Six-degree-of-freedom motion platform with high repeat positioning function

Abstract

The application discloses a six-degree-of-freedom motion platform with high repeated positioning function and belongs to the technical field of mechanical intelligentization. The application solves the problem of low positioning accuracy caused by mechanical structure error and insufficient driving system precision in the prior art, and through a BP neural network model, pose data fed back in real time can be used to more accurately calculate actuating cylinder elongation data, and the elongation of the actuating cylinder is controlled based on the elongation data, so that the error in the control process is reduced, and the control precision of the platform is improved. Based on the sensor, the pose data is fed back in real time, the elongation of the actuating cylinder is dynamically adjusted, a closed-loop control system is formed, deviations can be found and corrected in time, and the reliability and stability of the intelligent control unit are improved. Through the priority search decision tree model, known pose data and actuating cylinder elongation data can be quickly responded, so that the calculation time is reduced, the control efficiency and response speed are improved.

Description

Technical Field

[0001] This invention relates to the field of mechanical intelligence technology, specifically a six-degree-of-freedom motion platform with high repeatability positioning function. Background Technology

[0002] A six-degree-of-freedom (6DOF) motion platform is a device capable of achieving six degrees of freedom (translation along the X, Y, and Z axes and rotation around the X, Y, and Z axes) in three-dimensional space. It is widely used in robotics, aerospace, automotive manufacturing, medical equipment, and other fields. The high repeatability of a six-DOF motion platform refers to its ability to repeatedly reach the target position and orientation with extremely high accuracy when performing the same motion task multiple times; it is one of the important indicators for evaluating platform performance. The high repeatability of a six-DOF motion platform relies on the precise mapping relationship between its pose data and the actuator extension. However, existing six-DOF motion platforms have certain limitations in repeatability accuracy and cannot meet the requirements for high-precision repeatability.

[0003] For example, some platforms experience low positioning accuracy during repeated positioning due to factors such as errors in the mechanical structure, insufficient precision of the drive system, and imperfections in the control algorithm, which affects the performance and reliability of the equipment.

[0004] Therefore, to address the shortcomings of existing requirements, we proposed a six-degree-of-freedom motion platform with high repeatability. Summary of the Invention

[0005] The purpose of this invention is to provide a six-degree-of-freedom motion platform with high repeatability. By using a BP neural network model to calculate the actuator cylinder elongation more accurately based on real-time feedback pose data, the elongation of the actuator cylinder can be controlled accordingly, thereby reducing errors in the control process and improving the platform's control precision. Furthermore, based on timed feedback pose data from sensors, the elongation of the actuator cylinder can be dynamically adjusted, forming a closed-loop control system that can promptly detect and correct deviations, thus improving the reliability and stability of the intelligent control unit. By using a priority retrieval decision tree model, known pose data and actuator cylinder elongation data can be quickly responded to, thereby reducing computation time, improving control efficiency and response speed, and solving the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A six-degree-of-freedom motion platform with high repeatability positioning function, the platform includes: a platform body and an intelligent control unit, the intelligent control unit includes an angle calculation module, an intelligent positioning module and a pose feedback module; The angle calculation module is configured to analyze the nonlinear mapping relationship between the historical pose data of the platform body and the historical actuator elongation data based on the BP neural network model, so that the BP neural network model can calculate the actuator elongation data based on the pose data. The intelligent positioning module is configured to collect pose data in real time based on the pose sensor installed on the platform, input it into the BP neural network model for calculation, and output the elongation data of the actuator cylinder as a control command for the platform body to adjust the elongation of the actuator cylinder. Furthermore, the platform also includes: a pose feedback module; The posture feedback module is configured to dynamically adjust the extension of the actuator cylinder based on the posture data fed back by the posture sensor in real time, thereby forming a closed-loop control.

[0007] Furthermore, the angle calculation module includes: The model building module is configured to construct a three-layer BP neural network based on historical pose data and historical actuator elongation data, and to establish the activation function. The three-layer BP neural network includes an input layer, a hidden layer, and an output layer. The input layer is used to input the pose data of the platform itself; the hidden layer is used to select an appropriate number of neurons according to the complexity of the problem; and the output layer is used to output the elongation data of the actuator. The model training module is configured to take historical pose data as input and actuator elongation data as output, and input them into a BP neural network for training; calculate the output value through forward propagation and compare it with the actual value to obtain the error; adjust the network parameters through back propagation and repeat the iteration until the error converges; The model testing module is configured to use historical pose data that was not used in training as input to verify whether the model's calculation results are consistent with the historical actuator elongation data.

[0008] Furthermore, the angle calculation module also includes: The model validation module is configured to calculate the error between the predicted value and the actual value based on the mean square error method to evaluate the accuracy of the model. The model optimization module is configured to set an online learning mechanism, which collects pose data in real time based on pose sensors and updates neural network parameters and network structure periodically.

[0009] Furthermore, the error between the predicted and actual values ​​is calculated based on the mean square error method, specifically as follows: The system analyzes multiple sets of historical pose data using a backpropagation neural network to obtain the predicted value of the historical actuator elongation data for each set of historical pose data, and also obtains the actual value of the historical actuator elongation data for each set of historical pose data. The error between the predicted and actual values ​​is calculated using the mean square error method, with the following formula: in, Expressed as mean square error; This is represented as a combination of historical pose data and historical actuator elongation data; Represented as the first The actual value of the elongation of the actuator cylinder; Represented as the first Predicted value of the elongation of the actuator cylinder; A lower MSE value indicates a smaller error between the predicted and actual values, and a higher model accuracy. A high MSE value indicates a large error between the predicted and actual values, and thus a low model accuracy.

[0010] Furthermore, the intelligent control unit also includes: The data processing module is configured to clean the pose data and actuator elongation data, remove noise and outliers, and then normalize the pose data to the range of [0,1] or [-1,1]. The data storage module is configured to cluster and save each set of pose data and the corresponding actuator cylinder elongation data based on a decision tree model, so that each pose data has corresponding actuator cylinder elongation data.

[0011] Furthermore, the data storage module includes: The priority adjustment module is configured to, after acquiring the pose data of the platform body, prioritize retrieving the same pose data and corresponding actuator cylinder extension data from the decision tree model; if found, control the extension of the actuator cylinder based on the actuator cylinder extension data; if not found, calculate the new actuator cylinder extension data based on the pose data using the BP neural network model. The data update module is configured to update the platform's new pose data and corresponding actuator elongation data in real time, and optimize the content of the decision tree model.

[0012] Furthermore, the intelligent positioning module includes: The device integration module is configured to embed a BP neural network model into the intelligent control unit and integrate the pose sensor with the intelligent control unit based on wireless transmission technology, so that the real-time pose data of the pose sensor can be transmitted to the intelligent control unit for analysis. The model implementation module is configured to analyze and calculate real-time pose data based on a BP neural network model and output the actuator elongation data.

[0013] Furthermore, the intelligent positioning module also includes: The instruction drive module is configured to convert the calculation results based on the BP neural network model into a drive signal, and control the extension of the actuator cylinder based on the signal.

[0014] Furthermore, the pose feedback module includes: The timing feedback module is configured to periodically feed back real-time pose data to the intelligent control unit based on the pose sensor after the six-degree-of-freedom motion platform starts working. This data is used to control the extension of the actuator cylinder to achieve the expected pose and ensure the high repeatability of the platform body. The secondary adjustment module is configured to determine whether the extension of the actuator cylinder has reached the expected pose based on the feedback result; if not, it recalculates the current pose data based on the BP neural network model to obtain new actuator cylinder extension data; and adjusts the extension of the actuator cylinder based on this data until the extension of the actuator cylinder reaches the expected pose.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In this invention, the BP neural network model can more accurately calculate the actuator elongation data based on real-time feedback pose data, and use this to control the elongation of the actuator, thereby reducing errors in the control process and improving the control accuracy of the platform; and based on the timed feedback pose data from the sensor, the elongation of the actuator can be dynamically adjusted, thus forming a closed-loop control system that can monitor and adjust the state of the actuator in real time, promptly detect and correct deviations, thereby improving the reliability and stability of the intelligent control unit.

[0016] 2. In this invention, by prioritizing the retrieval of the decision tree model, it is possible to quickly respond to known pose data and actuator elongation data, thereby reducing computation time and improving control efficiency and response speed. Attached Figure Description

[0017] Figure 1 This is a diagram showing the composition of the intelligent control unit module for the six-degree-of-freedom motion platform of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] To address the technical problem in existing technologies where low positioning accuracy, impacting equipment performance and reliability, stems from factors such as mechanical structure errors, insufficient drive system precision, and imperfect control algorithms, please refer to [link to relevant documentation]. Figure 1 This embodiment provides the following technical solution: A six-degree-of-freedom motion platform with high repeatability positioning function. The platform includes: platform body and intelligent control unit. The platform body includes: base, upper platform, six actuators, universal hinges and posture sensor. The upper platform is connected to the base through the six actuators. The two ends of the actuators are connected to the base and the upper platform through universal hinges respectively, so as to realize six degrees of freedom of motion.

[0020] The intelligent control unit includes an angle calculation module, an intelligent positioning module, and a pose feedback module. The intelligent control unit also includes a data processing module configured to clean the pose data and actuator elongation data, removing noise and outliers, and then normalizing the pose data to the range of [0,1] or [-1,1] to improve the training efficiency and generalization ability of the BP neural network. Specifically, it collects historical pose data of the platform itself, including X, Y, and Z coordinates and rotation angles around the X, Y, and Z axes, as well as the corresponding actuator elongation data. The pose data and actuator elongation data are then normalized to eliminate dimensional differences and improve training efficiency. The dataset is divided into a training set and a test set, typically in a ratio of 70% training set and 30% test set.

[0021] The data storage module is configured to cluster and save each set of pose data and the corresponding actuator cylinder elongation data based on a decision tree model, so that each pose data has corresponding actuator cylinder elongation data.

[0022] The angle calculation module is configured to analyze the nonlinear mapping relationship between the historical pose data and historical actuator elongation data of the platform itself based on a BP neural network model, enabling the BP neural network model to calculate the actuator elongation data based on the pose data; the angle calculation module includes: The model building module is configured to construct a three-layer BP neural network based on historical pose data and historical actuator elongation data, and to establish activation functions. The three-layer BP neural network includes an input layer, a hidden layer, and an output layer. The input layer is used to input the pose data of the platform body, such as X, Y, and Z coordinates and rotation angles around the X, Y, and Z axes. The hidden layer uses the ReLU function to select an appropriate number of neurons based on the complexity of the problem. The output layer uses a linear function to output the actuator elongation data. Specifically, by analyzing the nonlinear mapping relationship between the historical pose data of the platform body and the historical actuator elongation data through the BP neural network model, and using this model to calculate the actuator elongation data based on the pose data, the control accuracy and repeatability of the six-degree-of-freedom motion platform can be significantly improved.

[0023] The model training module is configured to take historical pose data as input and actuator cylinder elongation data as output, and input them into a BP neural network for training. The output value is calculated through forward propagation and compared with the actual value to obtain the error. The network parameters are adjusted through backpropagation, and the iteration is repeated until the error converges to avoid getting trapped in local optima. Specifically, by inputting the normalized training set into the BP neural network model, the hidden layer output is calculated through the weights and biases from the input layer to the hidden layer, and the predicted value of the output layer is calculated through the weights and biases from the hidden layer to the output layer. The output layer obtains the predicted value of the actuator cylinder elongation, which is used as the training result.

[0024] The model testing module is configured to use historical pose data that was not used in training as input to verify whether the model's calculation results are consistent with historical actuator elongation data. Specifically, by inputting the test set into the BP neural network model, the final performance of the model is evaluated, and the prediction accuracy and error are calculated to ensure the model's performance on unseen data. The trained BP neural network model is saved as a file and deployed to the control system for real-time input of pose data to calculate actuator elongation and achieve precise control of the platform.

[0025] The model validation module is configured to calculate the error between predicted and actual values ​​based on the mean squared error method to evaluate the model's accuracy; specifically: The system analyzes multiple sets of historical pose data using a backpropagation neural network to obtain the predicted value of the historical actuator elongation data for each set of historical pose data, and also obtains the actual value of the historical actuator elongation data for each set of historical pose data. The error between the predicted and actual values ​​is calculated using the mean square error method, with the following formula: in, Expressed as mean square error; This is represented as a combination of historical pose data and historical actuator elongation data; Represented as the first The actual value of the elongation of the actuator cylinder; Represented as the first Predicted value of the elongation of the actuator cylinder; A lower MSE value indicates a smaller error between the predicted and actual values, and a higher model accuracy. A high MSE value indicates a large error between the predicted and actual values, and thus a low model accuracy.

[0026] The model optimization module is configured to set an online learning mechanism, which collects pose data in real time based on pose sensors and updates neural network parameters and network structure periodically to adapt to the dynamic changes of the system and improve the prediction accuracy and control performance of the model.

[0027] In one embodiment, it is assumed that there are the following three sets of historical pose data and corresponding actuator elongation data:

[0028] Table 1: Historical Data Prediction Results The MSE is calculated as follows: Based on the above calculation results, the low MSE value indicates that the model has high prediction accuracy.

[0029] By using the pose data analysis method of BP neural network, evaluating the model accuracy using mean square error, and combining it with an online learning mechanism, the prediction accuracy and control performance of the intelligent control unit can be effectively improved.

[0030] The intelligent positioning module is configured to collect pose data in real time based on pose sensors installed on the platform, input the data into a BP neural network model for calculation, and output the elongation data of the actuator cylinder as a control command for the platform body to adjust the elongation of the actuator cylinder; the intelligent positioning module includes: The device integration module is configured to convert the trained BP neural network model into a format executable by embedded devices and embed it into the intelligent control unit. Based on wireless transmission technologies such as Wi-Fi, Bluetooth, and ZigBee, it integrates and connects the pose sensor and the intelligent control unit, enabling the real-time pose data of the pose sensor to be transmitted to the intelligent control unit for analysis, thus achieving real-time data communication between the pose sensor and the intelligent control unit. Furthermore, by optimizing the communication protocol and data compression technology, it reduces transmission latency and ensures the real-time performance of the data.

[0031] The model implementation module is configured to analyze and calculate real-time pose data based on a BP neural network model and output the actuator elongation data.

[0032] In one embodiment, assuming that the robotic arm control system needs to monitor the robotic arm's pose in real time using a pose sensor and predict the extension of the actuator cylinder to achieve precise control; the process is as follows: Real-time pose data, such as position x, y, z, is sent through a pose sensor; the pose data is normalized to be within the range of [0, 1]; the preprocessed data is input into a BP neural network model to predict the elongation of the actuator cylinder; the predicted value is converted into a PWM signal to control the elongation of the actuator cylinder, thereby achieving accurate prediction and control of the elongation of the actuator cylinder.

[0033] The instruction-driven module is configured to convert the calculation results based on the BP neural network model into a drive signal, and control the extension of the actuator cylinder based on the signal. Specifically, the intelligent control unit receives real-time data sent by the pose sensor and parses it into a format that the neural network model can process; it uses the BP neural network model to predict the pose data and obtain a predicted value for the extension of the actuator cylinder; it generates control commands based on the prediction results and executes the corresponding commands based on the instruction-driven module to control the extension of the actuator cylinder; at the same time, it also records the pose data, prediction results, and control commands for subsequent analysis and optimization.

[0034] The pose feedback module is configured to dynamically adjust the extension of the actuator cylinder based on the pose data fed back by the pose sensor in real time, forming a closed-loop control; the pose feedback module includes: The timing feedback module is configured to periodically feed back real-time pose data to the intelligent control unit based on the pose sensor after the six-degree-of-freedom motion platform starts working. This data is used to control the extension of the actuator cylinder to achieve the expected pose and ensure the high repeatability of the platform body.

[0035] In one embodiment, assuming a six-degree-of-freedom motion platform, it is necessary to periodically feed back real-time pose data via a pose sensor to control the extension of the actuator cylinder, ensuring that the platform reaches the expected pose and maintains high repeatability positioning accuracy. The process is as follows: By configuring a timer, a data request is sent to the pose sensor every 10ms, and real-time pose data is received. The received pose data is sent to the intelligent control unit, which uses a BP neural network model to calculate and generate an adjustment command for the extension of the actuator cylinder. The adjustment command is converted into a PWM signal to control the extension of the actuator cylinder. Through a closed-loop control mechanism, the repeatability of the platform in multiple movements is ensured.

[0036] The secondary adjustment module is configured to determine whether the extension of the actuator cylinder has reached the expected pose based on the feedback result; if not, it recalculates the current pose data based on the BP neural network model to obtain new actuator cylinder extension data; and adjusts the extension of the actuator cylinder based on this data until the extension of the actuator cylinder reaches the expected pose.

[0037] In the above embodiment, after obtaining real-time pose data, the error between the current pose and the expected pose is calculated. If the error exceeds a threshold, the extension of the actuator cylinder is recalculated using a BP neural network model, and the actuator cylinder is adjusted according to the new extension data. The above process is repeated until the pose error is less than the threshold, so that the extension of the actuator cylinder gradually approaches the expected pose, thereby achieving precise adjustment of the actuator cylinder extension and ensuring that the six-degree-of-freedom motion platform reaches the expected pose.

[0038] The beneficial effects of the above are as follows: By using the BP neural network model to calculate the actuator elongation data more accurately based on the real-time feedback pose data, the elongation of the actuator can be controlled accordingly, thereby reducing errors in the control process and improving the control accuracy of the platform; and by using the sensor to provide timed feedback pose data, the elongation of the actuator can be dynamically adjusted, thus forming a closed-loop control system that can monitor and adjust the state of the actuator in real time, promptly detect and correct deviations, thereby improving the reliability and stability of the intelligent control unit.

[0039] Data storage module, including: The priority adjustment module is configured to, after acquiring the platform's pose data, prioritize retrieving identical pose data and corresponding actuator cylinder extension data from the decision tree model. If a match is found, the actuator cylinder extension is controlled based on this extension data. If not, a new actuator cylinder extension is calculated based on the pose data using a BP neural network model. Specifically, the current pose data is input into the decision tree model to check for a matching record. If a matching record is found, the actuator cylinder extension data is directly acquired. If no matching record is found, the actuator cylinder extension is calculated using the BP neural network model. The actuator cylinder extension is adjusted based on the acquired or calculated extension data. By prioritizing the use of the decision tree model for retrieval, the computational overhead of the neural network is reduced. Furthermore, combining the decision tree and neural network models improves the intelligence level of the control unit.

[0040] The data update module is configured to update the platform's new pose data and corresponding actuator elongation data in real time, and optimize the content of the decision tree model.

[0041] The beneficial effects of the above are: by prioritizing the retrieval of the decision tree model, it is possible to quickly respond to known pose data and actuator elongation data, thereby reducing computation time and improving control efficiency and response speed.

[0042] Working principle: The BP neural network model calculates the corresponding actuator cylinder extension data based on the real-time feedback pose data; the calculation result is converted into an execution command to control the extension of the actuator cylinder; and the pose data of the platform body is monitored in real time to ensure the positioning accuracy of the platform body, thereby realizing the high repeatability positioning function of the platform body.

[0043] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "possessing," or any other variations thereof are intended to cover non-exclusive possession, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0044] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that variations, modifications, substitutions and alterations may be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A six-degree-of-freedom motion platform with high repeatability positioning function, characterized in that, The platform includes: a platform body and an intelligent control unit, wherein the intelligent control unit includes an angle calculation module, an intelligent positioning module and a pose feedback module; The angle calculation module is configured to analyze the nonlinear mapping relationship between the historical pose data of the platform body and the historical actuator elongation data based on the BP neural network model, so that the BP neural network model can calculate the actuator elongation data based on the pose data. The intelligent positioning module is configured to collect pose data in real time based on the pose sensor installed on the platform, input it into the BP neural network model for calculation, and output the elongation data of the actuator cylinder as a control command for the platform body to adjust the elongation of the actuator cylinder. The posture feedback module is configured to dynamically adjust the extension of the actuator cylinder based on the posture data fed back by the posture sensor in real time, thereby forming a closed-loop control. The intelligent control unit also includes: The data storage module is configured to cluster and save each set of pose data and the corresponding actuator cylinder elongation data based on the decision tree model, so that there is corresponding actuator cylinder elongation data for different pose data. The data storage module includes: The priority adjustment module is configured to, after acquiring the pose data of the platform body, prioritize retrieving the same pose data and corresponding actuator cylinder extension data from the decision tree model; if found, control the extension of the actuator cylinder based on the actuator cylinder extension data; if not found, calculate the new actuator cylinder extension data based on the pose data using the BP neural network model. The pose feedback module includes: The secondary adjustment module is configured to determine whether the extension of the actuator cylinder has reached the expected pose based on the feedback result; if not, it recalculates the current pose data based on the BP neural network model to obtain new actuator cylinder extension data; and adjusts the extension of the actuator cylinder based on this data until the extension of the actuator cylinder reaches the expected pose.

2. The six-degree-of-freedom motion platform with high repeatability positioning function according to claim 1, characterized in that: The angle calculation module includes: The model building module is configured to construct a three-layer BP neural network based on historical pose data and historical actuator elongation data, and to establish the activation function. The three-layer BP neural network includes an input layer, a hidden layer, and an output layer. The input layer is used to input the pose data of the platform itself; the hidden layer is used to select an appropriate number of neurons according to the complexity of the problem; and the output layer is used to output the elongation data of the actuator. The model training module is configured to take historical pose data as input and actuator elongation data as output, and input them into a BP neural network for training; calculate the output value through forward propagation and compare it with the actual value to obtain the error; adjust the network parameters through back propagation and repeat the iteration until the error converges; The model testing module is configured to use historical pose data that was not used in training as input to verify whether the model's calculation results are consistent with the historical actuator elongation data.

3. The six-degree-of-freedom motion platform with high repeatability positioning function according to claim 2, characterized in that: The angle calculation module further includes: The model validation module is configured to calculate the error between the predicted value and the actual value based on the mean square error method to evaluate the accuracy of the model. The model optimization module is configured to set an online learning mechanism, which collects pose data in real time based on pose sensors and updates neural network parameters and network structure periodically.

4. The six-degree-of-freedom motion platform with high repeatability positioning function according to claim 3, characterized in that: The error between the predicted and actual values ​​is calculated based on the mean square error method, specifically as follows: The system analyzes multiple sets of historical pose data using a backpropagation neural network to obtain the predicted value of the historical actuator elongation data for each set of historical pose data, and also obtains the actual value of the historical actuator elongation data for each set of historical pose data. The error between the predicted and actual values ​​is calculated using the mean square error method, with the following formula: in, Expressed as mean square error; This is represented as a combination of historical pose data and historical actuator elongation data; Represented as the first The actual value of the elongation of the actuator cylinder; Represented as the first Predicted value of the elongation of the actuator cylinder; A lower MSE value indicates a smaller error between the predicted and actual values, and a higher model accuracy. A high MSE value indicates a large error between the predicted and actual values, and thus a low model accuracy.

5. The six-degree-of-freedom motion platform with high repeatability positioning function according to claim 1, characterized in that: The intelligent control unit also includes: The data processing module is configured to clean the pose data and actuator elongation data, remove noise and outliers, and then normalize the pose data to the range of [0,1] or [-1,1].

6. The six-degree-of-freedom motion platform with high repeatability positioning function according to claim 1, characterized in that: The data storage module includes: The data update module is configured to update the platform's new pose data and corresponding actuator elongation data in real time, and optimize the content of the decision tree model.

7. The six-degree-of-freedom motion platform with high repeatability positioning function according to claim 1, characterized in that: The intelligent positioning module includes: The device integration module is configured to embed a BP neural network model into the intelligent control unit and integrate the pose sensor with the intelligent control unit based on wireless transmission technology, so that the real-time pose data of the pose sensor can be transmitted to the intelligent control unit for analysis. The model implementation module is configured to analyze and calculate real-time pose data based on a BP neural network model and output the actuator elongation data.

8. The six-degree-of-freedom motion platform with high repeatability positioning function according to claim 7, characterized in that: The intelligent positioning module further includes: The instruction drive module is configured to convert the calculation results based on the BP neural network model into a drive signal, and control the extension of the actuator cylinder based on the signal.

9. The six-degree-of-freedom motion platform with high repeatability positioning function according to claim 1, characterized in that: The pose feedback module includes: The timing feedback module is configured to periodically feed back real-time pose data to the intelligent control unit based on the pose sensor after the six-degree-of-freedom motion platform starts working. This data is used to control the extension of the actuator cylinder to achieve the expected pose and ensure the high repeatability of the platform body.

Images