A method for compensating the full elevation angle error of the secondary reflector adjustment mechanism of an omnidirectionally movable radio telescope

Abstract

The present invention discloses a method for compensating the full elevation error of the sub-surface adjustment mechanism of an omnidirectional movable radio telescope, which determines the measurement posture points in the working space of the sub-surface adjustment mechanism, measures and obtains the posture measurement results of the moving platform at different elevation angles; based on the posture measurement results of the moving platform at a 90° elevation angle, identifies the geometric error, and modifies the control parameters of the kinematic model of the sub-surface adjustment mechanism; uses a neural network to predict the equivalent error of the drive rod at any pitch angle of the antenna and any posture of the moving platform, and assigns the correction value to the drive rod after superimposing it with the ideal drive rod length, and compensates the posture error of the moving platform in real time by correcting the length of the drive rod. The present invention effectively compensates for the error within the full elevation range of the sub-surface adjustment mechanism by performing graded compensation for geometric errors and non-geometric errors, greatly reduces the need for frequent calibration of the sub-reflecting surface of the radio telescope, reduces the maintenance cost in long-term use, and significantly improves the overall reliability and stability of the equipment.

Description

Technical Field

[0001] The invention belongs to the field of radio telescope antenna devices, and in particular relates to a method for compensating posture errors of a secondary surface adjustment mechanism of an omnidirectional movable radio telescope at full elevation angles. Background Art

[0002] Large-aperture omnidirectional movable radio telescope antennas usually need to work in open-air environments. Since the antenna structure is mainly composed of metal materials such as steel, it is easily affected by environmental loads such as gravity, temperature, wind load, rain and snow, which can cause deformation of the shape of the main reflector surface of the antenna and the supporting structure of the sub-reflector surface, deteriorate the surface accuracy of the reflector surface, and cause the microwave optical path to shift and produce misalignment errors, thereby reducing the gain of the antenna and causing pointing deviations, resulting in a decrease in the sensitivity of the antenna system. Therefore, it is necessary to install an adjustment mechanism between the sub-surface and the sub-surface support rod, adjust the sub-surface posture through the sub-surface adjustment mechanism, and actively correct the relative positions of the foci of the main and sub-reflector surfaces of the antenna and the feed phase center to compensate for the focusing error caused by structural deformation.

[0003] For antenna systems with high precision requirements, especially large-aperture omnidirectional movable radio telescope antennas, due to the large diameter of the sub-reflector, heavy weight, large adjustment required, and the need for adjustment in multiple degrees of freedom, the sub-reflector of the high-precision radio telescope antenna adopts a six-degree-of-freedom parallel mechanism with high rigidity, compact structure, high precision, and multiple degrees of freedom adjustment to adjust the position of the sub-reflector.

[0004] Since the antenna of the large-aperture omnidirectional movable radio telescope is in a complex outdoor working environment (gravity, temperature, wind load, rain and snow) for a long time, there are many factors that cause motion errors in the sub-surface adjustment mechanism in the complex environment, and the influencing mechanism is also relatively complex. After the sub-surface adjustment mechanism works for a long time, the mechanical transmission system and hinges of the electric cylinder will wear out, resulting in an increase in the hinge gap and a decrease in the transmission accuracy of the electric cylinder. Due to the influence of temperature load, the electric cylinder, hinge and other components will be thermally deformed, causing deviations in the mapping between the output posture of the sub-surface adjustment mechanism and the displacement of the motor. In addition, since the sub-surface adjustment mechanism is installed on the antenna, it is heavy and moves with the main surface of the antenna in pitch. The force characteristics of the sub-surface adjustment mechanism change with the change of the antenna pitch angle, and the terminal posture error also changes with the change of the antenna pitch angle, so that the movement of each dimension of the sub-surface adjustment mechanism will be affected, resulting in posture errors, and the adjusted sub-reflector surface cannot accurately reach the required posture, thereby affecting the implementation of the overall task.

[0005] Therefore, in response to the above technical problems, there is an urgent need to provide a method for compensating the posture error of the sub-surface adjustment mechanism of an omnidirectional movable radio telescope at all elevation angles. Summary of the invention

[0006] In order to solve the above-mentioned defects existing in the prior art, the purpose of the present invention is to provide a method for compensating the full elevation angle error of the sub-surface adjustment mechanism of an omnidirectional movable radio telescope. By establishing an error data set, n posture measurement results of a moving platform in the working space of the sub-surface adjustment mechanism at different elevation angles are obtained; a sub-surface adjustment mechanism error kinematic model, a sub-surface adjustment mechanism error model and an identification model are established to identify geometric parameter errors; the kinematic model parameter geometric error is compensated; by establishing a mapping model between the equivalent error of the driving rod and the posture error of the moving platform, the error is predicted and the length of the driving rod of the antenna sub-surface adjustment mechanism is corrected to compensate for the posture error of the moving platform.

[0007] The present invention is achieved through the following technical solutions.

[0008] One aspect of the present invention provides a method for compensating for the full elevation angle error of the secondary surface adjustment mechanism of an omnidirectional movable radio telescope, comprising:

[0009] Obtain n measurement pose points in the working space of the radio telescope sub-surface parallel adjustment mechanism;

[0010] Determine the posture transformation relationship between the fixed platform of the sub-surface adjustment mechanism and the measurement coordinate system and the posture transformation relationship between the target point and the moving platform;

[0011] Measure the target on the moving platform at a 90° elevation angle and obtain the position and posture parameters of n measurement points on the moving platform;

[0012] Adjust the pitch angle of the fixed platform, obtain n position measurement results of the moving platform in the working space of the secondary surface adjustment mechanism at different pitch angles, and establish an error data set;

[0013] The kinematic inverse solution model and geometric error model of the sub-surface adjustment mechanism are established, and the geometric parameter errors of the sub-surface adjustment mechanism are identified based on the n posture measurement results of the moving platform at a 90° elevation angle.

[0014] Modify the control parameters of the error kinematics model of the secondary surface adjustment mechanism;

[0015] Based on the multi-layer perceptron neural network algorithm, a mapping model between the equivalent error of the driving rod at all elevation angles and the attitude error of the moving platform is established to predict the equivalent error of the driving rod at any elevation angle of the antenna and any posture of the moving platform.

[0016] The equivalent error of the driving rod is used to correct the length of the driving rod and compensate for the position error of the moving platform in real time.

[0017] Another aspect of the present invention provides a device for compensating the full elevation angle error of the secondary surface adjustment mechanism of an omnidirectional movable radio telescope according to the method, comprising:

[0018] The error data set establishment module is used to obtain n measurement pose points in the workspace of the radio telescope's sub-surface parallel adjustment mechanism; determine the pose conversion relationship between the fixed platform and the measurement coordinate system and the pose conversion relationship between the target point and the moving platform; obtain the pose parameters of n measurement points of the moving platform; obtain n pose measurement results of the moving platform in the workspace of the sub-surface adjustment mechanism at different pitch angles;

[0019] The geometric error compensation module is used to establish the inverse kinematics model of the secondary surface adjustment mechanism and the error model of the secondary surface adjustment mechanism; identify the initial geometric parameter error of the secondary surface adjustment mechanism; and modify the control parameters of the error kinematics model of the secondary surface adjustment mechanism;

[0020] The non-geometric error compensation module is used to establish a mapping model between the equivalent error of the driving rod and the attitude error of the moving platform, predict the equivalent error of the driving rod under any pitch angle of the antenna and any posture of the moving platform; correct the length of the driving rod and compensate for the posture error of the moving platform in real time.

[0021] Another aspect of the present invention provides an electronic device, including a processor and a memory;

[0022] The memory is used to store a computer program, wherein the computer program includes program instructions;

[0023] The processor is used to call the program stored in the memory to execute the steps of the full elevation angle error compensation method of the sub-surface adjustment mechanism of the omnidirectional movable radio telescope.

[0024] The present invention adopts the above technical solution, which has the following beneficial effects:

[0025] In view of the non-geometric errors at different pitch angles, a multi-layer perceptron neural network (MLP) algorithm is designed. The neural network is used to predict the equivalent error of the driving rod at any pitch angle of the antenna and any posture of the moving platform. The correction amount is superimposed with the ideal driving rod length and assigned to the driving rod. The posture error of the moving platform is compensated in real time by correcting the length of the driving rod. The present invention effectively compensates for the posture error of the moving platform, especially the non-geometric error within the full elevation angle range, by performing graded compensation for geometric errors and non-geometric errors. Through precise error compensation and adjustment, the need for frequent calibration of the sub-reflector of the radio telescope is greatly reduced, the maintenance cost in long-term use is reduced, and the overall reliability and stability of the equipment are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of the present application, and do not constitute an improper limitation of the present invention. In the drawings:

[0027] Figure 1The present invention is a flow chart of the method for compensating the full elevation angle error of the secondary surface adjustment mechanism of an omnidirectional movable radio telescope;

[0028] Figure 2 It is a schematic diagram of a full elevation angle error compensation device of an omnidirectional movable radio telescope sub-surface adjustment mechanism shown in an embodiment of the present invention;

[0029] Figure 3 It is a schematic diagram of the structure of an error compensation electronic device shown in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. The exemplary embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0031] The embodiment of the present invention provides a method for compensating the full elevation angle error of the sub-surface adjustment mechanism of an omnidirectional movable radio telescope. The present invention divides the error sources that affect the absolute positioning accuracy of the end of the sub-surface adjustment mechanism into geometric errors and non-geometric errors. The geometric errors include connecting rod length, axis parallelism, joint offset, manufacturing error and assembly error, etc., while the non-geometric errors include load deformation, temperature, gear clearance and dynamic parameter errors, and non-linear errors caused by the change of the force of the sub-surface adjustment mechanism at different elevation angles of the antenna. The present invention performs graded compensation for geometric errors and non-geometric errors, compensates for geometric errors by modifying the control parameters of the kinematic model of the sub-surface adjustment mechanism, and compensates for non-geometric errors by correcting the equivalent error of the driving rod in real time.

[0032] like Figure 1 As shown, the method for compensating the posture error of the auxiliary surface adjustment mechanism of the omnidirectional movable radio telescope at full elevation angles of the present invention comprises the following steps:

[0033] Step 1: Collect error data set.

[0034] The specific steps include:

[0035] S11, according to the working space of the sub-surface adjustment mechanism of the radio telescope, a uniform selection method is adopted to uniformly select n measurement pose points in the working space of the six-degree-of-freedom sub-surface parallel adjustment mechanism;

[0036] S12, determining the posture conversion relationship of the fixed platform of the sub-surface adjustment mechanism relative to the measurement coordinate system and the posture conversion relationship of the target point relative to the moving platform;

[0037] The pose transformation relationship of the measurement coordinate system is:

[0038] In the formula, They are the rotation matrix and translation matrix of the measurement coordinate system relative to the fixed platform coordinate system;

[0039] The position transformation relationship of the target point relative to the moving platform is:

[0040] In the formula, They are respectively the rotation matrix and translation matrix of the target point coordinate system relative to the moving platform coordinate system.

[0041] S13, adjusting the moving platform to the above n measurement posture points at an elevation angle of 90°, measuring the target on the moving platform, and obtaining posture measurement results of the n moving platforms;

[0042] S14, uniformly adjust the pitch angle Φ of the fixed platform within the range of 0°-90°, obtain the position and posture measurement results of n moving platforms in the working space of the sub-surface adjustment mechanism at different pitch angles, and decompose the overall working space to obtain the position and posture errors of the moving platforms at different elevation angles, so as to quickly and accurately predict the position and posture error compensation amount of the sub-surface adjustment mechanism of the overall working space.

[0043] Step 2: Geometric error compensation.

[0044] The specific steps include:

[0045] S21, establish the kinematic model of the secondary surface adjustment mechanism based on the closed-loop vector method.

[0046] The coordinate systems P-xyz and B-xyz are established with the centers of the moving platform and the fixed platform as the origins; the center of the Hooke joint of the moving platform is expressed in the moving coordinate system as The center of the fixed platform Hooke's joint is expressed in the fixed coordinate system as i=1,2,…,6 represents the numbers of the 6 legs.

[0047] The kinematic inverse solution model of the secondary surface adjustment mechanism is:

[0048]

[0049] In the formula, l i is the length of the i-th rod, for point to The unit vector of is the coordinate of the center of the Hooke's hinge on the fixed platform; is the coordinate of the center of the spherical joint on the moving platform. The center of the Hooke joint of the moving platform is expressed in the moving platform coordinate system as The center of the fixed platform Hooke's joint is expressed in the fixed platform coordinate system as i=1,2,…,6 represents the numbers of the six legs. B , It is the rotation matrix and translation matrix of the moving platform coordinate system relative to the fixed platform coordinate system.

[0050] S22, based on the matrix method, the error model of the secondary surface adjustment mechanism is established as:

[0051]

[0052] Where: Δe is the position error vector of the moving platform, ΔL=[ΔL1ΔL2…ΔL6] T ∈R 6×1 , represents the rod length error vector of the six legs; is the upper and lower hinge position error vector, J s is the error Jacobian matrix, and the formula is as follows:

[0053]

[0054] J p is the hinge position error transformation matrix, the formula is as follows:

[0055]

[0056] S23, based on the error data set at 90° elevation, the Levenberg-Marquardt algorithm is used to identify the geometric parameter errors of the sub-surface adjustment mechanism, namely the initial rod length error and the upper and lower hinge position errors:

[0057] Initial rod length error: ΔL = [ΔL1 ΔL2…ΔL6] T ∈R 6×1

[0058] Upper and lower hinge position error:

[0059]

[0060] The iterative optimization steps of the Levenberg-Marquardt algorithm are as follows:

[0061] S231, calculate the geometric parameter matrix of the secondary surface adjustment mechanism Among them J s is the error Jacobian matrix, J p is the hinge point position error transformation matrix;

[0062] S232, calculate the residual vector of the geometric parameters of the secondary surface adjustment mechanism: k In this case, k is the number of cycles, and the difference vector r between the model prediction output and the actual measured value is calculated. k ;

[0063] The geometric parameter error P of the secondary surface adjustment mechanism is updated using the following formula: k = [ΔLΔd], i.e. 6 initial rod length errors and 36 hinge position errors:

[0064] P k+1 =P k -(HT H+λD) -1 H T r k

[0065] Where D is J T The diagonal elements of J, λ is a non-negative adjustment factor;

[0066] S233, adjusting the non-negative adjustment factor λ: adjusting the value of λ according to the error reduction to optimize the convergence speed and stability;

[0067] S234, using the new parameter P k+1 Verify the motion accuracy of the secondary surface adjustment mechanism to ensure that the error is within an acceptable range.

[0068] S24, modifying control parameters of the kinematic model of the sub-surface adjustment mechanism based on the identified geometric error of the sub-surface adjustment mechanism.

[0069] Step 3: Real-time compensation of non-geometric errors at all elevation angles.

[0070] The specific steps include:

[0071] S31, based on the multi-layer perceptron neural network algorithm, a mapping model between the equivalent error of the driving rod and the attitude error of the moving platform at all elevation angles is established.

[0072] When the position and posture of the moving platform changes, the elongation of the driving rod can be calculated by the change of the coordinates of the hinge in space. Therefore, the hinge error can also be converted into an easily controllable driving rod error, and compensation can be made by adjusting the length of the driving rod. Therefore, formula (2) can be simplified as follows:

[0073]

[0074] Where ΔL' is the equivalent error of the driving rod, which includes two errors: hinge error and driving rod error.

[0075] By adjusting the length of the driving rod, the position error of the hinge point of the moving and fixed platforms and the influence of the driving rod error on the posture error of the moving platform can be compensated.

[0076] S32, using neural network to predict the equivalent error of the driving rod at any pitch angle of the antenna and any posture of the moving platform.

[0077] The sine sin(Φ) and cosine cos(Φ) of the antenna elevation angle Φ and the normalized theoretical pole length L = (l1, l2, …, l6) are used as the input of the neural network, and the pole length equivalent error ΔL' is used as the network output. The Adam optimizer is used for training. In order to prevent overfitting, L2 regularization is introduced in the hidden layer, and a dynamic learning rate scheduling strategy is applied during the training process to adjust the learning rate according to the performance of the model during training.

[0078] The driving rod correction value ΔL' is superimposed on the theoretical rod length L and assigned to the controller. The position error of the moving platform is compensated in real time by correcting the rod length error.

[0079] There is a continuous mapping relationship between the ideal rod length L and the equivalent error ΔL' of the driving rod in the joint space. Therefore, the two can be used as the input and output of the MLP neural network at different elevation angles.

[0080] The MLP neural network maps input to output through several hidden layers. Each neuron in the hidden layer processes data through linear transformation and nonlinear activation function. The output of the hidden layer is:

[0081] h m =f(W m ·h m-1 +b m )(4)

[0082] where h m is the output of hidden layer m, W m is the weight of hidden layer m, b m is the bias term of the hidden layer m, f(·) is the activation function, which is the ReLU activation function in the present invention, and the final output layer obtains the predicted posture error through linear transformation:

[0083] ΔL'=W o ·h M +b o (5)

[0084] Where ΔL' is the predicted pose error, W o is the weight matrix of the output layer, b o is the bias term of the output layer, and M is the last hidden layer.

[0085] In order to optimize the model, the present invention uses mean square error (MSE) as the loss function. The loss function is defined as follows:

[0086]

[0087] In the formula is the expected output, ΔL' is the network output, is the input vector, and wi is the weight from hidden layer to output layer, c i (i=1, 2, ..., N) is the center point position, and σ is the expansion constant parameter of the Gaussian basis function.

[0088] S33, using the equivalent error of the driving rod to correct the length of the driving rod, and to compensate for the posture error of the moving platform in real time.

[0089] The trained MLP model is integrated into the control system. The equivalent error of the driving rod is calculated according to the real-time antenna elevation angle and the posture of the moving platform. The driving rod correction value ΔL' is superimposed on the theoretical rod length L and assigned to the controller. The posture error of the moving platform is compensated in real time by correcting the rod length error.

[0090] Real-time compensation of the moving platform posture error, the specific steps are as follows:

[0091] S331, using the sine sin(Φ) and cosine cos(Φ) of the antenna elevation angle Φ and the normalized theoretical pole length L = (l1, l2, ..., l6) as the input of the neural network, using the pole length equivalent error ΔL' as the network output, the data set measured at all elevation angles is divided into a training data set, a test data set, and a validation data set.

[0092] S332, for the selected posture point, the theoretical posture is obtained by inverse kinematics to obtain the theoretical rod length L, and the error between the theoretical posture and the measured posture is obtained by inverse kinematics to obtain the rod length equivalent error ΔL';

[0093] S333, select mean squared error (MSE) as the loss function to calculate the difference between the predicted value and the actual value;

[0094] S334, use Adam optimizer to train a multilayer perceptron neural network model.

[0095] In order to prevent overfitting, L2 regularization is introduced in the hidden layer, and a dynamic learning rate scheduling strategy is applied during the training process to adjust the learning rate according to the performance of the model during the training process. After the training is completed, the model parameters are saved;

[0096] S335, integrate the trained multi-layer perceptron neural network model in the control system, calculate the equivalent error of the driving rod length according to the real-time antenna elevation angle and the dynamic platform posture, and superimpose the correction value of the driving rod length equivalent error ΔL' with the theoretical rod length L and assign it to the controller. By correcting the equivalent error of the driving rod length, the dynamic platform posture error is compensated in real time.

[0097] According to an exemplary embodiment of the present invention, Figure 2 As shown, a device 100 for compensating the full elevation angle error of the secondary surface adjustment mechanism of an omnidirectional movable radio telescope, used to implement the method, comprises:

[0098] The error data set establishment module 110 is used to obtain n measurement pose points in the working space of the sub-surface parallel adjustment mechanism of the radio telescope; determine the pose conversion relationship between the fixed platform and the measurement coordinate system and the pose conversion relationship between the target point and the moving platform; obtain the pose parameters of the n measurement points of the moving platform; obtain the n pose measurement results of the moving platform in the working space of the sub-surface adjustment mechanism at different pitch angles;

[0099] The geometric error compensation module 120 is used to establish the inverse kinematics model of the secondary surface adjustment mechanism and the error model of the secondary surface adjustment mechanism; identify the initial geometric parameter error of the secondary surface adjustment mechanism; and modify the control parameters of the error kinematics model of the secondary surface adjustment mechanism;

[0100] The non-geometric error compensation module 130 is used to establish a mapping model between the equivalent error of the driving rod and the attitude error of the moving platform, predict the equivalent error of the driving rod under any pitch angle of the antenna and any posture of the moving platform; correct the length of the driving rod and compensate for the posture error of the moving platform in real time.

[0101] According to an exemplary embodiment of the present invention, Figure 3 As shown, the present invention provides an electronic device 200 for implementing the method for compensating the full elevation angle error of the sub-surface adjustment mechanism of an omnidirectional movable radio telescope, comprising: an acquisition card 210, a memory 220, a communication bus 230, and a processor 240.

[0102] The computer device includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used to implement the operation of a method for compensating the full elevation angle error of the secondary surface adjustment mechanism of an omnidirectional movable radio telescope.

[0103] The present invention is not limited to the above-mentioned embodiments. On the basis of the technical solution disclosed in the present invention, technicians in this field can make some substitutions and deformations to some technical features therein according to the disclosed technical content without creative labor, and these substitutions and deformations are all within the protection scope of the present invention.

Claims

1. A method for compensating the full elevation angle error of the secondary adjustment mechanism of an omnidirectionally movable radio telescope, characterized in that: include: Obtain n measurement pose points in the working space of the radio telescope sub-surface parallel adjustment mechanism; Determine the posture transformation relationship between the fixed platform of the sub-surface adjustment mechanism and the measurement coordinate system and the posture transformation relationship between the target point and the moving platform; Measure the target on the moving platform at a 90° elevation angle and obtain the position and posture parameters of n measuring points on the moving platform; Adjust the pitch angle of the fixed platform, obtain n position measurement results of the moving platform in the working space of the secondary surface adjustment mechanism at different pitch angles, and establish an error data set; Establish the kinematic inverse model and geometric error model of the secondary surface adjustment mechanism; Based on the n pose measurement results of the moving platform at a 90° elevation angle, the geometric parameter errors of the sub-surface adjustment mechanism are identified; Modify the control parameters of the kinematic model of the secondary surface adjustment mechanism and correct the geometric errors; Based on the multi-layer perceptron neural network algorithm, a mapping model between the equivalent error of the driving rod at all elevation angles and the attitude error of the moving platform is established to predict the equivalent error of the driving rod at any elevation angle of the antenna and any posture of the moving platform. The equivalent error of the driving rod is used to correct the length of the driving rod and compensate for the position error of the moving platform in real time.

2. The method for compensating the full elevation angle error of the secondary adjustment mechanism of an omnidirectional movable radio telescope according to claim 1, characterized in that: The pose transformation relationship of the measurement coordinate system is: In the formula, They are the rotation matrix and translation matrix of the measurement coordinate system relative to the fixed platform coordinate system; The position transformation relationship of the target point relative to the moving platform is: In the formula, They are respectively the rotation matrix and translation matrix of the target point coordinate system relative to the moving platform coordinate system.

3. The method for compensating the full elevation angle error of the secondary adjustment mechanism of an omnidirectional movable radio telescope according to claim 1, characterized in that: The kinematic inverse solution model of the secondary surface adjustment mechanism is as follows: In the formula, l i is the length of the i-th rod, for point to The unit vector of is the coordinate of the center of the Hooke's hinge on the fixed platform, is the center coordinate of the spherical joint on the moving platform, p B , They are the rotation matrix and translation matrix of the moving platform coordinate system relative to the fixed platform coordinate system respectively.

4. The method for compensating the full elevation angle error of the secondary adjustment mechanism of an omnidirectional movable radio telescope according to claim 1, characterized in that: The geometric error model of the secondary surface adjustment mechanism is as follows: Where: Δe is the position error vector of the moving platform, ΔL represents the length error vector of the six legs, Δd is the position error vector of the upper and lower hinges, J s is the error Jacobian matrix, J p is the hinge point position error transformation matrix.

5. The method for compensating the full elevation angle error of the secondary adjustment mechanism of an omnidirectional movable radio telescope according to claim 1, characterized in that: Based on the n posture measurement results of the moving platform at an elevation angle of 90°, the Levenberg-Marquardt algorithm is used to iteratively optimize the geometric parameter errors of the sub-surface adjustment mechanism. The geometric parameter errors include the initial rod length error and the hinge position error of the moving platform.

6. The method for compensating the full elevation angle error of the auxiliary adjustment mechanism of an omnidirectional movable radio telescope according to claim 5, characterized in that: The iterative optimization steps of the Levenberg-Marquardt algorithm are as follows: Calculate the error Jacobian matrix of the sub-surface adjustment mechanism and the hinge position error transformation matrix; Calculate the residual vector of the geometric parameters of the sub-surface adjustment mechanism, and calculate the difference vector between the predicted output of the sub-surface adjustment mechanism geometric error model and the actual measured value; Update the geometric parameter errors of the sub-surface adjustment mechanism, including initial rod length error, hinge position error, diagonal elements and non-negative adjustment factors; According to the error reduction, the non-negative adjustment factor of the model after the geometric error of the sub-surface adjustment mechanism is adjusted to optimize the convergence speed and stability.

7. The method for compensating the full elevation angle error of the secondary adjustment mechanism of an omnidirectionally movable radio telescope according to claim 1, characterized in that: The mapping model between the equivalent error of the driving rod and the attitude error of the moving platform at full elevation angle is as follows: Where: Δe is the position error vector of the moving platform, ΔL' is the equivalent error of the driving rod, J s is the error Jacobian matrix.

8. The method for compensating the full elevation angle error of the secondary adjustment mechanism of an omnidirectionally movable radio telescope according to claim 1, characterized in that: The equivalent error of the driving rod is used to correct the length of the driving rod and compensate the posture error of the moving platform in real time, including: The sine and cosine of the antenna elevation angle and the normalized theoretical pole length are used as the input of the neural network, and the equivalent error of the pole length is used as the network output. The data set measured at all elevation angles is divided into a training data set, a test data set, and a validation data set. For the selected posture points, the theoretical posture is used to obtain the theoretical rod length through the inverse kinematic solution, and the error between the theoretical posture and the measured posture is used to obtain the rod length equivalent error through the inverse kinematic solution; Select mean square error as the loss function to calculate the difference between the predicted value and the actual value; Use Adam optimizer to train a multilayer perceptron neural network model; A trained multi-layer perceptron neural network model is integrated into the control system. The equivalent error of the driving rod length is calculated according to the real-time antenna elevation angle and the moving platform posture. The driving rod length equivalent error correction amount is superimposed with the theoretical rod length and assigned to the controller. By correcting the rod length equivalent error, the moving platform posture error can be compensated in real time.

9. A device for compensating the full elevation angle error of the secondary surface adjustment mechanism of an omnidirectionally movable radio telescope according to the method of any one of claims 1 to 8, characterized in that: include: The error data set establishment module is used to obtain n measurement pose points in the working space of the radio telescope sub-surface parallel adjustment mechanism; Determine the posture conversion relationship of the fixed platform relative to the measurement coordinate system and the posture conversion relationship of the target point relative to the moving platform; obtain the posture parameters of n measurement points of the moving platform; obtain n posture measurement results of the moving platform in the working space of the sub-surface adjustment mechanism at different pitch angles; The geometric error compensation module is used to establish the inverse kinematics model of the secondary surface adjustment mechanism and the error model of the secondary surface adjustment mechanism; identify the initial geometric parameter error of the secondary surface adjustment mechanism; and modify the control parameters of the error kinematics model of the secondary surface adjustment mechanism; The non-geometric error compensation module is used to establish a mapping model between the equivalent error of the driving rod and the attitude error of the moving platform, predict the equivalent error of the driving rod under any pitch angle of the antenna and any posture of the moving platform; correct the length of the driving rod and compensate for the posture error of the moving platform in real time.

10. An electronic device, characterized in that: Including processor and memory: The memory is used to store a computer program, wherein the computer program includes program instructions; The processor is used to call the program stored in the memory to execute the steps of the full elevation angle error compensation method of the sub-disc adjustment mechanism of the omnidirectional movable radio telescope according to any one of claims 1 to 8.

Images