the current particle fitness value is greater than the average particle fitness value T, and the current particle random value is greater than the crossover probability factor P. c If so, the crossover operation of the genetic algorithm is executed to expand the particle optimization range;

[0086] If the current particle fitness value is less than the average particle fitness value T, then the global optimum value p' is obtained. gd ;

[0087] 5.4) Within the particle placement range and optimization range, based on the optimal value p' of each particle... id The global optimum p' is calculated using the particle swarm optimization algorithm. gd :

[0088] p' gd =min(p' id )

[0089] 5.5) From the global optimum p' gd Find the optimal position (x, y) of the corresponding particle:

[0090] Let the global optimum be p' gd The optimal value for the nth individual is given by the particle position (x). n ,y n ),but

[0091] (x,y)=(x n ,y n )

[0092] Where n∈[1,P];

[0093] 5.6) Deploy UAVs based on the optimal position (x,y), that is, find the optimal deployment position based on the real-time movement of the UAVs.

[0094] The numbers used in the above steps are only for the purpose of more clearly indicating the implementation scheme of the present invention, and their order is not limited.

[0095] The effects of the present invention will be further described below with reference to simulation experiments.

[0096] 1. Simulation experimental conditions:

[0097] The hardware platform for the simulation experiment of this invention is: CPU is Intel Core i7-7700, RAM is 8GB.

[0098] The software platform for the simulation experiments of this invention is Windows 10 operating system and Matlab R2019a.

[0099] The target phased array radar is located at (150, 150). The three UAV observation stations are located approximately 200 kilometers away from the target phased array radar. Let the positions of the three UAV observation stations be (0, 0), (-6, 8), and (6, 8), respectively, in kilometers.

[0100] The aforementioned location information is assigned to the particles, and station placement optimization is performed using both the existing particle swarm optimization (PSO) algorithm and the genetic particle swarm optimization (GSO) algorithm. The population size is set to 15, the maximum number of iterations to 2000, the inertia weight ω = 0.8 for the PSO algorithm, and the upper and lower bounds of the inertia weight for the GSO algorithm are ωGSO and ωGSO, respectively. max =0.95 and ω min =0.35, the two learning factors are set to the same size c1=c2=2, the particle velocity range is [-1,1], the time error of each UAV reconnaissance station is 10ns, and the station location error is 0.01km.

[0101] 2. Simulation content and results:

[0102] Simulation 1: Under the above simulation experimental conditions, the target localization velocity was simulated using the present invention and existing particle swarm optimization algorithms. The results are as follows: Figure 4 As shown, where:

[0103] Figure 4 (a) shows the fitness curve of the existing particle swarm optimization algorithm;

[0104] Figure 4 (b) shows the fitness curve of the genetic particle swarm algorithm of this invention;

[0105] from Figure 4 As can be seen from (a), with the continuous iteration of the particle swarm algorithm, the existing particle swarm algorithm can only find the best placement location after the 668th generation.

[0106] from Figure 4 As can be seen in (b), the genetic particle swarm algorithm of the present invention can find the optimal placement location after 150 generations, which is faster than the existing particle swarm algorithm that finds the optimal placement location after 668 generations.

[0107] Simulation 2: Under the above simulation conditions, the UAV observation station was deployed using the present invention and existing particle swarm optimization algorithm. The geometric accuracy factor values ​​before and after deployment were simulated, and the results are as follows. Figure 5 As shown, where:

[0108] Figure 5 (a) is a diagram of the geometric accuracy factor before the deployment of the UAV observation station;

[0109] Figure 5 (b) is a diagram of the geometric accuracy factor of the UAV observation station after deployment using the existing particle swarm algorithm;

[0110] Figure 5 (c) is a diagram of the geometric accuracy factor of the UAV observation station after being deployed using the genetic particle swarm algorithm of this invention.

[0111] from Figure 5 As can be seen in (a), the geometric accuracy factor value of the UAV observation station before its deployment is greater than 20, indicating very low positioning accuracy.

[0112] from Figure 5 As can be seen in (b), the geometric precision factor of the UAV observation station is approximately 11 after deployment using the existing particle swarm algorithm.

[0113] from Figure 5 As can be seen in (c), the geometric accuracy factor of the UAV observation station after deployment using the genetic particle swarm algorithm of this invention is about 6, which is less than the geometric accuracy factor of the UAV observation station after deployment using the existing particle swarm algorithm, thus improving the positioning accuracy.

[0114] Simulation 3: Under the above simulation experimental conditions, the optimization capability of the particle swarm optimization algorithm in this invention and existing methods for target localization was simulated. The results are as follows: Figure 6 As shown;

[0115] from Figure 6 As can be seen from the results of 50 experimental tests, the average fitness curve of the existing particle swarm optimization algorithm fluctuates greatly, while the average fitness curve of the genetic particle swarm optimization algorithm of this invention fluctuates less, indicating that the optimization effect of this invention is more stable.

[0116] The simulation results above show that the present invention can accelerate the positioning speed, improve the positioning accuracy, and improve the optimization effect.

[0117] The above description is merely a specific example of the present invention and does not constitute any limitation on the present invention. Obviously, those skilled in the art, after understanding the content and principles of the present invention, may make various modifications and changes in form and details without departing from the principles and structure of the present invention. However, these modifications and changes based on the ideas of the present invention are still within the scope of protection of the claims of the present invention.

Claims

1. A site deployment optimization method based on star-space joint positioning, characterized in that, Including the following: (1) Set up a space-air joint optimization deployment scenario that includes multiple UAV observation stations, reconnaissance satellites and phased array radar targets; (2) In the scenario of station deployment, one UAV is the main observation station and the others are secondary observation stations. The position information of the phased array radar target is relayed to the main UAV observation station through a reconnaissance satellite. The main UAV observation station calculates the geometric precision factor of the UAV at the initial position based on the position information. The implementation steps include the following: (2a) Establish a coordinate system based on the deployment scenario. Assume the scenario contains three UAV observation stations, one of which is the main UAV observation station, and its coordinates are: The other two are unmanned aerial vehicle (UAV) auxiliary observation stations, with coordinates as follows: and The coordinates of the target phased array radar are ; (2b) Based on the optimized UAV deployment scenario, calculate the distances from the three UAV observation stations to the phased array radar target. ; (2c) Based on the distances from the three UAV observation stations to the target phased array radar Calculate the first distance exist Length on the axis , No. distance exist Length on the axis and the Measurement error of UAV observation station The total differential formula is obtained. : ; in, ; (2d) Apply the total differential formula Rewritten as the total differential formula for distance difference : ; in, The first distance exist Length on the axis, The first distance exist Length on the axis, This represents the measurement error of the first UAV observation station; (2e) According to the total differential formula of distance difference Calculate the matrix between the three distances Measurement error matrix between the three UAV observation stations The total differential formula is obtained. matrix form : ; in, ; (2f) in matrix form The target position estimation error was calculated. : ; in, For matrix transpose; (2g) Target position estimation error The covariance matrix of the radiation target positioning error is obtained. : (2h) From the covariance matrix Calculated at shaft and Estimated positioning error of radiation target on axis and : ; in, for Time covariance matrix The value, for Time covariance matrix The value; (2i) Estimated value of the positioning error of the radiation target and The geometric accuracy factor of the initial position of the UAV was calculated. : ; in, To obtain the trace of the matrix; (3) Set constraints on the location distribution of UAVs according to the deployment scenario; (4) Based on the site layout scenario and geometric precision factor Construct an objective function model for the location distribution of UAVs; (5) Based on the constraints of the UAV location distribution and the objective function model, the genetic algorithm and the particle swarm algorithm are combined to form the genetic particle swarm algorithm. The genetic particle swarm algorithm is used to optimize the UAV deployment, that is, to find the optimal deployment location based on the real-time mobile UAV.

2. The method according to claim 1, characterized in that, In step (2b), the distances from the three UAV observation stations to the phased array radar target are calculated. The calculation formula is as follows: ; in, For the first drone observation station Length on the axis For the second drone observation station Length on the axis For the third drone observation station Length on the axis For the first drone observation station Length on the axis, For the second drone observation station Length on the axis For the third drone observation station The length on the axis.

3. The method according to claim 1, characterized in that, In step (2c), the calculation of the first distance exist Length on the axis , No. distance exist Length on the axis and the Measurement error of UAV observation station The calculation formula is as follows: ; ; ; in, For the first A drone in Length on the axis, For the first A drone in Length on the axis, .

4. The method according to claim 1, characterized in that, In step (2e), the matrix between the three distances is calculated. Measurement error matrix between the three UAV observation stations The calculation formula is as follows: ; ; in, The second distance exist Length on the axis, The third distance exist Length on the axis, The second distance exist Length on the axis, The third distance exist Length on the axis, The measurement error is due to the second UAV observation station. This refers to the measurement error of the third UAV observation station.

5. The method according to claim 1, characterized in that, The steps to implement step (2g) include the following: (2g1) Calculate the standard deviation matrix between measurement errors of the observation stations. and observation station axis, axis, Standard deviation matrix of measurement errors on the axis : ; ; in, The correlation coefficient between measurement errors. The standard deviation of measurement errors between each observation station. For each observation station axis, axis, Standard deviation of measurement error on the shaft; (2g2) is the standard deviation matrix of measurement errors between observation stations. and observation station axis, axis, Standard deviation matrix of measurement errors on the axis The covariance matrix of the radiation target positioning error is calculated. : ; in, .

6. According to the method of claim 1, in step (3), the constraints on the location distribution of the UAV are set, and the formula is as follows: ; in, For existence, and The distances from two different drones to the phased array radar target. For the first drone observation station Length on the axis, For the second drone observation station Length on the axis, For the third drone observation station Length on the axis, For the first drone observation station Length on the axis, For the second drone observation station Length on the axis, For the third drone observation station The length on the axis.

7. The method according to claim 1, wherein the objective function model for the location distribution of the UAV in step (4) includes the following steps: (4a) by geometric precision factor Get the first Geometric precision factor value of each target location point ; (4b) Allocation of the first Weight of each target location point ; (4c) by the first Geometric precision factor value of each target location point and the Weight of each target location point Construct an objective function model for the location distribution of UAVs : ; in, The number of target location points.

8. According to the method of claim 1, in step (5), the genetic algorithm and the particle swarm algorithm are mixed to form a genetic particle swarm algorithm, and the genetic particle swarm algorithm is used to deploy UAVs. The implementation steps include the following: (5a) Set the average particle fitness value Population size Maximum number of iterations The position of the particles particle velocity Crossover probability factor in genetic algorithms and the probability factor of variation ; (5b) Based on the initialized particle velocity and location Within the maximum number of iterations, the particle swarm optimization algorithm is used to find the optimal value for each individual. The particle position corresponding to the optimal value for each individual is: , ; (5c) Determine the difference between the current particle's fitness value and the average fitness value of all particles. : If the current particle fitness value is greater than the average particle fitness value. And the current particle random value is greater than the mutation probability factor. If so, the mutation operation of the genetic algorithm is executed to increase the particle placement range; If the current particle fitness value is greater than the average particle fitness value And the current particle random value is greater than the crossover probability factor. If so, the crossover operation of the genetic algorithm is executed to expand the particle optimization range; If the current particle fitness value is less than the average particle fitness value Then the global optimal value is obtained. ; (5d) Within the particle placement range and optimization range, based on the optimal value of each particle... The global optimum is obtained by calculating the particle swarm optimization algorithm. : ; (5e) From the global optimum Find the optimal position of the corresponding particle : ; (5f) Based on the optimal position Deploy drone stations.