A method for correcting pointing of a rotating double prism based on a swarm optimization algorithm

Abstract

The present application relates to a kind of based on rotating double-prism pointing correction method of swarm optimization algorithm.Provide a rotating double-prism system includes camera, first prism, second prism, first gear, first pinion, second gear, second pinion, first motor, second motor, lower computer, upper computer;The method first establishes the ideal model and assembly error model of rotating double-prism system, identifies assembly error according to the deviation between the ideal angle of target pointing and actual angle, and finally optimizes the parameter of prism based on swarm optimization algorithm, reduces the influence of prism installation error, improves the actual pointing accuracy of rotating double-prism system.

Description

Technical Field

[0001] This invention belongs to the field of photoelectric tracking technology, specifically relating to a rotating biprism pointing correction method based on a mouse swarm optimization algorithm. Background Technology

[0002] Traditional beam pointing devices mainly include gimbal assemblies and mirror assemblies. Gimbal assemblies change the beam propagation direction by rotating the pitch and azimuth axes of the gimbal, which houses detectors and lasers. However, these devices are large, heavy, and have poor dynamic performance. Mirror assemblies change the beam direction by altering the attitude of a mirror mounted in front of the detector. However, these devices have a simple structure, are easily damaged, and are sensitive to mechanical errors.

[0003] A rotating double prism changes the direction of beam propagation by controlling the coaxial independent rotation of two prisms. It has the characteristics of compact structure, fast response and good dynamic performance. Existing technology ([1] Qiu Sai, Sheng Lei, Gao Shijie et al. Rotational double prism pointing system rotation angle compensation deviation correction method [J]. Optical Communication Technology, 2021, 45(02): 41-45.) uses mathematical compensation method to analyze the mathematical relationship between pointing error and prism rotation angle error, and calculates the prism compensation rotation angle to correct the system deviation. This method can effectively improve the pointing accuracy of the rotating double prism system, but the solution process of prism compensation rotation angle is complicated and the amount of calculation is large.

[0004] The existing technology (Shen Ying, Li Leqian, Huang Feng, et al. A pointing correction method based on particle swarm optimization algorithm, CN202111001376.0[P / OL].2021-08-30.) proposes a pointing error correction method based on particle swarm optimization algorithm. This method adjusts the prism setting value of the reverse solution algorithm based on the mathematical model of the experimental prototype, thereby improving the pointing accuracy. However, the particle swarm optimization algorithm involved in this method is prone to getting trapped in local optima, and the setting of algorithm parameters will seriously affect the convergence speed and convergence accuracy, requiring a lot of time to adjust the parameters. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a rotating biprism pointing correction method based on a mouse swarm optimization algorithm.

[0006] To achieve the above objectives, the technical solution of the present invention is: a rotating double prism pointing correction method based on a mouse swarm optimization algorithm, providing a rotating double prism system, including a camera (1), a first prism (2), a second prism (3), a first large gear (4), a first small gear (5), a second large gear (6), a second small gear (7), a first motor (8), a second motor (9), a lower-level machine (10), and a higher-level machine (11); wherein the two prisms, the two large gears, and the camera are coaxially mounted; the extended shafts of the first motor (8) and the second motor (9) Each is connected to the first pinion (5) and the second pinion (7) via a coupling; the first large gear (4) and the second large gear (6) are respectively connected to the first prism (2) and the second prism (3), the first prism (2) and the second prism (3) have the same apex angle and refractive index, the first motor (8) and the second motor (9) each have a built-in Hall sensor to measure the rotation angles θ1 and θ2 of the first prism (2) and the second prism (3); the camera (1) sends the acquired image to the host computer (11), according to the coordinates (x, y, x) of the target's imaging point on the image. t ,y t The host computer (11) calculates the angles θ1 and θ2 that the first prism (2) and the second prism (3) need to rotate to make the target imaging point located at the center of the camera's field of view, and sends θ1 and θ2 to the slave computer (10). The slave computer controls the rotational speeds V1 and V2 of the first motor (8) and the energizing times t1 and t2 of the second motor (9) to drive the first prism (2) and the second prism (3) to rotate by the specified angle. The method includes the following steps:

[0007] Step S1: Establish an ideal model and assembly error model for the beam deflection of the rotating double prism system;

[0008] Step S2: Construct an error database of the ideal and actual rotation angles of the prism;

[0009] Step S3: Identify assembly errors using a mouse swarm optimization algorithm based on error database information;

[0010] Step S4: Use a mouse swarm optimization algorithm to optimize the apex angle and refractive index of the prism.

[0011] In one embodiment of the present invention, the specific implementation of establishing the ideal model and assembly error model of the rotating double prism system for beam deflection in step S1 is as follows:

[0012] Step S11: Establish an ideal model for beam deflection by the rotating double prism system:

[0013] Ideally, the line of sight s0 emitted from the camera (1) will point to the center point of the target after being refracted four times by the first prism (2) and the second prism (3). The calculation process is as follows:

[0014]

[0015] Among them, T i This represents the four refraction processes through the prism surface. The refraction vector is calculated based on Snell's law in vector form. The parameters involved in the calculation process include: the apex angle α1 of the first prism (2), the apex angle α2 of the second prism (3), the refractive index n1 of the first prism (2), the refractive index n2 of the second prism (3), the rotation angle θ1 of the first prism (2), and the rotation angle θ2 of the second prism (3).

[0016] set up K1, L1, M1 are vectors Direction cosine, The corresponding azimuth angle Θ and elevation angle Φ are calculated by the following formula:

[0017]

[0018] Step S12: Establish the assembly error model of the rotating double prism system:

[0019] In practice, improper prism installation will introduce assembly errors, which include prism tilt error and bearing tilt error. Under the influence of assembly errors, the line of sight s0 emitted by the camera (1) will deviate from the target center point after being deflected by the prism. Therefore, assembly errors should be considered during the beam deflection process. The calculation process is as follows:

[0020]

[0021] Among them, H i This represents the four refraction processes after passing through the first prism (2) and the second prism (3). The refraction vector is calculated using Snell's law in vector form, involving parameters including: the apex angle α1 of the first prism (2), the apex angle α2 of the second prism (3), the refractive index n1 of the first prism (2), the refractive index n2 of the second prism (3), the rotation angle θ1 of the first prism (2), and the rotation angle θ2 of the second prism (3). p,g This represents the prism tilt error matrix, which is related to the prism tilt angle θ. p,g and the prism tilt direction λ p,g Related to M b,g This represents the bearing error matrix, related to the bearing tilt angle θ. b,g and bearing tilt direction λ b,g Relevant, g = 1 or 2, indicating the first prism or the second prism;

[0022] set up K2, L2, M2 are vectors The direction cosine, considering prism tilt error and bearing tilt error. azimuth Θ e and elevation angle Φ e Calculated by the following formula:

[0023]

[0024] In one embodiment of the present invention, the specific process of constructing the error database of the ideal rotation angle and the actual rotation angle of the prism in step S2 is as follows:

[0025] S21. Adjust the prism angle to zero.

[0026] S22. Record the pixel coordinates (x, y) of the target imaging point in the image. t ,y t );

[0027] S23. Based on the pixel coordinates of the target point, and using the inverse kinematics algorithm based on the ideal model of beam deflection of the rotating double prism system, calculate the ideal angle that the two prisms need to rotate to make the target image at the center of the camera's field of view. and

[0028] S24, The host computer rotates the prism. and The signal is sent to the lower-level machine (10), which controls the first motor (8) and the second motor (9) to move, causing them to drive the first large gear (4), the first small gear (5), the second large gear (6), and the second small gear (7) to rotate, thereby driving the first prism (2) and the second prism (3) to rotate by an angle. and

[0029] S25. Due to systematic errors such as processing and assembly errors, the camera's field of view center will deviate from the target center point after the prism is rotated into position. Record the deviation distances Δx and Δy between the target point and the camera's field of view center in the x and y directions, and record the corrected pixel coordinates of the target point (x, y). c ,y c ):

[0030]

[0031] S26. Based on the corrected target point pixel coordinates, and using the inverse angle algorithm based on the assembly error model of the rotating double prism system, calculate the actual angle that the two prisms need to rotate to make the target image at the center of the camera's field of view. and

[0032] S27. Randomly and uniformly select J target points within the pointing range, and repeat steps S21-S24 for each target point to construct an error database containing J sets of ideal and actual prism rotation angles.

[0033] In one embodiment of the present invention, the specific implementation method of calculating the prism rotation angle using the inverse angle solution algorithm in step S23 is as follows:

[0034] Step S231: Based on the reversibility of light, the pixel coordinates (x, y) of the target point are used to determine the direction of light. t ,y t The equivalent focal length f of the rotating biprism system at zero position. c Calculate the elevation angle Φ of the target point relative to the exit center of the second prism beam. t and azimuth Θ t :

[0035]

[0036] Step S232: Based on the ideal model of beam deflection using a rotating double prism system, the elevation angle and azimuth angle can be obtained by traversing the prism rotation angles. The elevation angle is then fitted based on the angle information. - Relative rotation angle Δθ” curve;

[0037] Step S233, Φ t Substitute the fitted "elevation angle" The relative rotation angle Δθ is obtained by using the "relative rotation angle Δθ" curve;

[0038] Step S234: Let the first prism rotation angle θ1 = 0 and the second prism rotation angle θ2 = Δθ. Based on the ideal model of beam deflection by the rotating double prism system, calculate the azimuth angle θw.

[0039] Step S235, the solution for the prism angle is:

[0040] or

[0041] In one embodiment of the present invention, the specific process of identifying assembly errors using a mouse swarm optimization algorithm based on error database information in step S3 is as follows:

[0042] Step S31: Based on the error database of the ideal and actual rotation angles of the prism, the objective function for assembly error identification is defined as:

[0043]

[0044] Where J represents the number of target points, indicating that J static target pointing experiments are conducted. and These are the ideal rotation angles of the first prism (2) and the second prism (3) in the error database, respectively. and Let n ∈ (1, J) be the actual rotation angle of the first prism (2) and the second prism (3) in the error database, and let n represent the ideal rotation angle and actual rotation angle corresponding to the nth static target pointing in the error database.

[0045] Step S32: Use the following formula as the population of the mouse swarm optimization algorithm. Use the mouse swarm optimization algorithm to find the minimum value of the objective function Fit1 for assembly error identification, and the assembly error can be identified.

[0046]

[0047] Where X represents the population of the mouse swarm optimization algorithm, i∈(1,N), and N is the population size. For the prism tilt error matrix and bearing tilt error matrix Calculate the relevant parameters. This represents the position vector of the i-th mouse in the population. This indicates the tilt angle and tilt direction of the first prism (2). This indicates the prism tilt angle and tilt direction of the second prism (3). This indicates the bearing tilt angle and bearing tilt direction of the first prism (2). This indicates the bearing tilt angle and bearing tilt direction of the second prism (3).

[0048] In one embodiment of the present invention, the process of optimizing the apex angle and refractive index of the prism using a mouse swarm optimization algorithm in step S4 is as follows:

[0049] Step S41: Define the objective function for optimizing the prism parameters as follows:

[0050]

[0051] Wherein, Φ' k and Θ' k To obtain the elevation and azimuth angles by traversing the prism rotation angles using the assembly error model based on the assembly error identification results, Φ' e,k and Θ' e,k The elevation and azimuth angles are obtained by traversing the prism angles using an ideal model. The value of U is determined based on the step size of the traversed prism angles.

[0052] Step S42: Use the following formula as the population of the mouse swarm optimization algorithm. Use the mouse swarm optimization algorithm to find the minimum value of the objective function Fit2 for optimizing the prism parameters, and the optimized prism apex angle and prism refractive index parameters can be obtained.

[0053]

[0054] Where W represents the population of the mouse swarm optimization algorithm, r∈(1,P), and P is the population size. This represents the position vector of the r-th mouse in the population. The apex angles of the first prism (2), the second prism (3), the refractive index of the first prism (2), and the refractive index of the second prism (3) are respectively represented by the vector at that position.

[0055] In one embodiment of the present invention, the calculation process of U in step S41 is specifically as follows:

[0056] The rotation angle θ1 of the first prism (2) ranges from [0°, 360°], and the relative rotation angle between the first prism (2) and the second prism (3) ranges from [0°, 180°]. The rotation angle θ1 of the second prism (3) is θ1 + Δθ, where Δθ is the relative rotation angle between the first prism (2) and the second prism (3). Let the step size for traversing the rotation angle of the first prism (2) be a, and the step size for traversing the relative rotation angle be b. The formula for calculating U is as follows:

[0057]

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. This invention adopts mathematical modeling ideas and optimization method theory, utilizes the global search capability of the mouse swarm optimization algorithm, takes the assembly error model as the control object, and takes the deviation between the ideal and actual values ​​of the output beam elevation angle and azimuth angle as the control error to optimize the apex angle and refractive index parameters of the prism, thereby realizing the adjustment of the beam control parameters of the rotating double prism system. The method of this invention has a fast convergence speed and can significantly improve the pointing accuracy when optimizing prism parameters.

[0060] 2. The mouse swarm optimization algorithm used in the method of this invention has the advantages of simple algorithm theory, few parameter settings, strong adaptability, and fast algorithm convergence speed. Attached Figure Description

[0061] Figure 1 This is a schematic diagram of a rotating double prism system.

[0062] Figure 2 This is a diagram of the beam propagation path after refraction by a prism.

[0063] Figure 3 It is an iterative curve diagram of the particle swarm optimization algorithm and the mouse swarm optimization algorithm in the process of identifying assembly errors;

[0064] Figure 4 This is an iterative curve diagram of the process of optimizing prism parameters using the particle swarm optimization algorithm and the mouse swarm optimization algorithm. Detailed Implementation

[0065] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings.

[0066] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0067] To address the issue of large pointing errors in large-aperture rotating biprism systems and the tendency of algorithms such as particle swarm optimization to get trapped in local optima, this invention provides a pointing error correction method for rotating biprisms based on a mouse swarm optimization algorithm from the perspective of system identification. Considering that online compensation methods for prism rotation angles are computationally complex and time-consuming, this invention uses a simple mouse swarm optimization algorithm as its foundation. It identifies the assembly error of the rotating biprism system through system identification and then optimizes the apex angle and refractive index parameters of the prisms to reduce the pointing error of the rotating biprism system.

[0068] The rotating double prism system is composed of Figure 1 The main components shown include a camera (1), a first prism (2), a second prism (3), a first large gear (4), a first small gear (5), a second large gear (6), a second small gear (7), a first motor (8), a second motor (9), a lower-level machine (10), and a higher-level machine (11). The two prisms, two large gears, and the camera are coaxially mounted. The extended shafts of the first motor (8) and the second motor (9) are connected to the first small gear (5) and the second small gear (7) respectively via couplings. The first large gear (4) and the second large gear (6) are connected to the first prism (2) and the second prism (3) respectively. The first prism (2) and the second prism (3) have the same apex angle and refractive index. The first motor (8) and the second motor (9) each have a built-in Hall sensor for measuring the rotation angles θ1 and θ2 of the first prism (2) and the second prism (3). The camera (1) sends the acquired image to the higher-level machine (11), which then determines the target's position based on the image point coordinates (x, y, y). t ,y t The host computer (11) calculates the angles θ1 and θ2 that the first prism (2) and the second prism (3) need to rotate when the target imaging point is located at the center of the camera's field of view through an algorithm, and sends θ1 and θ2 to the slave computer (10). The slave computer controls the rotation speeds V1 and V2 of the first motor (8) and the second motor (9) as well as the power-on times t1 and t2, driving the first prism (2) and the second prism (3) to rotate by a specified angle.

[0069] This example presents a rotating biprism pointing error correction method based on a mouse swarm optimization algorithm, which includes the following steps:

[0070] Step S1: Establish an ideal model and assembly error model for the beam deflection of the rotating double prism system;

[0071] Step S2: Construct an error database of the ideal and actual rotation angles of the prism;

[0072] Step S3: Identify assembly errors using a mouse swarm optimization algorithm based on error database information;

[0073] Step S4: Use a mouse swarm optimization algorithm to optimize the apex angle and refractive index of the prism.

[0074] In step S1 provided in this embodiment of the invention, the specific process of establishing the ideal model and assembly error model of the rotating double prism system for beam deflection is as follows:

[0075] Step S11, establishing an ideal model for beam deflection by the rotating double prism system, specifically:

[0076] Ideally, the line of sight s0 emitted from the camera (1) will point to the center of the target after being refracted four times by the first prism (2) and the second prism (3), and the propagation path of the beam is as follows: Figure 2 As shown, the calculation process is as follows:

[0077]

[0078] Among them, T i This represents the four refraction processes through the prism surface. The refraction vector is calculated based on Snell's law in vector form. The parameters involved in the calculation process include: the apex angle α1 of the first prism (2), the apex angle α2 of the second prism (3), the refractive index n1 of the first prism (2), the refractive index n2 of the second prism (3), the rotation angle θ1 of the first prism (2), and the rotation angle θ2 of the second prism (3).

[0079] set up K1, L1, M1 are vectors Direction cosine, The azimuth angle Θ and elevation angle Φ are calculated by the following formula:

[0080]

[0081] Step S12, the process of establishing the assembly error model of the rotating double prism system is as follows:

[0082] In practice, improper prism installation will introduce assembly errors, which include prism tilt error and bearing tilt error. Under the influence of assembly errors, the line of sight s0 emitted by the camera (1) will deviate from the target center point after being deflected by the prism. Therefore, assembly errors should be considered during the beam deflection process. The calculation process is as follows:

[0083]

[0084] Among them, Hi This represents the four refraction processes after passing through the first prism (2) and the second prism (3). The refraction vector is calculated using Snell's law in vector form, involving parameters including: the apex angle α1 of the first prism (2), the apex angle α2 of the second prism (3), the refractive index n1 of the first prism (2), the refractive index n2 of the second prism (3), the rotation angle θ1 of the first prism (2), and the rotation angle θ2 of the second prism (3). p,g This represents the prism tilt error matrix, which is related to the prism tilt angle θ. p,g and the prism tilt direction λ p,g Related to M b,g This represents the bearing error matrix, related to the bearing tilt angle θ. b,g and bearing tilt direction λ b,g Relevant, g = 1 or 2, indicating the first prism or the second prism;

[0085] set up K2, L2, M2 are vectors The direction cosine, considering prism tilt error and bearing tilt error. azimuth Θ e and elevation angle Φ e Calculated by the following formula:

[0086]

[0087] In step S2 provided in this embodiment of the invention, the specific process of constructing the error database of the ideal rotation angle and the actual rotation angle of the prism is as follows:

[0088] S21. Adjust the prism angle to zero.

[0089] S22. Record the pixel coordinates (x, y) of the target imaging point in the image. t ,y t );

[0090] S23. Based on the pixel coordinates of the target point, and using the inverse kinematics algorithm based on the ideal model of the rotating double prism system, calculate the ideal angle that the two prisms need to rotate to make the target image at the center of the camera's field of view. and

[0091] S24, The host computer rotates the prism. and The signal is sent to the lower-level machine (10), which controls the first motor (8) and the second motor (9) to move, causing them to drive the first large gear (4), the first small gear (5), the second large gear (6), and the second small gear (7) to rotate, thereby driving the first prism (2) and the second prism (3) to rotate by an angle. and

[0092] S25. Due to systematic errors such as processing and assembly errors, the camera's field of view center will deviate from the target center point after the prism is rotated into position. Record the deviation distances Δx and Δy between the target point and the camera's field of view center in the x and y directions, and record the corrected pixel coordinates of the target point (x, y). c ,y c ):

[0093]

[0094] S26. Based on the corrected target point pixel coordinates, and using the inverse angle algorithm based on the assembly error model of the rotating double prism system, calculate the actual angle that the two prisms need to rotate to make the target image at the center of the camera's field of view. and

[0095] S27. Randomly and uniformly select J target points within the pointing range, and repeat steps S21-S24 for each target point to construct an error database containing J sets of ideal and actual prism rotation angles.

[0096] In step S23 provided in this embodiment of the invention, the process of calculating the prism rotation angle using the inverse angle algorithm is as follows:

[0097] Step S231: Based on the reversibility of light, the pixel coordinates (x, y) of the target point are used to determine the direction of light. t ,y t The equivalent focal length f of the rotating biprism system at zero position. c Calculate the elevation angle Φ of the target point relative to the center of the second prism. t and azimuth Θ t :

[0098]

[0099] Step S232: Based on the ideal model of beam deflection using a rotating double prism system, the elevation angle and azimuth angle can be obtained by traversing the prism rotation angles. The elevation angle is then fitted based on the angle information. - Relative rotation angle Δθ” curve;

[0100] Step S233, Φ t Substitute the fitted "elevation angle" The relative rotation angle Δθ can be obtained from the curve of "-relative rotation angle Δθ";

[0101] Step S234: Let the first prism rotation angle θ1 = 0 and the second prism rotation angle θ2 = Δθ, and calculate the azimuth angle θw based on the ideal model of beam deflection;

[0102] Step S235, the solution for the prism angle is:

[0103] or

[0104] In step S3 provided in this embodiment of the invention, the process of identifying assembly errors based on error database information using a mouse swarm optimization algorithm is as follows:

[0105] Step S31: Based on the error database of the ideal and actual rotation angles of the prism, the objective function for assembly error identification is defined as follows:

[0106]

[0107] A total of J static target pointing experiments were conducted. and These are the ideal rotation angles of the first prism (2) and the second prism (3) in the error database, respectively. and The actual rotation angles of the first and second prisms in the error database;

[0108] Step S32: Use the following formula as the population of the mouse swarm optimization algorithm. Use the mouse swarm optimization algorithm to find the minimum value of the assembly error identification objective function Fit1, and the assembly error can be identified.

[0109]

[0110] Where X represents the population of the mouse swarm optimization algorithm, i∈(1,N), and N is the population size. For the prism tilt error matrix and bearing tilt error matrix Calculate the relevant parameters. This represents the position vector of the i-th mouse in the population. This indicates the tilt angle and tilt direction of the first prism (2). This indicates the prism tilt angle and tilt direction of the second prism (3). This indicates the bearing tilt angle and bearing tilt direction of the first prism (2). This indicates the bearing tilt angle and bearing tilt direction of the second prism (3).

[0111] In step S4 provided in this embodiment of the invention, the process of optimizing the prism's apex angle and refractive index using a mouse swarm optimization algorithm is as follows:

[0112] Step S41: Define the objective function for optimizing the prism parameters as follows:

[0113]

[0114] Wherein, Φ' k and Θ'k To obtain the elevation and azimuth angles by traversing the prism rotation angles using the assembly error model based on the assembly error identification results, Φ' e,k and Θ' e,k The elevation and azimuth angles are obtained by traversing the prism angles using the ideal pointing model. The value of U is determined by the step size of the traversed prism angles.

[0115] Step S42: Use the following formula as the population of the mouse swarm optimization algorithm. Use the mouse swarm optimization algorithm to find the minimum value of the objective function Fit2 for optimizing the prism parameters, and the optimized prism apex angle and prism refractive index parameters can be obtained.

[0116]

[0117] Where W represents the population of the mouse swarm optimization algorithm, e∈(1,P), and P is the population size. This represents the position vector of the r-th mouse in the population. The apex angles of the first prism (2), the second prism (3), the refractive index of the first prism (2), and the refractive index of the second prism (3) are respectively represented by the vector at that position.

[0118] In step S41 provided in this embodiment of the invention, the calculation process of U is specifically as follows:

[0119] The rotation angle θ1 of the first prism (2) ranges from [0°, 360°], and the relative rotation angle between the first prism (2) and the second prism (3) ranges from [0°, 180°]. The rotation angle θ1 of the second prism (3) is θ1 + Δθ, where Δθ is the relative rotation angle between the first prism (2) and the second prism (3). Let the step size for traversing the rotation angle of the first prism (2) be a, and the step size for traversing the relative rotation angle be b. The formula for calculating U is as follows:

[0120]

[0121] Specifically, specific implementation examples of the present invention are as follows:

[0122] Based on the established prism parameters, an ideal model and an assembly error model for beam deflection of the rotating double prism system are established, wherein the apex angles of the first and second prisms are α1=α2=14.85°, and the refractive indices of the first and second prisms are n1=n2=1.515.

[0123] Adjust the first and second prisms to zero initial rotation angle, and randomly and evenly select 50 target points within the camera's field of view. Then, based on the pixel coordinates of the targets in the image... The host computer uses an inverse kinematic algorithm to determine the angle that the first and second prisms need to rotate to make the camera's field of view center coincide with the target. The specific process of solving the prism rotation angle using the inverse kinematics algorithm is as follows:

[0124] From the pixel coordinates (x) of the target point t ,y t The equivalent focal length f of the rotating biprism system at zero position. c =670 Calculate the elevation angle Φ of the target point with respect to the center of the second prism. t and azimuth Θ t :

[0125]

[0126] Based on an ideal model of beam deflection using a rotating double prism system, the elevation and azimuth angles can be obtained by iterating through the prism rotation angles. The elevation angle can then be fitted using this angle information. - Relative rotation angle Δθ” curve;

[0127] Φ t Substitute the fitted "elevation angle" The relative rotation angle Δθ can be obtained from the curve of "-relative rotation angle Δθ";

[0128] Rotate the first prism Second prism angle The azimuth angle θw is obtained based on an ideal model of beam deflection.

[0129] The solution to the prism rotation angle is:

[0130] or

[0131] The host computer rotates the prism. and The data is sent to the lower-level machine, which controls the movement of the first and second motors, causing them to drive the first large gear, the first small gear, the second large gear, and the second small gear to rotate, thereby driving the first and second prisms to rotate by an angle. and

[0132] Record the deviation distances Δx and Δy between the target point and the camera's field of view center in the x and y directions, respectively, and record the corrected pixel coordinates (x, y) of the target point. c ,y c ), construct an error database for ideal and actual turning angles;

[0133]

[0134] Based on the corrected target point pixel coordinates, the assembly error model of the rotating double prism system uses an inverse angle algorithm to calculate the actual angle that the two prisms need to rotate to make the target image at the center of the camera's field of view. and

[0135] The objective function for assembly error identification is as follows:

[0136]

[0137] in, and These are the ideal rotation angles of the first and second prisms in the error database, respectively. and The actual rotation angles of the first and second prisms in the error database;

[0138] Based on the analysis results of prism tilt error and bearing tilt error, the objective function for identifying assembly error needs to be established with a dimension of 8, and the population matrix X is specifically as follows:

[0139]

[0140] Where X is the population of the mouse swarm optimization algorithm, i∈(1,N), and N is the population size, N=50. For the prism tilt error matrix and bearing tilt error matrix Calculate the relevant parameters. This represents the position vector of the i-th mouse in the population. This indicates the tilt angle and tilt direction of the first prism. This indicates the tilt angle and tilt direction of the second prism. This indicates the bearing tilt angle and bearing tilt direction of the first prism. This indicates the bearing tilt angle and bearing tilt direction of the second prism.

[0141] Based on the established objective function, the mouse swarm optimization algorithm is called to identify assembly errors.

[0142] The objective function change curve of the error identification process is as follows: Figure 3 As shown, the error identification result is [θ] p,1 λ p,1 θ p,2 λ p,2 θ b,1 λ b,1 θ b,2 λ b,2 ] = [8.95° 10.11° 2.61° 10.83° - 8.87° 11.36° 8.95° 8.56°];

[0143] The specific process of optimizing prism parameters using the mouse swarm optimization algorithm is as follows:

[0144] The objective function for optimizing the prism parameters is as follows:

[0145]

[0146] Wherein, Φ' k and Θ' k To obtain the elevation and azimuth angles by traversing the prism rotation angles using the assembly error model based on the assembly error identification results, Φ' e,k and Θ' e,k The elevation and azimuth angles are obtained by traversing the prism corners using an ideal pointing model;

[0147] The first prism rotation step is 1°, and the relative rotation step is 90°. At this time, U = 1083.

[0148] The prism parameters are optimized by calling the mouse swarm optimization algorithm. The specific population settings of the mouse swarm optimization algorithm are as follows. The mouse swarm optimization algorithm is used to find the minimum value of the objective function Fit2 for optimizing the prism parameters, and the optimized prism apex angle and prism refractive index parameters can be obtained.

[0149]

[0150] Where W is the population of the mouse swarm optimization algorithm, r∈(1,P), P is the population size, P=50. This represents the position vector of the r-th mouse in the population. The first prism (2) vertex angle, the second prism vertex angle, the first prism refractive index, and the second prism refractive index are respectively represented by the vector at that position.

[0151] The objective function changes during the optimization of prism parameters as follows: Figure 4 As shown, the optimized prism parameters are: first prism apex angle Second prism apex angle First prism refractive index The refractive index of the second prism

[0152] The above are preferred embodiments of the present invention. Any changes made to the technical solution of the present invention that do not exceed the scope of the technical solution of the present invention shall fall within the protection scope of the present invention.

Claims

1. A rotating biprism pointing correction method based on a mouse swarm optimization algorithm, characterized in that, A rotating double prism system is provided, including a camera (1), a first prism (2), a second prism (3), a first large gear (4), a first small gear (5), a second large gear (6), a second small gear (7), a first motor (8), a second motor (9), a lower-level machine (10), and a higher-level machine (11); wherein the two prisms, the two large gears, and the camera are coaxially mounted; the extended shafts of the first motor (8) and the second motor (9) are respectively connected to the first small gear (5) and the second small gear (7) via couplings; the first large gear (4) and the second large gear (6) are respectively connected to the first prism (2) and the second prism (3), the first prism (2) and the second prism (3) have the same apex angle and refractive index, and the first motor (8) and the second motor (9) each have a built-in Hall sensor for measuring the rotation angle of the first prism (2) and the second prism (3). and The camera (1) sends the acquired image to the host computer (11), and the image is processed according to the coordinates of the target's imaging point on the image. The host computer (11) calculates the angle that the first prism (2) and the second prism (3) need to rotate to make the target imaging point located at the center of the camera's field of view. and and will and The data is sent to the lower-level machine (10), which controls the speed of the first motor (8) and the second motor (9). and and power-on time and The method involves driving the first prism (2) and the second prism (3) to rotate by a specified angle; the method includes the following steps: Step S1: Establish an ideal model and assembly error model for the beam deflection of the rotating double prism system; Step S2: Construct an error database of the ideal and actual rotation angles of the prism; Step S3: Identify assembly errors using a mouse swarm optimization algorithm based on error database information; Step S4: Optimize the prism's apex angle and refractive index using a mouse swarm optimization algorithm; In step S1, the specific implementation method for establishing the ideal model and assembly error model of the rotating double prism system for beam deflection is as follows: Step S11: Establish an ideal model for beam deflection by the rotating double prism system: Ideally, the line of sight emitted by camera (1) After four refractions through the first prism (2) and the second prism (3), it will point to the center point of the target. The calculation process is as follows: ， =1，2，3，4 in, This represents the four refraction processes through the prism surface. The refraction vector is calculated based on Snell's law in vector form. The parameters involved in the calculation include: the apex angle of the first prism (2). The apex of the second prism (3) The refractive index of the first prism (2) The refractive index of the second prism (3) The rotation angle of the first prism (2) The rotation angle of the second prism (3) ; set up , For vectors Direction cosine, corresponding azimuth angle and elevation angle Calculated by the following formula: Step S12: Establish the assembly error model of the rotating double prism system: In practice, improper prism installation will introduce assembly errors, which include prism tilt error and bearing tilt error. Under the influence of assembly errors, the line of sight emitted by the camera (1) will be affected. After being deflected by the prism, the beam will deviate from the target center point. Therefore, assembly errors should be considered during the beam deflection process. The calculation process is as follows: ， =1，2，3，4； =1，2 in, This represents the four refraction processes after passing through the first prism (2) and the second prism (3). The refraction vector is calculated using Snell's law in vector form, and the parameters involved include: the apex angle of the first prism (2). The apex of the second prism (3) The refractive index of the first prism (2) The refractive index of the second prism (3) The rotation angle of the first prism (2) The rotation angle of the second prism (3) , This represents the prism tilt error matrix, which is related to the prism tilt angle. and the tilt direction of the prism , This represents the bearing error matrix, along with the bearing tilt angle. and bearing tilt direction related, =1 or 2 indicates the first prism or the second prism; set up , For vectors The direction cosine, considering prism tilt error and bearing tilt error. azimuth and elevation angle Calculated by the following formula: In step S3, the specific process of identifying assembly errors using a mouse swarm optimization algorithm based on error database information is as follows: Step S31: Based on the error database of the ideal and actual rotation angles of the prism, the objective function for assembly error identification is defined as: in, The number of target points indicates the number of points to be processed. Sub-static target pointing experiment and These are the ideal rotation angles of the first prism (2) and the second prism (3) in the error database, respectively. and The actual rotation angles of the first prism (2) and the second prism (3) in the error database are given. , representing the first in the error database The sub-static target points to the corresponding ideal turning angle and actual turning angle; Step S32: Use the following formula as the population for the mouse swarm optimization algorithm, and use the mouse swarm optimization algorithm as the objective function for identifying assembly errors. By finding the minimum value, assembly errors can be identified; in, For the mouse swarm optimization algorithm, , For population size, For the prism tilt error matrix , and bearing tilt error matrix , Calculate the relevant parameters. Indicates the first in the population The position vector of the mouse. This indicates the tilt angle and tilt direction of the first prism (2). This indicates the prism tilt angle and tilt direction of the second prism (3). This indicates the bearing tilt angle and bearing tilt direction of the first prism (2). Indicates the bearing tilt angle and bearing tilt direction of the second prism (3); In step S4, the process of optimizing the prism's apex angle and refractive index using a mouse swarm optimization algorithm is as follows: Step S41: Define the objective function for optimizing the prism parameters as follows: in, and To obtain the elevation and azimuth angles by traversing the prism rotation angles based on the assembly error identification results, the assembly error model is used. and To obtain the elevation and azimuth angles by traversing the prism's rotation using an ideal model, at this point... The value of is determined based on the step size of the prism rotation angle during traversal; Step S42: Use the following formula as the population for the mouse swarm optimization algorithm, and apply the mouse swarm optimization algorithm to the objective function of optimizing the prism parameters. By finding the minimum value, the optimized prism apex angle and prism refractive index parameters can be obtained; in, For the mouse swarm optimization algorithm, , For population size, Indicates the first in the population The position vector of the mouse. , , , The apex angles of the first prism (2), the second prism (3), the refractive index of the first prism (2), and the refractive index of the second prism (3) are respectively represented by the vector at that position.

2. The rotating biprism pointing correction method based on mouse swarm optimization algorithm according to claim 1, characterized in that, In step S2, the specific process of constructing the error database of the ideal rotation angle and the actual rotation angle of the prism is as follows: S21. Adjust the prism angle to zero. S22. Record the pixel coordinates of the target imaging point in the image. ; S23. Based on the pixel coordinates of the target point, and using the inverse kinematics algorithm based on the ideal model of beam deflection of the rotating double prism system, calculate the ideal angle that the two prisms need to rotate to make the target image at the center of the camera's field of view. and ; S24, The host computer rotates the prism. and The signal is sent to the lower-level machine (10), which controls the first motor (8) and the second motor (9) to move, causing them to drive the first large gear (4), the first small gear (5), the second large gear (6), and the second small gear (7) to rotate, thereby driving the first prism (2) and the second prism (3) to rotate by an angle. and ; S25. Due to systematic errors such as processing and assembly errors, the camera's field of view center will deviate from the target center point after the prism is rotated into position. Record the distance between the target point and the camera's field of view center. direction and Deviation between directions , Record the corrected target point pixel coordinates. : S26. Based on the corrected target point pixel coordinates, and using the inverse angle algorithm based on the assembly error model of the rotating double prism system, calculate the actual angle that the two prisms need to rotate to make the target image at the center of the camera's field of view. and ; S27. Randomly and uniformly select within the pointing range. For each target point, repeat steps S21-S24 to construct a target point containing... An error database of the ideal and actual rotation angles of the prism assembly.

3. The rotating biprism pointing correction method based on mouse swarm optimization algorithm according to claim 2, characterized in that, In step S23, the specific implementation method for calculating the prism rotation angle using the inverse angle solution algorithm is as follows: Step S231: Based on the reversibility of light, the pixel coordinates of the target point are used to determine the direction of light. The equivalent focal length of the rotating biprism system at zero position Calculate the elevation angle of the target point relative to the exit center of the second prism beam. and azimuth : Step S232: Based on the ideal model of beam deflection using a rotating double prism system, the elevation angle and azimuth angle can be obtained by traversing the prism rotation angles. The elevation angle is then fitted based on the angle information. - Relative corner "curve; Step S233, will Substitute the fitted "elevation angle" - Relative corner "The relative turning angle is obtained by curve calculation" ; Step S234: Rotate the first prism. The second prism's angle The azimuth angle is obtained based on an ideal model of beam deflection using a rotating double prism system. ; Step S235, the solution for the prism angle is: or .

4. The rotating biprism pointing correction method based on mouse swarm optimization algorithm according to claim 1, characterized in that, In step S41, The calculation process is as follows: The rotation angle of the first prism (2) The range is The relative rotation angle range between the first prism (2) and the second prism (3) is as follows: The angle of the second prism (3) , Let be the relative rotation angle between the first prism (2) and the second prism (3), and let the step size for traversing the rotation angle of the first prism (2) be . The traversal step size relative to the corner is , The calculation formula is as follows: 。

Images