Geomagnetic matching method based on improved data group intelligent search strategy optimization

Abstract

The geomagnetic matching navigation method, based on an improved data group intelligent search strategy optimization, comprises the following steps: 1. Reading the position data information of the integrated navigation system, the navigation error of the vehicle, and the detection data measurement values ​​of the geomagnetic sensor; 2. Extracting the geomagnetic values ​​of the matching area from the geomagnetic reference map for trajectory matching; 3. Based on the Grey Wolf Algorithm theory, the variables and parameters of the data group intelligent search strategy optimization algorithm are redesigned, and the relevant definitions and parameters are rewritten to make it suitable for geomagnetic matching. 4. Using the improved matching algorithm to perform geomagnetic matching, the optimal geomagnetic navigation trajectory can be obtained. The present invention utilizes three types of motion behaviors of the data group to achieve a closer approach to the optimal solution. This algorithm can obtain the desired optimal trajectory, thereby improving the accuracy of geomagnetic navigation matching and, in turn, the positioning accuracy of the integrated navigation system.

Description

Technical Field

[0001] The present invention belongs to the technical field of geomagnetic navigation matching algorithms, and in particular is a geomagnetic matching method based on optimized intelligent search strategy of improved data group. Background Art

[0002] In the early days, before rapid technological advancement, humans primarily relied on nearby reference objects and their own speed and time to determine their current location. In modern times, with the development of the economy and society, the rapid advancement of science and technology, the increasing sophistication of transportation, and the proliferation of diverse means of transportation, humans can no longer rely on prior knowledge to determine their location, relying instead on various navigation systems. Consequently, the need for navigation has become increasingly important. Simultaneously, the rapid advancement of key technologies such as computers, communications systems, and information processing has enabled humans to employ electronic, mechanical, acoustic, optical, and magnetic methods to determine parameters related to the location of moving objects for positioning and navigation. Navigation and positioning technology has diversified, and the demand for navigation technology has expanded from surface navigation, rivers, and lakes to high altitudes and even the deep sea. Furthermore, the performance requirements for navigation systems have become increasingly demanding.

[0003] Current navigation systems generally include satellite navigation, inertial navigation, terrain matching navigation, celestial navigation, gravity matching navigation, and geomagnetic matching navigation. Satellite navigation and inertial navigation are the most widely used and common navigation systems. While satellite navigation systems offer high accuracy and coverage, their satellite signals are weak and their interference resistance is poor, so navigation information is often blocked under certain circumstances. Inertial navigation systems primarily use two inertial devices, gyroscopes and accelerometers, to achieve navigation by observing the ship's acceleration and angular velocity. While they perform well in terms of passivity, they can generate cumulative errors, necessitating the use of other auxiliary navigation methods for correction. Terrain matching navigation systems rely on terrain measurements for navigation, but they must account for various factors, including topography, climate, and seasonal variations. Celestial navigation refers to navigation that determines position and orientation through observation of natural celestial bodies. While it can provide position data and even attitude information, it can also suffer from issues such as signal discontinuity and interference from climatic conditions. The technical methods of geomagnetic and gravity matching navigation are very similar. Geomagnetism mainly relies on calculating geomagnetic characteristics, while gravity achieves matching navigation by measuring the magnitude of gravity. The two navigation systems can refer to each other, but the gravimeters used in the gravity navigation system are very expensive, which has a certain inhibitory effect on the widespread application of the system, especially in the civilian field.

[0004] With the cross-development of various disciplines, the demand for navigation and positioning accuracy is increasing, and at the same time, the development trend of navigation systems is also moving towards passive, real-time, and concealed aspects. Since various single navigation technologies have their own shortcomings and scope of application, the current method of cross-combining two or more navigation technologies is usually adopted to achieve combined navigation, that is, to optimize the selection of various types of sensors, and complement each other's strengths and weaknesses, so as to enhance the various performance of combined navigation. At present, the common combined navigation methods mostly use inertial navigation as the main navigation system, and assist with satellite navigation, terrain navigation, etc. Although the characteristics and application scopes of different auxiliary navigation methods are different, geomagnetic navigation can make up for the shortcomings of existing combined navigation systems to a certain extent. It is a new auxiliary navigation method with great military potential. The basic structure of the underwater geomagnetic auxiliary navigation system is as follows. Figure 1 As shown, a real-time map is first formed based on the geomagnetic feature data sequence measured by the aircraft during navigation. Then, the correlation analysis method is used to match the information of the real-time map and the geomagnetic reference map in the navigation computer. The similarity between the two is calculated according to appropriate method rules to obtain the "best" information position matching point.

[0005] The geomagnetic matching algorithm is a core technology in geomagnetic matching navigation, significantly impacting navigation positioning accuracy and matching performance. Geomagnetic matching is essentially matching with a digital map. Compared to commonly used filtering methods, its principle does not require linearization of the geomagnetic field and is also applicable to intermittent matching problems. Currently, research on geomagnetic matching algorithms has yet to establish a comprehensive system, with most studies focusing on simulations. Numerous practical challenges remain to be addressed, such as the matching algorithm's varying matching probabilities across various geomagnetic regions, large errors between measured magnetic field values ​​and the reference map, and the long distance between the track to be matched and the actual trajectory of the vehicle, which prevents the matching algorithm from meeting its requirements. To address these issues, existing matching algorithms must be optimized or new ones developed.

[0006] In recent years, another key research approach for finding optimal solutions has been to utilize intelligent random search techniques. Since their introduction, swarm intelligence optimization algorithms, with their superior performance and global convergence, have been extensively and specifically applied in fields such as neural network optimization, signal processing, signal control, and optimal path selection, achieving impressive research results. These applications have demonstrated the algorithm's exceptional optimization capabilities.

[0007] At present, with the further development of modern computer information technology, bionics, genetics and other disciplines, intelligent optimization algorithms have become increasingly rich and complete, and have a certain room for growth. At the same time, they have gradually become the best method for dealing with practical optimization problems. Intelligent optimization algorithms mainly include evolutionary algorithms, swarm intelligence algorithms, artificial neural network algorithms, etc.

[0008] Taking the actual problems that arise in the environment of geomagnetic matching and positioning as the starting point, some parameters and related concepts in the algorithm have been optimized and improved to make them more in line with practical significance. At the same time, the objective function value is used to guide its individuals to search for target objects, thus meeting the processing requirements of geomagnetic matching, and thus making the designed schemes more efficient in achieving geomagnetic matching.

[0009] 1. Patent Name: "Improved ICCP Underwater Geomagnetic Matching Method Based on Triangle Matching Algorithm" Patent Number: 202010116489.4 uses the traditional ICCP geomagnetic matching algorithm, supplemented by the PSO optimization algorithm to initialize the particle constraint model. The ICCP algorithm has the following main characteristics: 1. It is sensitive to errors in the geomagnetic measurement sequence. When the geomagnetic measurement sequence has large errors, the ICCP matching success rate is low and the matching error increases. 2. The path provided by the inertial navigation system on the carrier should have a small error compared to the actual path. If the initial error is large, this method may cause the error to increase.

[0010] Based on the traditional image matching algorithm as the main method and the intelligent optimization algorithm as the auxiliary method, this application proposes a geomagnetic matching algorithm model based on the intelligent optimization algorithm and optimized based on the improved data group intelligent search strategy. In order to meet the needs of underwater geomagnetic matching, the Grey Wolf algorithm is improved and an objective function suitable for geomagnetic matching is established. This solves the shortcomings of the existing technology, such as the traditional geomagnetic matching algorithm being greatly affected by the initial positioning error, and improves the accuracy of geomagnetic navigation matching.

[0011] 2. Patent name "Geomagnetic matching navigation method based on PSO and ICCP" Patent number 202111553624.2 uses the PSO optimization algorithm to first perform coarse matching to obtain candidate trajectories, and then uses the traditional ICCP algorithm for multiple matching models. The real-time performance is limited. At the same time, when the PSO algorithm handles multi-dimensional optimization problems, the particles will show jumpy movement during the search process, which is easy to fall into the local optimal solution, resulting in a decrease in algorithm performance. The improved gray wolf algorithm used in this application is significantly superior to the PSO algorithm in terms of convergence performance and stability in handling complex multi-dimensional problems.

[0012] 3. The patent name is “A positioning system for indoor public service places” and the patent number is 201911028605.0. The process adopted is that the geomagnetic database module provides geomagnetic reference data, the data preprocessing module provides geomagnetic measurement data, and the INS module provides inertial positioning position information. Finally, the geomagnetic matching algorithm module calculates this information and sends the result to the mobile terminal. Its usage scenario is indoors. The usage scenario of the present invention is underwater. At the same time, the geomagnetic information on land is far more than the information underwater. In order to solve this problem, the present invention uses an effective spatial bilinear interpolation method. This method can use the spatial interpolation method to perform fine-grained expansion under the premise of relatively sparse reference points to ensure that the accuracy of the library meets the requirements of matching positioning. At the same time, the bilinear interpolation method is simple to calculate and has a good interpolation effect when the grid spacing is small. It solves the problem that the amount of underwater geomagnetic information is insufficient and needs real-time processing. Summary of the Invention

[0013] In order to solve the above problems, in view of the shortcomings of the existing technology, such as the traditional geomagnetic matching algorithm being greatly affected by the initial positioning error, the present invention proposes a geomagnetic matching method based on the improved data group intelligent search strategy optimization, and adopts the geomagnetic matching algorithm method with the data group intelligent search strategy improved by the Grey Wolf algorithm to perform matching, so as to improve the accuracy of geomagnetic navigation matching.

[0014] To achieve the above object, the technical solution adopted by the present invention is:

[0015] The geomagnetic matching method based on the improved data group intelligent search strategy optimization includes the following steps:

[0016] (1) Obtain the geomagnetic value information to be matched from the geomagnetic map;

[0017] (2) Establish the objective function based on the obtained data information;

[0018] (3) Based on the geomagnetic matching algorithm of the data group intelligent search strategy improved by the Grey Wolf algorithm, the data group position is initialized, and the fitness value is calculated according to the objective function to determine the optimal data element position;

[0019] (4) Update the position of the data element through the entire data group hunting process, and determine the updated optimal fitness value and optimal data element position;

[0020] (5) The position of the optimal data element is updated once according to the preset maximum number of iterations to obtain the optimal fitness vector and determine the optimal matching track result.

[0021] As a further improvement of the present invention, the step (1) of obtaining the geomagnetic value information to be matched is specifically as follows:

[0022] The bilinear interpolation method is used to obtain the geomagnetic value information to be matched. The geomagnetic value of any point inside the grid is expressed as follows:

[0023] .

[0024] in, is the geomagnetic value of the point being sought, is the bilinear interpolation coefficient, is the geomagnetic reference map coordinate.

[0025] As a further improvement of the present invention, the method using bilinear interpolation is applicable to the specific calculation formula of geomagnetic matching as follows:

[0026] A grid in the geomagnetic map, set the coordinates The geomagnetic value of the four grid intersection points is , then the coordinates of the four grid points in this area and the geomagnetic values ​​of the magnetic field at the grid points are substituted into:

[0027]

[0028] in, For coordinates The geomagnetic value of are the bilinear interpolation coefficients.

[0029] Arranging the formula we can get Four coefficients:

[0030]

[0031] After using this formula to obtain the four coefficients, the geomagnetic values ​​of all nodes within each grid can be obtained. ,After processing the grid by this geomagnetic value extraction ,technology, the geomagnetic value to be matched is obtained for ,matching trajectory.

[0032] As a further improvement of the present invention, the objective function of step (2) is established as follows:

[0033] The improved MAD algorithm is used as the correlation metric, which is expressed as

[0034]

[0035] The above formula is the number of trajectory sampling points, For the Geomagnetic measurement sequence matching points, To be matched Geomagnetic measurement sequence matching points, is the average value of the geomagnetic value sequence obtained by the magnetic sensor, It represents the average value of all geomagnetic map sequences to be matched. The metric is to subtract the mean value of the corresponding geomagnetic values ​​of all tracks in the matching process from their respective means, and match them using the characteristics of relative changes. The above formula gives the evaluation set of all tracks to be matched, and the track to be matched corresponding to the minimum value is the one we are looking for:

[0036]

[0037] Where, is the number of tracks to be matched by the aircraft. After improvement, when setting the constant error in the noise of the magnetic sensor measurement sequence, an arbitrary constant can be given. The random error in the noise is set to the noise whose probability density function obeys the Gaussian distribution. On this basis, the fitness function is rewritten as

[0038]

[0039] The larger the fitness function is, the closer it is to the optimal solution.

[0040] As a further improvement of the present invention, the gray wolf algorithm in step (3) is improved as follows:

[0041] When found Twiddle factors, scaling factor, 、 After the value of the translation factor is determined, the actual trajectory is obtained by the transformation, and the data element and the transformation factor are combined. The variable to be optimized is the individual state of the data element:

[0042] (1) Set the data group size to vectors, maximum number of iterations , the search space is 4-dimensional, :

[0043] (2) Initial data group, i.e. initialization vectors ,in is the data group vector;

[0044] (3) The rewritten fitness value formula is given:

[0045] .

[0046] As a further improvement of the present invention, the position update of the data element in the whole data group hunting process in step (4) is specifically as follows:

[0047] 1) Following, chasing, and approaching prey;

[0048] 2) Chasing, trapping, and harassing prey until they cease their movements;

[0049] 3) Attacking the target prey means finding the target value;

[0050] The Gray Wolf Algorithm is an optimization algorithm based on the tracking, encircling and hunting of the wolf pack, i.e. the data group. The fitness value of the entire group is divided into the main data elements from top to bottom. , secondary data element , common data elements and underlying data elements , master data element The fitness value of the data element is set to the highest to determine the active movement direction of the data group; and data elements The fitness value of Indicates the reference direction; data element The fitness value is the lowest, subject to 、 、 Data elements and provide stability to the data group;

[0051] The basic idea of ​​the algorithm is 、 、 Locating the prey is the optimal solution, guiding the data element Surround and hunt to find the optimal solution:

[0052] The process of data group surrounding prey is expressed as:

[0053]

[0054]

[0055]

[0056]

[0057] in, represents the relative position between the individual and the prey, is the data element position update formula, where, is the current iterative algebra, and is the synergy coefficient vector; represents the position vector of the prey; Represents the position vector of the current data element; during the entire iteration It decreases linearly from 2 to 0. and yes Random vectors in ;

[0058] After the data group surrounds the prey, the data element with the highest fitness value is calculated. 、 、 The position of the optimal solution is determined by:

[0059]

[0060]

[0061] Finally, the data element 、 、 The location of the data group is jointly determined as:

[0062] .

[0063] As a further improvement of the present invention, the coefficient vector of the process of the data group surrounding the prey , in the early stages of the Gray Wolf Optimization algorithm, once the coefficient vector If it is greater than 1, the algorithm's ability to explore the global situation may be affected: In the middle and late stages of the algorithm, if the coefficient vector If it is less than 1, the algorithm will converge prematurely;

[0064] Therefore, a new coefficient vector is proposed The generation formula makes it more likely to maintain less than 1 in the early stage of the algorithm and greater than 1 in the middle and late stages, thereby improving the algorithm's exploration ability. The coefficient vector in the process of the data group surrounding the optimal solution The specific improvements are as follows:

[0065] Set coefficient vector The formula for generating

[0066]

[0067] in, is the total number of iterations, for A random number between .

[0068] This application has the following beneficial effects:

[0069] This invention discloses a geomagnetic matching method based on an improved data group intelligent search strategy. This method addresses existing shortcomings in existing geomagnetic matching algorithms, such as the significant impact of initial positioning errors. By treating the geomagnetic matching problem as an optimization problem, this method proposes an improved gray wolf search optimization algorithm suitable for geomagnetic matching. By selecting several movement behaviors of the wolf pack, the algorithm achieves convergence to the optimal solution. This algorithm can obtain the desired optimal track, improve the accuracy of geomagnetic navigation matching, and thereby enhance the positioning accuracy and robustness of the integrated navigation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is an overall flow chart of the method disclosed in the present invention;

[0071] Figure 2 is an affine motion parameter model graph constructed in the method disclosed in the present invention;

[0072] Figure 3 It is a grid area in the geomagnetic map in the method disclosed in the present invention. DETAILED DESCRIPTION

[0073] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0074] The present invention discloses a geomagnetic matching algorithm based on an improved data group intelligent search strategy of the Grey Wolf algorithm, comprising the following steps:

[0075] Step 1: From Figure 2 From the affine motion parameter model, we can know that the inertial trajectory transformation and geomagnetic value extraction method can be used to determine Twiddle factors, scaling factor, 、 The value of the translation factor.

[0076] Step 2: Use bilinear interpolation to extract the geomagnetic value on the grid line. The geomagnetic value of any point inside the grid is expressed as follows:

[0077]

[0078] in, is the geomagnetic value of the point being sought, is the bilinear interpolation coefficient, is the geomagnetic reference map coordinate.

[0079] Figure 3 Represents a grid in a geomagnetic map, with coordinates The geomagnetic values ​​of the four grid intersections are , and substituting the coordinates of the four grid points in this area and the magnetic field values ​​of the grid points into the equation:

[0080]

[0081] in, For coordinates The geomagnetic value of are the bilinear interpolation coefficients.

[0082] Arranging the formula we can get Four coefficients:

[0083]

[0084] After obtaining the four coefficients by this method, the geomagnetic value of any point inside a specific grid area can be obtained ,After processing the grid by this geomagnetic value extraction technology, the ,geomagnetic values ​​to be matched can be obtained for ,matching trajectories.

[0085] Step 3: The steps to establish the objective function are as follows;

[0086] The present invention adopts the improved MAD algorithm as the correlation metric, which is expressed as

[0087]

[0088] The above formula is the number of trajectory sampling points, For the Geomagnetic measurement sequence matching points, To be matched Geomagnetic measurement sequence matching points, is the average value of the geomagnetic value sequence obtained by the magnetic sensor, It represents the average value of all the geomagnetic map sequences to be matched. This metric is to subtract the mean value of the geomagnetic values ​​corresponding to all the tracks in the matching process, and use the relatively changing characteristics to match to reduce the influence caused by the constant error. The above formula can be used to obtain the evaluation set of all the tracks to be matched, and the track to be matched corresponding to the minimum value is the track to be matched.

[0089]

[0090] Where, is the number of tracks to be matched by the aircraft. After improvement, when setting the constant error in the noise of the magnetic sensor measurement sequence, an arbitrary constant can be given, and the random error in the noise is set to the noise whose probability density function obeys the Gaussian distribution. On this basis, the fitness function can be rewritten as

[0091]

[0092] The larger the fitness function is, the closer it is to the optimal solution.

[0093] Step 4: Initialization steps of the geomagnetic matching algorithm based on the data group intelligent search strategy improved by the Grey Wolf algorithm are as follows:

[0094] According to step 1, find Twiddle factors, scaling factor, 、 After the value of the translation factor is determined, the true trajectory can be obtained by the transformation. The data element is then combined with the transformation factor, and the variable to be optimized is the individual state of the data element.

[0095] (1) Set the data group size to vectors, maximum number of iterations , the search space is 4-dimensional, :

[0096] (2) Initial data group, i.e. initialization vectors ,in is the data group vector;

[0097] (3) The rewritten fitness value formula is given:

[0098]

[0099] Step 5: The process of the entire data group behavior and the formula for updating the data element position, and the steps for determining the updated optimal fitness value and optimal data element position are as follows:

[0100] The algorithm is an optimization algorithm based on the three steps of tracking, encircling and hunting of data group hunting behavior. The fitness value of the entire group is divided into main data elements from top to bottom. , secondary data element , common data elements and underlying data elements . Master Data Element The fitness value of the data element is set to the highest to determine the active movement direction of the data group; and The fitness value decreases in turn, responsible for sending data elements Indicates the reference direction; data element The fitness value is the lowest, subject to 、 、 , and provide stability for the data group. The basic idea of ​​the algorithm is to 、 、 Locate prey (optimal solution) and guide data elements Surround and hunt (find the optimal solution).

[0101] The process of data group surrounding prey can be expressed as:

[0102]

[0103]

[0104]

[0105]

[0106] in, represents the distance between the individual and the prey, is the data element position update formula. is the current iterative algebra, and is the synergy coefficient vector; represents the position vector of the prey; Represents the position vector of the current data element; during the entire iteration It decreases linearly from 2 to 0. and yes A random vector in .

[0107] The coefficient vector in the process of the data group surrounding the prey The specific improvements are as follows: In the early stage of the algorithm, if the coefficient vector , then the algorithm's exploration ability may be affected: In the middle and late stages of the algorithm, if the coefficient vector , it may cause the algorithm to converge prematurely and thus affect the results. Therefore, the present invention proposes a new coefficient vector The generation method of , so that it can be kept less than 1 in the early stage of the algorithm and greater than 1 in the middle and late stages with a high probability, so as to achieve the survey capability of the stable algorithm and prevent premature convergence. Set the coefficient vector The way to generate is

[0108]

[0109] in, is the total number of iterations, for A random number between .

[0110] After the data group surrounds the prey, the data element with the highest fitness value is calculated. 、 、 The position of the optimal solution is determined by:

[0111]

[0112]

[0113] Finally, the data element 、 、 The location of the data group is jointly determined as:

[0114]

[0115] Step 5: Update the position of the optimal data element once according to the preset maximum number of iterations to obtain the optimal fitness vector and determine the optimal matching track result.

[0116] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent variation based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A geomagnetic matching method based on an improved data group intelligent search strategy optimization, characterized in that: The following steps are involved: (1) Obtain the geomagnetic value information to be matched from the geomagnetic map; (2) Establish the objective function based on the obtained data information; (3) Based on the geomagnetic matching algorithm of the data group intelligent search strategy improved by the Grey Wolf algorithm, the data group position is initialized, and the fitness value is calculated according to the objective function to determine the optimal data element position; (4) Update the position of the data element through the entire data group hunting process, and determine the updated optimal fitness value and optimal data element position; (5) The position of the optimal data element is updated once according to the preset maximum number of iterations to obtain the optimal fitness vector and determine the optimal matching track result.

2. The geomagnetic matching method based on improved data group intelligent search strategy optimization according to claim 1, characterized in that: The step (1) of obtaining the geomagnetic value information to be matched is specifically as follows: The bilinear interpolation method is used to obtain the geomagnetic value information to be matched. The geomagnetic value of any point inside the grid is expressed as follows: ， in, is the geomagnetic value of the point being sought, is the bilinear interpolation coefficient, is the geomagnetic reference map coordinate.

3. The geomagnetic matching method based on improved data group intelligent search strategy optimization according to claim 2, characterized in that: The specific calculation formula for the bilinear interpolation method applicable to geomagnetic matching is as follows: A grid in the geomagnetic map, with coordinates The geomagnetic value of the four grid intersection points is , then substitute the coordinates of the four grid points in this area and the geomagnetic values ​​of the magnetic field at the grid points into: ； in, For coordinates The geomagnetic value of is the bilinear interpolation coefficient; Arranging the formula we can get Four coefficients: ； After using this formula to obtain the four coefficients, the geomagnetic values ​​of all nodes within each grid can be obtained. After processing the grid using the geomagnetic value extraction technology, the geomagnetic values ​​must be matched to match the trajectory.

4. The geomagnetic matching method based on improved data group intelligent search strategy optimization according to claim 1, characterized in that: The objective function of step (2) is established as follows: The improved MAD algorithm is used as the correlation metric and is expressed as ； The above formula is the number of trajectory sampling points, For the Geomagnetic measurement sequence matching points, To be matched Geomagnetic measurement sequence matching points, is the average value of the geomagnetic value sequence obtained by the magnetic sensor, It represents the average value of all geomagnetic map sequences to be matched. The metric is to subtract the mean value of the corresponding geomagnetic values ​​of all tracks in the matching process from their respective means, and match them using the characteristics of relative changes. The above formula is the evaluation set of all tracks to be matched, and the track to be matched corresponding to the minimum value is the one we are looking for: ； Where, is the number of tracks to be matched by the aircraft. After improvement, when setting the constant error in the noise of the magnetic sensor measurement sequence, an arbitrary constant can be given. The random error in the noise is set to the noise whose probability density function obeys the Gaussian distribution. On this basis, the fitness function is ； The larger the fitness function is, the closer it is to the optimal solution.

5. The geomagnetic matching method based on improved data group intelligent search strategy optimization according to claim 4, characterized in that: The improvement of the grey wolf algorithm in step (3) is as follows: When found Twiddle factors, scaling factor, 、 After the value of the translation factor is adjusted, the actual trajectory is transformed, and the data element and the transformation factor are combined. The variable to be optimized is the individual state of the data element: (1) Set the data group size to vectors, maximum number of iterations , the search space is 4-dimensional, : (2) Initial data group, i.e. initialization vectors ,in is the data group vector; (3) The rewritten fitness value formula is given: 。 6. The geomagnetic matching method based on improved data group intelligent search strategy optimization according to claim 1, characterized in that: The specific update of the position of the data element in the whole data group hunting process in step (4) is as follows: 1) Following, chasing, and approaching prey; 2) Chasing, trapping, and harassing prey until they cease their movements; 3) Attacking the target prey means finding the target value; The Gray Wolf Algorithm is an optimization algorithm based on the tracking, encircling and hunting of the wolf pack, i.e. the data group. The fitness value of the entire group is divided into the main data elements from top to bottom. , secondary data element , common data elements and underlying data elements , master data element The fitness value of the data element is set to the highest to determine the active movement direction of the data group; and data elements The fitness value of Indicates the reference direction; data element The fitness value is the lowest, subject to 、 、 Data elements and provide stability to the data group; The basic idea of ​​the algorithm is 、 、 Locating the prey is the optimal solution, guiding the data element Surround and hunt to find the optimal solution: The process of the data group surrounding the prey is: ； ； ； ； in, Indicates the relative position between the individual and the prey, is the data element position update formula, where, is the current iterative algebra, and is the synergy coefficient vector; represents the position vector of the prey; Represents the position vector of the current data element; Throughout the iterative process It decreases linearly from 2 to 0. and yes Random vectors in ; After the data group surrounds the prey, the data element with the highest fitness value is calculated. 、 、 The position of the optimal solution is determined by: ； ； Finally, the data element 、 、 The location of the data group is jointly determined as: 。 7. The geomagnetic matching method based on improved data group intelligent search strategy optimization according to claim 6, characterized in that: The coefficient vector of the data group surrounding the prey process , in the early stages of the Gray Wolf Optimization algorithm, once the coefficient vector If it is greater than 1, the algorithm's ability to explore the global situation may be affected: In the middle and late stages of the algorithm, if the coefficient vector If it is less than 1, the algorithm will converge prematurely; Therefore, a new coefficient vector is proposed The generation formula of , makes it have a greater probability of maintaining less than 1 in the early stage of the algorithm and greater than 1 in the middle and late stages, thereby improving the algorithm's exploration ability, and the coefficient vector in the process of group surrounding the optimal solution The specific improvements are as follows: Set coefficient vector The formula for generating ； in, is the total number of iterations, for A random number between .

Images