Robot motion control system and method based on motion capture and digital twinning

Abstract

The invention relates to the technical field of robots and digital twinning, and discloses a robot motion control system and method based on motion capture and digital twinning. The method comprises the following steps: establishing a robot motion capture module, a digital twin robot model design module, a robot data collection module, a robot data analysis module, a robot data judgment module, a digital twin robot model optimization module and a robot execution module; a robot operation virtual model is constructed in a digital twin robot model design module, a robot data collection module collects data of the modules, a robot data analysis module calculates the data collected by the modules, a robot data judgment module judges whether the current state of a robot meets the expectation or not, and if yes, the robot is started. The digital twin robot model optimization module performs defect supplementation on the scheme of the digital twin robot model, and the robot execution module executes the optimized scheme in actual operation.

Description

Technical Field

[0001] The present invention relates to the fields of robot technology and digital twin technology, and specifically to a robot motion control system and method based on motion capture and digital twin. Background Art

[0002] In the field of robot motion control, motion capture technology is one of the key means to achieve precise control. However, the current motion capture technology still faces some challenges. Although the technology continues to develop, in complex environments, its capture accuracy will be affected by various factors. For example, changes in environmental light may interfere with the performance of optical motion capture systems, resulting in errors in the captured motion data. Especially in high-speed movements, subtle motion details are often difficult to be accurately captured, thus affecting the precise replication of actions by robots. In addition, different motion capture devices have their own accuracy limitations. For application scenarios that require extremely high-precision motion control, such as precision surgical robots, the existing motion capture accuracy may not meet the requirements. Device calibration is also a key issue in motion capture. Before installation and use, motion capture devices need to be precisely calibrated to ensure the accuracy of data. However, the calibration process is usually complex and requires professional technicians to operate, and the calibration parameters may change with the use of the device and environmental changes, and need to be recalibrated regularly. This increases the maintenance cost and operation difficulty of the device and may lead to motion capture errors due to inaccurate calibration. Another challenge is the occlusion problem. When the object of motion capture is partially occluded, the capture system may not be able to obtain complete motion information. In scenarios where multiple people are performing motion capture simultaneously, mutual occlusion between people or occlusion by objects in the environment may cause the capture system to lose information about some marker points, thus affecting the accurate judgment and capture of the entire action. This is a major challenge for applications that require motion capture in complex scenarios. Therefore, improving the anti-interference ability of motion capture technology, simplifying the calibration process, solving the occlusion problem, and improving the overall accuracy of the system have become urgent problems to be solved in this field. Summary of the Invention

[0003] (1) Technical Problems to be Solved

[0004] Aiming at the deficiencies of the prior art, the present invention provides a robot motion control system and method based on motion capture and digital twin, which have the advantages of high capture accuracy, efficient device calibration, and occlusion prediction, and solve the problems of limited capture accuracy of robots, complex device calibration, and occlusion of the object of motion capture in the prior art.

[0005] (2) Technical Solutions

[0006] To achieve the above object, the present invention provides the following technical solutions: A robot motion control system based on motion capture and digital twin, including a robot motion capture module, a digital twin robot model design module, a robot data collection module, a robot data analysis module, a robot data judgment module, a digital twin robot model optimization module, and a robot execution module;

[0007] The robot motion capture module adopts optical motion capture technology, acoustic motion capture technology, and magnetic induction technology to achieve millimeter-level accurate motion data acquisition;

[0008] The digital twin robot model design module uses digital twin technology to build a virtual model of the robot's operation, inputs the data collected by the motion capture module into this model, runs the model in a virtual environment, simulates the robot's motion state according to the input data, and generates a preliminary motion control scheme;

[0009] The robot data collection module collects the data in the robot motion capture module and the digital twin robot model design module. The robot data collection module includes a motion data acquisition unit, a model operation data acquisition unit, and an environmental data acquisition unit;

[0010] The robot data analysis module calculates the data collected by the robot data collection module. The robot data analysis module includes an accuracy optimization unit, a calibration efficiency improvement unit, and an occlusion processing unit;

[0011] The robot data judgment module judges whether the current state of the robot meets the expectation and whether adjustment is needed according to the calculation result of the robot data analysis module;

[0012] The digital twin robot model optimization module supplements the defects of the digital twin robot model solution according to the result of the robot data judgment module;

[0013] The robot execution module implements the optimized instructions and executes the optimized solution in actual operation.

[0014] Preferably, the motion data acquisition unit obtains the robot motion data through optical motion capture technology, acoustic motion capture technology, and magnetic induction technology. The motion data acquisition unit numbers the actual parameters of the robot according to the characteristics of the robot motion data. The actual parameter number of the robot is U 2 .

[0015] Preferably, the model operation data acquisition unit collects model operation data by connecting to the robot operation virtual model through a network. The model operation data acquisition unit numbers the system state observation values and robot virtual model parameters actually measured by the robot through sensors at time j according to the characteristics of the model operation data. The system state observation value actually measured by the robot through sensors at time j is numbered as E j , and the robot virtual model parameter is numbered as U 1 .

[0016] Preferably, the environmental data acquisition unit collects robot operation environmental data through a vision sensor, an ultrasonic sensor, and a temperature and humidity sensor. The environmental data acquisition unit numbers the area of the occluder being occluded and the total area of the occluder according to the characteristics of the robot operation environmental data. The area of the occluder being occluded and the total area of the occluder are numbered as D z and D

[0017] Preferably, the accuracy optimization unit calculates the optimal system state value Zg according to the robot action data j , and its calculation formula is:

[0018] Zg j = Zg j / j-1 + R j * (E j - H j * Zg j / j-1 )

[0019] In the formula, Zg j / j represents the optimal system state value, that is, the optimal value of the system state at time j. Zg j / j-1 represents the optimal value of the system state at time j - 1. R j represents the Kalman gain. E j represents the system state observation value actually measured by the robot through sensors at time j. H j represents the measurement matrix

[0020] Preferably, the calibration efficiency improvement unit calculates the optimal calibration parameter Sl of the robot according to the model operation data. The calculation formula is:

[0021] Sl = w * U 1 + (1 - w) * U 2

[0022] In the formula, Sl represents the optimal calibration parameter of the robot. U 1 represents the robot virtual model parameter. U 2 represents the actual parameter of the robot. w represents the influence weight of the system to balance the virtual model parameter and the actual parameter

[0023] Preferably, the occlusion processing unit calculates the action data Bl after the robot occlusion compensation according to the robot operation environment data, and its calculation formula is:

[0024]

[0025] In the formula, Bl represents the action data after the robot occlusion compensation, and A y represents the action data predicted by the system, and D z represents the occluded area of the occluder, and D represents the total area of the occluder.

[0026] Preferably, the robot data judgment module compares the current actual operation state of the robot with the optimal value Zg of the system state j , determines whether the current state of the robot meets the expectation, and determines whether adjustment is needed if there is a deviation.

[0027] Preferably, the robot data judgment module compares the current calibration parameter of the robot with the optimal calibration parameter Sl of the robot, determines whether the current calibration state of the robot meets the expectation, and further determines whether adjustment of the calibration is needed;

[0028] The robot data judgment module analyzes the action completion degree and accuracy of the robot in the occlusion situation according to the action data Bl after the robot occlusion compensation, determines whether the current state of the robot meets the expectation, and decides whether to adjust the action strategy.

[0029] A robot motion control method based on motion capture and digital twin includes the following steps:

[0030] Step 1: Establish a robot motion capture module, a digital twin robot model design module, a robot data collection module, a robot data analysis module, a robot data judgment module, a digital twin robot model optimization module, and a robot execution module;

[0031] Step 2: In the robot motion capture module, use optical motion capture technology, acoustic motion capture technology, and magnetic induction technology to collect action data;

[0032] Step 3: In the digital twin robot model design module, use digital twin technology to build a virtual model of the robot operation, input the data collected by the motion capture module into the model, run the model in the virtual environment, simulate the motion state of the robot according to the input data, and generate a preliminary motion control scheme;

[0033] Step 4: The robot data collection module collects the data in the robot motion capture module and the digital twin robot model design module;

[0034] Step 5: The robot data analysis module calculates the data collected by the robot data collection module to obtain the optimal system state value Zg j , the optimal robot calibration parameter Sl, and the action data Bl after robot occlusion compensation;

[0035] Step 6: The robot data judgment module determines whether the current state of the robot meets the expectations and whether adjustment is needed according to the calculation results of the robot data analysis module;

[0036] Step 7: The digital twin robot model optimization module supplements the deficiencies in the solution of the digital twin robot model according to the results of the robot data judgment module;

[0037] Step 8: The robot execution module implements the optimized instructions and executes the optimized solution during actual operation.

[0038] Compared with the prior art, the present invention provides a robot motion control system and method based on motion capture and digital twin, having the following beneficial effects:

[0039] 1. By calculating the optimal system state value Zg j , establishing an error compensation model, and using the Kalman filtering algorithm to process the action data, the present invention eliminates noise and measurement errors. At the same time, the robot data judgment module compares the current actual operating state of the robot with the optimal system state value Zg j to determine whether the current state of the robot meets the expectations. If there is a deviation, it is judged whether adjustment is needed. When it is determined that the current state of the robot does not meet the expectations, the digital twin robot model optimization module will automatically input the optimal system state value Zg j into the digital twin robot model for system state value adjustment and determine the adjustment direction. The introduction and adjustment of the optimal system state value Zg j enable the digital twin robot model to more accurately simulate the motion of the robot, thereby achieving the effect of improving the accuracy of robot motion capture.

[0040] 2. By comparing the parameters of the virtual model and the actual device, the present invention quickly determines the optimal calibration parameter through an optimization algorithm. At the same time, the robot data judgment module compares the current calibration parameter of the robot with the optimal robot calibration parameter Sl to determine whether the current calibration state of the robot meets the expectations, and further judges whether calibration adjustment is needed. When it is judged that the current calibration state of the robot does not meet the expectations, the digital twin robot model optimization module will automatically input the optimal robot calibration parameter Sl into the digital twin robot model for calibration parameter supplementation, which helps to shorten the calibration and debugging time of the robot, enables the robot to quickly complete calibration during use, reduces downtime, and improves production efficiency.

[0041] 3. When the motion capture object is occluded, the present invention predicts the motion trend of the unoccluded part based on the prediction and compensation algorithm, compensates the motion data of the occluded part. At the same time, the robot data judgment module analyzes the action completion degree and accuracy of the robot in the occluded situation according to the action data Bl after the robot occlusion compensation, judges whether the current state of the robot meets the expectation, and decides whether to adjust the action strategy. When it is judged that the current state of the robot does not meet the expectation, the digital twin robot model optimization module will automatically input the action data Bl after the robot occlusion compensation into the digital twin robot model for supplementing the occlusion parameters. By supplementing the occluder parameters, the robot can better adapt to the complex occlusion environment and continuously and stably work in the presence of occlusion. Brief Description of the Drawings

[0042] Figure 1 It is a system flowchart of the present invention. Detailed Embodiment

[0043] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0044] Please refer to Figure 1 , the robot motion control system based on motion capture and digital twin includes a robot motion capture module, a digital twin robot model design module, a robot data collection module, a robot data analysis module, a robot data judgment module, a digital twin robot model optimization module, and a robot execution module;

[0045] The robot motion capture module realizes the acquisition of motion data with millimeter-level accuracy by adopting optical motion capture technology, acoustic motion capture technology, and magnetic induction technology;

[0046] The digital twin robot model design module uses digital twin technology to construct a virtual model of the robot operation, inputs the data collected by the motion capture module into the model, runs the model in the virtual environment, simulates the motion state of the robot according to the input data, and generates a preliminary motion control scheme;

[0047] The robot data collection module collects the data in the robot motion capture module and the digital twin robot model design module. The robot data collection module includes an action data acquisition unit, a model operation data acquisition unit, and an environment data acquisition unit;

[0048] The robot data analysis module calculates the data collected by the robot data collection module. The robot data analysis module includes a precision optimization unit, a calibration efficiency improvement unit, and an occlusion processing unit;

[0049] The robot data judgment module determines whether the current state of the robot meets the expectations and whether adjustment is needed based on the calculation results of the robot data analysis module;

[0050] The digital twin robot model optimization module supplements the deficiencies of the digital twin robot model solution according to the results of the robot data judgment module;

[0051] The robot execution module implements the optimized instructions and executes the optimized solution during actual operation.

[0052] The motion data acquisition unit obtains the robot motion data through optical motion capture technology, acoustic motion capture technology, and magnetic induction technology. The motion data acquisition unit numbers the actual parameters of the robot according to the characteristics of the robot motion data. The actual parameter number of the robot is U 2 。

[0053] The model operation data acquisition unit collects the model operation data by connecting to the robot operation virtual model through the network. The model operation data acquisition unit numbers the system state observation values actually measured by the sensor and the robot virtual model parameters at the j-th moment according to the characteristics of the model operation data. The system state observation value actually measured by the sensor at the j-th moment of the robot is numbered E j ,The robot virtual model parameter number is U 1 。

[0054] The environment data acquisition unit collects the robot operation environment data through visual sensors, ultrasonic sensors, and temperature and humidity sensors. The environment data acquisition unit numbers the area of the occluder being occluded and the total area of the occluder according to the characteristics of the robot operation environment data. The area of the occluder being occluded and the total area of the occluder are numbered D z 、D。

[0055] The precision optimization unit calculates the optimal system state value Zg based on the robot motion data j ,By establishing an error compensation model and using the Kalman filter algorithm to process the motion data, noise and measurement errors are eliminated, and the accuracy of motion capture is improved. Its calculation formula is:

[0056] Zg j =Zg j / j-1 +R j *(E j -H j *Zg j / j-1 )

[0057] In the formula, Zg j / j represents the optimal value of the system state, that is, the optimal value of the system state at time j, which represents the state of the robot after Kalman filtering processing. Zg j / j-1 represents the optimal value of the system state at time j - 1, R j represents the Kalman gain, which is a weight coefficient matrix that determines the proportion of the measured value E j in updating the estimated value of the system state. E j represents the observed value of the system state actually measured by the robot through sensors at time j. H j represents the measurement matrix. The role of this data is to map the system state from the state space to the observation space, that is, to relate the internal state variables of the system to the actually measurable observation variables.

[0058] The advantages are as follows: By calculating the optimal value of the system state Zg j , by establishing an error compensation model and using the Kalman filtering algorithm to process the motion data, noise and measurement errors can be eliminated. At the same time, the robot data judgment module compares the current actual operating state of the robot with the optimal value of the system state Zg j , determines whether the current state of the robot meets the expectations. If there is a deviation, it judges whether adjustment is needed. When it is determined that the current state of the robot does not meet the expectations, the digital twin robot model optimization module will automatically input the optimal value of the system state Zg j into the digital twin robot model to adjust the system state value and determine the adjustment direction. The introduction and adjustment of the optimal value of the system state Zg j enable the digital twin robot model to more accurately simulate the motion of the robot, thereby achieving the effect of improving the motion capture accuracy of the robot.

[0059] The calibration efficiency improvement unit calculates the optimal calibration parameter Sl of the robot according to the model operation data, compares the parameters of the virtual model with those of the actual device, and quickly determines the optimal calibration parameter through an optimization algorithm. The calculation formula is:

[0060] Sl = w * U 1 +(1 - w) * U 2

[0061] In the formula, Sl represents the optimal calibration parameter of the robot, U 1 represents the parameter of the robot virtual model, U 2 represents the actual parameter of the robot, and w represents the influence weight of the system to balance the virtual model parameter and the actual parameter.

[0062] The advantages are as follows: By calculating the optimal calibration parameter Sl of the robot, comparing the parameters of the virtual model with those of the actual device, and quickly determining the optimal calibration parameter through an optimization algorithm. At the same time, the robot data judgment module compares the current calibration parameter of the robot with the optimal calibration parameter Sl of the robot to determine whether the current calibration state of the robot meets the expectation, and further determines whether it is necessary to adjust the calibration. When it is determined that the current calibration state of the robot does not meet the expectation, the digital twin robot model optimization module will automatically input the optimal calibration parameter Sl of the robot into the digital twin robot model to supplement the calibration parameter, which helps to shorten the calibration and debugging time of the robot, enables the robot to quickly complete calibration during use, reduces the downtime, and improves the production efficiency.

[0063] The occlusion processing unit calculates the action data Bl of the robot after occlusion compensation based on the robot operation environment data. Based on the prediction and compensation algorithm, when the action capture object is occluded, by predicting the motion trend of the unoccluded part, the action data of the occluded part is compensated. The calculation formula is as follows:

[0064]

[0065] In the formula, Bl represents the action data of the robot after occlusion compensation, A y represents the action data predicted by the system, D z represents the area of the occluder that is occluded, and D represents the total area of the occluder.

[0066] The advantages are as follows: By calculating the action data Bl of the robot after occlusion compensation, based on the prediction and compensation algorithm, when the action capture object is occluded, by predicting the motion trend of the unoccluded part, the action data of the occluded part is compensated. At the same time, the robot data judgment module analyzes the action completion degree and accuracy of the robot in the occluded situation based on the action data Bl of the robot after occlusion compensation, determines whether the current state of the robot meets the expectation, and decides whether it is necessary to adjust the action strategy. When it is determined that the current state of the robot does not meet the expectation, the digital twin robot model optimization module will automatically input the action data Bl of the robot after occlusion compensation into the digital twin robot model to supplement the occlusion parameter. By supplementing the occluder parameter, the robot can better adapt to the complex occlusion environment and continuously work stably in the presence of occlusion.

[0067] A robot motion control method based on action capture and digital twin includes the following steps:

[0068] Step 1: Establish a robot motion capture module, a digital twin robot model design module, a robot data collection module, a robot data analysis module, a robot data judgment module, a digital twin robot model optimization module, and a robot execution module. Different modules can be replaced or improved according to actual needs, enabling quick adaptation to different application scenarios and task requirements, laying a solid foundation for building an efficient and stable robot motion control system;

[0069] Step 2: In the robot motion capture module, adopt optical motion capture technology, acoustic motion capture technology, and magnetic induction technology to collect motion data. The above capture technologies combine three different motion capture technologies, which can give full play to the advantages of each technology;

[0070] Step 3: In the digital twin robot model design module, use digital twin technology to build a virtual model of the robot's operation, and input the data collected by the motion capture module into this model. Run the model in a virtual environment, simulate the robot's motion state according to the input data, and generate a preliminary motion control plan. This model can predict the robot's motion state under different working conditions in advance, thus avoiding possible errors and risks in actual operation. Moreover, the generated preliminary motion control plan can serve as a basis for subsequent adjustments, effectively shortening the R & D cycle and cost;

[0071] Step 4: The robot data collection module collects the data in the robot motion capture module and the digital twin robot model design module, providing a comprehensive information source for subsequent data calculations;

[0072] Step 5: The robot data analysis module calculates the data collected by the robot data collection module to obtain the optimal system state value Zg j , the optimal calibration parameter Sl of the robot, and the motion data Bl of the robot after occlusion compensation. The calculation of the above values helps to improve the motion accuracy, stability, and adaptability of the robot, enabling it to better complete various tasks;

[0073] Step 6: The robot data judgment module judges whether the current state of the robot meets the expectations and whether adjustment is needed according to the calculation results of the robot data analysis module. By comparing the calculation results with the current parameters of the robot's motion, it can quickly discover the problems and deviations existing in the robot's operation process and make a decision on whether adjustment is needed in a timely manner. This real-time monitoring and feedback mechanism can ensure that the robot is always in the best operating state, thereby improving the stability of the system;

[0074] Step Seven: The digital twin robot model optimization module supplements the deficiencies of the digital twin robot model solution based on the results of the robot data judgment module. By supplementing the deficiencies and shortcomings in the model solution, the virtual model can more accurately reflect the actual operation of the robot, providing more precise guidance for the motion control of the robot;

[0075] Step Eight: The robot execution module implements the optimized instructions and executes the optimized solution during actual operation. By executing the optimized solution, the robot can give full play to its performance advantages, improve work efficiency and quality, and reduce the occurrence of errors and failures. At the same time, the feedback information during actual operation can also provide a reference for subsequent optimization, forming a closed-loop optimization process to continuously improve the motion control level of the robot.

[0076] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Robot motion control system based on motion capture and digital twin, characterized by: It includes robot motion capture module, digital twin robot model design module, robot data collection module, robot data analysis module, robot data judgment module, digital twin robot model optimization module and robot execution module; The robot motion capture module uses optical motion capture technology, acoustic motion capture technology and magnetic induction technology to achieve millimeter-level precision motion data collection; The digital twin robot model design module uses digital twin technology to build a robot operation virtual model, inputs the data collected by the motion capture module into the model, runs the model in a virtual environment, simulates the robot's motion state according to the input data, and generates a preliminary motion control plan; The robot data collection module collects data from the robot motion capture module and the digital twin robot model design module, and the robot data collection module includes a motion data collection unit, a model operation data collection unit and an environment data collection unit; The robot data analysis module calculates the data collected by the robot data collection module, and the robot data analysis module includes an accuracy optimization unit, a calibration efficiency improvement unit and an occlusion processing unit; The robot data judgment module judges whether the current state of the robot meets expectations and whether adjustment is required based on the calculation results of the robot data analysis module; The digital twin robot model optimization module supplements the shortcomings of the digital twin robot model solution according to the results of the robot data judgment module; The robot execution module implements the optimized instructions and executes the optimized solution in actual operation.

2. The robot motion control system based on motion capture and digital twinning according to claim 1, characterized in that: The motion data acquisition unit acquires the robot motion data through optical motion capture technology, acoustic motion capture technology and magnetic induction technology. The motion data acquisition unit numbers the actual parameters of the robot according to the characteristics of the robot motion data. The actual parameters of the robot are numbered U2.

3. The robot motion control system based on motion capture and digital twinning according to claim 1, characterized in that: The model operation data acquisition unit collects model operation data by connecting to the robot operation virtual model through the network. The model operation data acquisition unit performs data numbering on the system state observation value and the robot virtual model parameter actually measured by the sensor at time j according to the model operation data characteristics. The system state observation value actually measured by the sensor at time j is numbered as E. j , the robot virtual model parameter number is U1.

4. The robot motion control system based on motion capture and digital twinning according to claim 1, characterized in that: The environmental data acquisition unit collects robot operating environment data through visual sensors, ultrasonic sensors, and temperature and humidity sensors. The environmental data acquisition unit numbers the area blocked by the obstruction and the total area of ​​the obstruction according to the characteristics of the robot operating environment data. The area blocked by the obstruction and the total area of ​​the obstruction are numbered D and D respectively. z 、D.

5. The robot motion control system based on motion capture and digital twinning according to claim 2, characterized in that: The precision optimization unit calculates the optimal value Zg of the system state according to the robot motion data j , and its calculation formula is: Zg j =Zg j / j-1 +R j *(E j -H j *Zg j / j-1 ) In the formula, Zg j / j represents the optimal value of the system state, that is, the optimal value of the system state at time j, Zg j / j-1 represents the optimal value of the system state at time j-1, R j represents the Kalman gain, E j H represents the system state observation value actually measured by the sensor at time j, j represents the measurement matrix.

6. The robot motion control system based on motion capture and digital twinning according to claim 3 is characterized in that: The calibration efficiency improvement unit calculates the robot's optimal calibration parameter S1 according to the model operation data, and the calculation formula is: Sl=w*U1+(1-w)*U2 In the formula, Sl represents the optimal calibration parameters of the robot, U1 represents the virtual model parameters of the robot, U2 represents the actual parameters of the robot, and w represents the influence weight of the virtual model parameters and the actual parameters of the system balance.

7. The robot motion control system based on motion capture and digital twinning according to claim 4, characterized in that: The occlusion processing unit calculates the action data Bl of the robot after occlusion compensation according to the robot operating environment data, and the calculation formula is: In the formula, Bl represents the action data of the robot after occlusion compensation, A y represents the action data predicted by the system, D z represents the area blocked by the occluder, and D represents the total area of ​​the occluder.

8. The robot motion control system based on motion capture and digital twinning according to claim 5, characterized in that: The robot data judgment module is based on the optimal value Zg of the system state. j , compare the current actual operating status of the robot with it to determine whether the current status of the robot meets expectations. If there is a deviation, determine whether adjustment is needed.

9. The robot motion control system based on motion capture and digital twinning according to claim 7, characterized in that: The robot data judgment module compares the current robot calibration parameters with the robot's optimal calibration parameters S1 to determine whether the robot's current calibration state meets expectations, and then determines whether the calibration needs to be adjusted; The robot data judgment module analyzes the robot's action completion and accuracy under occlusion according to the robot's action data Bl after occlusion compensation, determines whether the robot's current state meets expectations, and determines whether the action strategy needs to be adjusted.

10. A robot motion control method based on motion capture and digital twin, characterized in that: The following steps are involved: Step 1: Establish a robot motion capture module, a digital twin robot model design module, a robot data collection module, a robot data analysis module, a robot data judgment module, a digital twin robot model optimization module and a robot execution module; Step 2: Using optical motion capture technology, acoustic motion capture technology and magnetic induction technology in the robot motion capture module to collect motion data; Step 3: Use digital twin technology to build a virtual model of the robot in the digital twin robot model design module, input the data collected by the motion capture module into the model, run the model in a virtual environment, simulate the robot's motion state according to the input data, and generate a preliminary motion control plan; Step 4: The robot data collection module collects data from the robot motion capture module and the digital twin robot model design module; Step 5: The robot data analysis module calculates the data collected by the robot data collection module to obtain the optimal value Zg of the system state j , the robot's optimal calibration parameters S1 and the robot's motion data Bl after occlusion compensation; Step 6: The robot data judgment module determines whether the current state of the robot meets expectations and whether it needs to be adjusted based on the calculation results of the robot data analysis module; Step 7: The digital twin robot model optimization module supplements the shortcomings of the digital twin robot model solution based on the results of the robot data judgment module; Step 8. The robot execution module implements the optimized instructions and executes the optimized plan in actual operation.

Images