Rehabilitation robot data acquisition method based on intention perception

Abstract

The invention relates to a rehabilitation robot data acquisition method based on intention perception, and belongs to the technical field of rehabilitation robots. Rehabilitation exercise comfort rehearsal and analysis are carried out based on exercise parameters of a rehabilitation robot and a patient specific biomechanical intention model, and the rehabilitation exercise comfort of a patient is obtained; and finally, analyzing and configuring the motion parameters of the rehabilitation robot based on the rehabilitation motion comfort of the patient, and performing control optimization on the rehabilitation robot according to the final motion parameters of the rehabilitation robot. Noise data collected by the sensor are processed through the Karl filtering algorithm, so that the recognition precision of user intentions can be improved, then motion simulation analysis can be conducted on processed real-time muscle force signal characteristics of a patient through the virtual reality technology and the digital twinning technology, control over the rehabilitation robot can be optimized, and the rehabilitation robot can be more accurately recognized. The exercise comfort of the user in the rehabilitation process is guaranteed, and it is guaranteed that the exercise amount of the patient is within a preset range.

Description

Technical Field

[0001] The present invention relates to the field of rehabilitation robot data technology, and in particular to a rehabilitation robot data acquisition method based on intention perception. Background Art

[0002] Rehabilitation training plays a great role in the early rehabilitation of hemiplegic stroke patients and seriously injured patients. In the past, walking ability training and standing balance training were mostly used. Conventional rehabilitation training can improve the patient's limb motor function and balance ability to a certain extent by training the patient's gait and balance function, but the effect is not obvious, and the lack of necessary feedback exercises can easily form a bad gait. The rehabilitation robot is an intelligent lower limb feedback rehabilitation training system that can combine movement patterns and visual feedback to trigger the lower limb motor function, so that the patient can regain the walking function, and has a high application value. However, rehabilitation robots often need to collect data through sensors, and the data collected by the sensors often have certain noise, which is not conducive to perceiving the user's action intention, and thus cannot coordinate and assist the operation with the rehabilitation robot better; secondly, the existing technology does not consider the user's comfort when doing rehabilitation exercises, resulting in poor experience during rehabilitation exercises. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a rehabilitation robot data collection method based on intention perception.

[0004] To achieve the above object, the technical solution adopted by the present invention is: The first aspect of the present invention provides a rehabilitation robot data collection method based on intention perception, characterized in that it includes the following steps: By arranging a flexible piezoresistive sensor on the joint of the patient, collecting the patient's real-time muscle force signal characteristic data through the flexible piezoresistive sensor, and processing the patient's real-time muscle force signal characteristic data through the Kalman filter method; Acquiring processed real-time muscle force signal characteristic data of the patient, and constructing a patient-specific biomechanical intention model according to the processed real-time muscle force signal characteristic data of the patient; Initializing the motion parameters of the rehabilitation robot, performing a comfort preview and analysis of the rehabilitation movement based on the motion parameters of the rehabilitation robot and the patient-specific biomechanical intention model, and obtaining the patient's rehabilitation movement comfort; The motion parameters of the rehabilitation robot are analyzed and configured based on the rehabilitation motion comfort of the patient, the final motion parameters of the rehabilitation robot are obtained, and the rehabilitation robot is controlled and optimized according to the final motion parameters of the rehabilitation robot.

[0005] Further, in the present method, a flexible piezoresistive sensor is arranged on the joint of the patient, and the real-time muscle force signal characteristic data of the patient is collected by the flexible piezoresistive sensor, and the real-time muscle force signal characteristic data of the patient is processed by the Kalman filtering method, specifically: A flexible piezoresistive sensor is arranged on the joint of the patient to form a flexible piezoresistive sensor sequence, and the real-time muscle force signal characteristic data of the patient is collected through the flexible piezoresistive sensor to determine the effective muscle force signal characteristic data range; Setting a system state vector based on the effective muscle force signal characteristic data range by the Kalman filtering method, setting the patient's real-time muscle force signal characteristic data as an observation vector, and setting a system model based on the system state vector; An observation model is set based on the observation vector, a state estimate and an error covariance matrix at a current moment are predicted based on the system model and the observation model, and a Kalman gain is calculated based on the state estimate and the error covariance matrix at the current moment; Based on the Kalman gain update state estimation and covariance matrix, adaptive filtering is performed in combination with the weights updated by the Kalman filter to complete the denoising of the patient's real-time muscle force signal feature data.

[0006] Furthermore, in the present method, the processed real-time muscle force signal characteristic data of the patient is obtained, and a patient-specific biomechanical intention model is constructed according to the processed real-time muscle force signal characteristic data of the patient, specifically: Acquire the processed real-time muscle force signal characteristic data of the patient, construct a biological joint model of the patient through three-dimensional modeling software, and preview the processed real-time muscle force signal characteristic data of the patient in the biological joint model of the patient; Based on the processed real-time muscle force signal characteristic data of the patient, the finite element analysis method is used to preview the force distribution of the tissue, obtain a schematic diagram of the force distribution of the tissue, and construct a patient-specific biomechanical intention model based on the schematic diagram of the force distribution of the tissue.

[0007] Furthermore, in the present method, the motion parameters of the rehabilitation robot are initialized, and the rehabilitation motion comfort preview and analysis are performed based on the motion parameters of the rehabilitation robot and the patient-specific biomechanical intention model to obtain the patient's rehabilitation motion comfort, specifically: Setting an exercise time threshold, obtaining the patient's tissue force threshold range in each rehabilitation state through big data, and setting comfort evaluation index characteristics according to the patient's tissue force threshold range and exercise time threshold in each rehabilitation state; Initializing and configuring the motion parameters of the rehabilitation robot including motion speed, motion torque information, and motion angle information according to the patient-specific biomechanical intention model; Perform motion preview according to the motion parameters of the rehabilitation robot, including motion speed, motion torque information, and motion angle information; By performing exercise preview within the exercise time threshold, the comfort level of the patient-specific biomechanical intention model is evaluated based on the comfort level evaluation index characteristics to obtain the patient's rehabilitation exercise comfort level.

[0008] Furthermore, in this method, a motion time threshold is set, specifically: The best exercise time for patients in each rehabilitation state is obtained through big data, and a knowledge graph is constructed. A graph neural network is introduced to input the best exercise time for patients in each rehabilitation state into the graph neural network for processing; Taking the rehabilitation status and the optimal exercise time as graph nodes of the graph neural network, constructing a directed description relationship, connecting the graph nodes based on the directed description relationship, constructing a topological structure graph, and inputting the topological structure graph into the knowledge graph for node representation; Obtain the rehabilitation status information of the current patient, input the rehabilitation status information of the current patient into the knowledge graph for analysis, obtain the best exercise time for the patient during the current rehabilitation state, and set the exercise time threshold according to the best exercise time for the patient during the current rehabilitation state.

[0009] Furthermore, in the present method, the motion parameters of the rehabilitation robot are analyzed and configured based on the rehabilitation motion comfort of the patient to obtain the final motion parameters of the rehabilitation robot, specifically: Introducing a particle swarm algorithm, setting the number of iterations based on the particle swarm algorithm, and judging whether the patient's rehabilitation exercise comfort is not lower than a preset rehabilitation exercise comfort evaluation index; When the patient's rehabilitation exercise comfort level is not lower than a preset rehabilitation exercise comfort level evaluation index, exercising according to the current exercise parameters of the rehabilitation robot; When the patient's rehabilitation exercise comfort is lower than a preset rehabilitation exercise comfort evaluation index, iteration is performed based on the number of iterations to reset the motion parameters of the rehabilitation robot until the patient's rehabilitation exercise comfort is no lower than the preset rehabilitation exercise comfort evaluation index.

[0010] The second aspect of the present invention provides a rehabilitation robot data acquisition system based on intention perception, comprising a memory and a processor, wherein the memory includes a rehabilitation robot data acquisition method program based on intention perception, and when the rehabilitation robot data acquisition method program based on intention perception is executed by the processor, any step of the rehabilitation robot data acquisition method based on intention perception is implemented.

[0011] The third aspect of the present invention provides a computer-readable storage medium, including a rehabilitation robot data collection method program based on intention perception. When the rehabilitation robot data collection method program based on intention perception is executed by a processor, the steps of any one of the rehabilitation robot data collection methods based on intention perception are implemented.

[0012] The present invention solves the defects existing in the background technology and has the following beneficial effects: The present invention sets a flexible piezoresistive sensor on the joint of the patient, collects the patient's real-time muscle force signal characteristic data through the flexible piezoresistive sensor, processes the patient's real-time muscle force signal characteristic data through the Kalman filter method, and then obtains the processed patient's real-time muscle force signal characteristic data, and constructs a patient-specific biomechanical intention model according to the processed patient's real-time muscle force signal characteristic data, thereby initializing the motion parameters of the rehabilitation robot, performing a comfortable preview and analysis of rehabilitation movement based on the motion parameters of the rehabilitation robot and the patient-specific biomechanical intention model, obtaining the patient's rehabilitation movement comfort, and finally analyzing and configuring the rehabilitation robot's motion parameters based on the patient's rehabilitation movement comfort, obtaining the final rehabilitation robot's motion parameters, and controlling and optimizing the rehabilitation robot according to the final rehabilitation robot's motion parameters. The present invention processes the noise data collected by the sensor through the Kalman filter algorithm, thereby improving the recognition accuracy of the user's intention, and secondly, can perform motion simulation analysis on the processed patient's real-time muscle force signal characteristics through virtual reality technology and digital twin technology, can optimize the control of the rehabilitation robot, ensure the user's movement comfort during the rehabilitation process, and ensure that the patient's exercise volume is within a predetermined range. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0014] Figure 1 The overall flow chart of the rehabilitation robot data acquisition method based on intention perception is shown; Figure 2 The system block diagram of the rehabilitation robot data acquisition system based on intention perception is shown. DETAILED DESCRIPTION

[0015] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0016] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0017] like Figure 1 As shown, the first aspect of the present invention provides a rehabilitation robot data collection method based on intention perception, characterized in that it includes the following steps: S102: a flexible piezoresistive sensor is arranged on the joint of the patient, the real-time muscle force signal characteristic data of the patient is collected by the flexible piezoresistive sensor, and the real-time muscle force signal characteristic data of the patient is processed by a Kalman filter method; S104: Acquire the processed real-time muscle force signal characteristic data of the patient, and construct a patient-specific biomechanical intention model according to the processed real-time muscle force signal characteristic data of the patient; S106: Initializing the motion parameters of the rehabilitation robot, performing a comfort preview and analysis of rehabilitation motion based on the motion parameters of the rehabilitation robot and a patient-specific biomechanical intention model, and obtaining the patient's rehabilitation motion comfort; S108: Analyze and configure the motion parameters of the rehabilitation robot based on the patient's rehabilitation movement comfort, obtain the final motion parameters of the rehabilitation robot, and control and optimize the rehabilitation robot according to the final motion parameters of the rehabilitation robot.

[0018] It should be noted that the present invention processes the noise data collected by the sensor through the Karl filter algorithm, so as to improve the recognition accuracy of the user's intention. Secondly, it can perform motion simulation analysis on the real-time muscle force signal characteristics of the processed patient through virtual reality technology and digital twin technology, optimize the control of the rehabilitation robot, ensure the user's exercise comfort during the rehabilitation process, and ensure that the patient's exercise volume is within the predetermined range.

[0019] Furthermore, in this method, a flexible piezoresistive sensor is arranged on the joint of the patient, and the real-time muscle force signal characteristic data of the patient is collected by the flexible piezoresistive sensor, and the real-time muscle force signal characteristic data of the patient is processed by the Kalman filter method, specifically: By setting flexible piezoresistive sensors on the joints of the patient to form a flexible piezoresistive sensor sequence, the patient's real-time muscle force signal characteristic data is collected through the flexible piezoresistive sensors to determine the effective muscle force signal characteristic data range; The system state vector is set based on the effective muscle force signal characteristic data range by using the Kalman filter method, the patient's real-time muscle force signal characteristic data is set as the observation vector, and the system model is set based on the system state vector; An observation model is set based on the observation vector, a state estimate and an error covariance matrix at the current moment are predicted based on the system model and the observation model, and a Kalman gain is calculated based on the state estimate and the error covariance matrix at the current moment; Based on the Kalman gain update state estimation and covariance matrix, adaptive filtering is performed in combination with the weights updated by the Kalman filter to complete the denoising of the patient's real-time muscle force signal feature data.

[0020] It should be noted that the Kalman filtering method can be used to denoise the patient's real-time muscle force signal feature data, thereby improving the accuracy of identifying the patient's movement intention.

[0021] Furthermore, in the present method, the processed real-time muscle force signal characteristic data of the patient is obtained, and a patient-specific biomechanical intention model is constructed according to the processed real-time muscle force signal characteristic data of the patient, specifically: Obtain the processed real-time muscle force signal characteristic data of the patient, build a biological joint model of the patient through three-dimensional modeling software (including virtual reality technology, digital twin technology, etc.), and preview the processed real-time muscle force signal characteristic data of the patient in the biological joint model of the patient; Based on the processed real-time muscle force signal characteristic data of the patient, the finite element analysis method is used to preview the force distribution of the tissue, obtain the schematic diagram of the tissue force distribution, and build a patient-specific biomechanical intention model based on the schematic diagram of the tissue force distribution.

[0022] It should be noted that the finite element analysis method is used to preview the force distribution of the patient's joint tissue, so as to construct a patient-specific biomechanical intention model. The patient-specific biomechanical intention model can display a schematic diagram of the force distribution of each tissue of the patient, thereby enabling dynamic display.

[0023] Furthermore, in this method, the motion parameters of the rehabilitation robot are initialized, and the rehabilitation motion comfort preview and analysis are performed based on the motion parameters of the rehabilitation robot and the patient-specific biomechanical intention model to obtain the patient's rehabilitation motion comfort, specifically: Set the exercise time threshold, obtain the patient's tissue force threshold range in each rehabilitation state through big data, and set the comfort evaluation index characteristics according to the patient's tissue force threshold range and exercise time threshold in each rehabilitation state; Initialize and configure the motion parameters of the rehabilitation robot including motion speed, motion torque information, and motion angle information according to the patient-specific biomechanical intention model; Perform motion preview according to the motion parameters of the rehabilitation robot, including motion speed, motion torque information, and motion angle information; By rehearsing the exercise within the exercise time threshold, the comfort level of the patient-specific biomechanical intention model is evaluated based on the characteristics of the comfort evaluation index to obtain the patient's rehabilitation exercise comfort.

[0024] It should be noted that the rehabilitation state includes the initial rehabilitation state, semi-recovery rehabilitation state, highly recovered rehabilitation state, full recovery rehabilitation state, etc. This method is used to obtain the patient's rehabilitation movement comfort, so as to better determine the impact of the current rehabilitation robot's movement parameters such as movement speed, movement torque information, and movement angle information on the user's movement comfort.

[0025] Furthermore, in this method, a motion time threshold is set, specifically: The best exercise time for patients in various rehabilitation states is obtained through big data, and a knowledge graph is constructed. The graph neural network is introduced to input the best exercise time for patients in various rehabilitation states into the graph neural network for processing; The rehabilitation status and the optimal exercise time are taken as graph nodes of the graph neural network, a directed description relationship is constructed, the graph nodes are connected based on the directed description relationship, a topological structure graph is constructed, and the topological structure graph is input into the knowledge graph for node representation; Obtain the current patient's rehabilitation status information, input the current patient's rehabilitation status information into the knowledge graph for analysis, obtain the patient's best exercise time during the current rehabilitation state, and set the exercise time threshold according to the patient's best exercise time during the current rehabilitation state.

[0026] It should be noted that patients in different rehabilitation states have different optimal exercise times. This method can further optimize the exercise time threshold and improve the rationality of comfort evaluation.

[0027] Furthermore, in this method, the motion parameters of the rehabilitation robot are analyzed and configured based on the patient's rehabilitation motion comfort, and the final motion parameters of the rehabilitation robot are obtained, specifically: The particle swarm algorithm is introduced, and the number of iterations is set based on the particle swarm algorithm to determine whether the patient's rehabilitation exercise comfort is not lower than the preset rehabilitation exercise comfort evaluation index; When the patient's rehabilitation exercise comfort level is not lower than the preset rehabilitation exercise comfort level evaluation index, the rehabilitation robot exercises according to the current exercise parameters; When the patient's rehabilitation movement comfort is lower than the preset rehabilitation movement comfort evaluation index, iterations are performed based on the number of iterations, and the movement parameters of the rehabilitation robot are reset until the patient's rehabilitation movement comfort is no lower than the preset rehabilitation movement comfort evaluation index.

[0028] It should be noted that the particle swarm optimization (PSO), also known as the particle swarm algorithm, comes from the simulation of a simplified social model. The particle swarm optimization algorithm originated from the process of bird flocks foraging. A core mechanism is that each bird forages for food and remembers a position closest to the food. By communicating with other birds, it obtains the best known position of the entire flock and guides the flock to continue searching in this direction. There are also two key settings: the historical best position of the particle (pBest vector) and the historical best position of the group (gBest vector). Here, the pBest vector is a set of vectors that contains the historical best position of each particle, and the gBest vector is the vector with the highest fitness value in the pBest vector, that is, the global optimum. Note: In the algorithm, the objective function to be optimized is generally taken as the fitness function, the size of the fitness value is evaluated, and then the pBest vector and gBest vector are updated. This method can optimize the motion parameters of the rehabilitation robot, ensure the comfort of the user's rehabilitation exercise, and avoid excessive exercise.

[0029] In addition, the method further comprises: Introducing a Bayesian network to obtain characteristic data of changes in the patient's rehabilitation exercise comfort level within a preset time, and inputting the characteristic data of changes in the patient's rehabilitation exercise comfort level within the preset time into the Bayesian network; Using the rehabilitation exercise comfort characteristic data of each time stamp in the rehabilitation exercise comfort change characteristic data of the patient within the preset time as an observation vector in the Bayesian network, and calculating a transfer probability value of the observation vector transferring to another observation vector; Determine whether the transition probability value is greater than a preset transition probability value, when the transition probability value is not greater than the preset transition probability value, maintain the current observation vector unchanged, when the transition probability value is greater than the preset transition probability value, update to another observation vector; An updated observation vector is obtained, and based on the updated observation vector, the rehabilitation exercise comfort change characteristic data of the patient within a preset time is updated.

[0030] It should be noted that by updating the patient's rehabilitation movement comfort state through the Bayesian network, the user's comfort level can be updated in a timely manner so that the rehabilitation robot can be controlled more in line with the user's movement intention.

[0031] like Figure 2 As shown, the second aspect of the present invention provides a rehabilitation robot data acquisition system 4 based on intention perception, including a memory 41 and a processor 42. The memory 41 includes a rehabilitation robot data acquisition method program based on intention perception. When the rehabilitation robot data acquisition method program based on intention perception is executed by the processor 42, the following steps are implemented: By setting a flexible piezoresistive sensor on the patient's joint, the patient's real-time muscle force signal characteristic data is collected through the flexible piezoresistive sensor, and the patient's real-time muscle force signal characteristic data is processed through the Kalman filter method; Acquire the processed real-time muscle force signal characteristic data of the patient, and construct a patient-specific biomechanical intention model according to the processed real-time muscle force signal characteristic data of the patient; Initialize the motion parameters of the rehabilitation robot, perform comfort preview and analysis of rehabilitation movements based on the motion parameters of the rehabilitation robot and the patient-specific biomechanical intention model, and obtain the patient's rehabilitation movement comfort level; The motion parameters of the rehabilitation robot are analyzed and configured based on the patient's rehabilitation movement comfort, the final motion parameters of the rehabilitation robot are obtained, and the rehabilitation robot is controlled and optimized according to the final motion parameters of the rehabilitation robot.

[0032] It should be noted that the present invention processes the noise data collected by the sensor through the Karl filter algorithm, so as to improve the recognition accuracy of the user's intention. Secondly, it can perform motion simulation analysis on the real-time muscle force signal characteristics of the processed patient through virtual reality technology and digital twin technology, optimize the control of the rehabilitation robot, ensure the user's exercise comfort during the rehabilitation process, and ensure that the patient's exercise volume is within the predetermined range.

[0033] Furthermore, in this system, a flexible piezoresistive sensor is arranged on the joint of the patient, and the real-time muscle force signal characteristic data of the patient is collected by the flexible piezoresistive sensor, and the real-time muscle force signal characteristic data of the patient is processed by the Kalman filter method, specifically: By setting flexible piezoresistive sensors on the joints of the patient to form a flexible piezoresistive sensor sequence, the patient's real-time muscle force signal characteristic data is collected through the flexible piezoresistive sensors to determine the effective muscle force signal characteristic data range; The system state vector is set based on the effective muscle force signal characteristic data range by using the Kalman filter method, the patient's real-time muscle force signal characteristic data is set as the observation vector, and the system model is set based on the system state vector; An observation model is set based on the observation vector, a state estimate and an error covariance matrix at the current moment are predicted based on the system model and the observation model, and a Kalman gain is calculated based on the state estimate and the error covariance matrix at the current moment; Based on the Kalman gain update state estimation and covariance matrix, adaptive filtering is performed in combination with the weights updated by the Kalman filter to complete the denoising of the patient's real-time muscle force signal feature data.

[0034] It should be noted that the Kalman filtering method can be used to denoise the patient's real-time muscle force signal feature data, thereby improving the accuracy of identifying the patient's movement intention.

[0035] Furthermore, in this system, the processed real-time muscle force signal characteristic data of the patient is obtained, and a patient-specific biomechanical intention model is constructed according to the processed real-time muscle force signal characteristic data of the patient, specifically: Obtain the processed real-time muscle force signal characteristic data of the patient, build a biological joint model of the patient through three-dimensional modeling software (including virtual reality technology, digital twin technology, etc.), and preview the processed real-time muscle force signal characteristic data of the patient in the biological joint model of the patient; Based on the processed real-time muscle force signal characteristic data of the patient, the finite element analysis method is used to preview the force distribution of the tissue, obtain the schematic diagram of the tissue force distribution, and build a patient-specific biomechanical intention model based on the schematic diagram of the tissue force distribution.

[0036] It should be noted that the finite element analysis method is used to preview the force distribution of the patient's joint tissue, so as to construct a patient-specific biomechanical intention model. The patient-specific biomechanical intention model can display a schematic diagram of the force distribution of each tissue of the patient, thereby enabling dynamic display.

[0037] Furthermore, in this system, the motion parameters of the rehabilitation robot are initialized, and the rehabilitation motion comfort preview and analysis are performed based on the motion parameters of the rehabilitation robot and the patient-specific biomechanical intention model to obtain the patient's rehabilitation motion comfort, specifically: Set the exercise time threshold, obtain the patient's tissue force threshold range in each rehabilitation state through big data, and set the comfort evaluation index characteristics according to the patient's tissue force threshold range and exercise time threshold in each rehabilitation state; Initialize and configure the motion parameters of the rehabilitation robot including motion speed, motion torque information, and motion angle information according to the patient-specific biomechanical intention model; Perform motion preview according to the motion parameters of the rehabilitation robot, including motion speed, motion torque information, and motion angle information; By rehearsing the exercise within the exercise time threshold, the comfort level of the patient-specific biomechanical intention model is evaluated based on the characteristics of the comfort evaluation index to obtain the patient's rehabilitation exercise comfort.

[0038] It should be noted that the rehabilitation status includes the initial rehabilitation status, semi-recovery rehabilitation status, highly recovered rehabilitation status, full recovery rehabilitation status, etc. This system is used to obtain the patient's rehabilitation movement comfort, so as to better determine the impact of the current rehabilitation robot's movement parameters such as movement speed, movement torque information, and movement angle information on the user's movement comfort.

[0039] Furthermore, in this system, a motion time threshold is set, specifically: The best exercise time for patients in various rehabilitation states is obtained through big data, and a knowledge graph is constructed. The graph neural network is introduced to input the best exercise time for patients in various rehabilitation states into the graph neural network for processing; The rehabilitation status and the optimal exercise time are taken as graph nodes of the graph neural network, a directed description relationship is constructed, the graph nodes are connected based on the directed description relationship, a topological structure graph is constructed, and the topological structure graph is input into the knowledge graph for node representation; Obtain the current patient's rehabilitation status information, input the current patient's rehabilitation status information into the knowledge graph for analysis, obtain the patient's best exercise time during the current rehabilitation state, and set the exercise time threshold according to the patient's best exercise time during the current rehabilitation state.

[0040] It should be noted that patients in different rehabilitation states have different optimal exercise times. This system can further optimize the exercise time threshold and improve the rationality of comfort evaluation.

[0041] Furthermore, in this system, the motion parameters of the rehabilitation robot are analyzed and configured based on the patient's rehabilitation motion comfort, and the final motion parameters of the rehabilitation robot are obtained, specifically: The particle swarm algorithm is introduced, and the number of iterations is set based on the particle swarm algorithm to determine whether the patient's rehabilitation exercise comfort is not lower than the preset rehabilitation exercise comfort evaluation index; When the patient's rehabilitation exercise comfort level is not lower than the preset rehabilitation exercise comfort level evaluation index, the rehabilitation robot exercises according to the current exercise parameters; When the patient's rehabilitation movement comfort is lower than the preset rehabilitation movement comfort evaluation index, iterations are performed based on the number of iterations, and the movement parameters of the rehabilitation robot are reset until the patient's rehabilitation movement comfort is no lower than the preset rehabilitation movement comfort evaluation index.

[0042] It should be noted that the particle swarm optimization (PSO), also known as the particle swarm algorithm, comes from the simulation of a simplified social model. The particle swarm optimization algorithm originated from the process of bird flocks foraging. A core mechanism is that each bird forages for food and remembers a position closest to the food. By communicating with other birds, it obtains the best known position of the entire flock and guides the flock to continue searching in this direction. There are also two key settings: the historical best position of the particle (pBest vector) and the historical best position of the group (gBest vector). Here, the pBest vector is a set of vectors that contains the historical best position of each particle, and the gBest vector is the vector with the highest fitness value in the pBest vector, that is, the global optimum. Note: In the algorithm, the objective function to be optimized is generally taken as the fitness function, the size of the fitness value is evaluated, and then the pBest vector and gBest vector are updated. This system can optimize the motion parameters of the rehabilitation robot, ensure the comfort of the user's rehabilitation exercise, and avoid excessive exercise.

[0043] In addition, the system also includes: Introducing a Bayesian network to obtain characteristic data of changes in the patient's rehabilitation exercise comfort level within a preset time, and inputting the characteristic data of changes in the patient's rehabilitation exercise comfort level within the preset time into the Bayesian network; Using the rehabilitation exercise comfort characteristic data of each time stamp in the rehabilitation exercise comfort change characteristic data of the patient within the preset time as an observation vector in the Bayesian network, and calculating a transfer probability value of the observation vector transferring to another observation vector; Determine whether the transition probability value is greater than a preset transition probability value, when the transition probability value is not greater than the preset transition probability value, maintain the current observation vector unchanged, when the transition probability value is greater than the preset transition probability value, update to another observation vector; An updated observation vector is obtained, and based on the updated observation vector, the rehabilitation exercise comfort change characteristic data of the patient within a preset time is updated.

[0044] It should be noted that by updating the patient's rehabilitation movement comfort state through the Bayesian network, the user's comfort level can be updated in a timely manner so that the rehabilitation robot can be controlled more in line with the user's movement intention.

[0045] The third aspect of the present invention provides a computer-readable storage medium, including a rehabilitation robot data collection method program based on intention perception. When the rehabilitation robot data collection method program based on intention perception is executed by a processor, any step of the rehabilitation robot data collection method based on intention perception is implemented.

[0046] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0047] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0048] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0049] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0050] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0051] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A rehabilitation robot data collection method based on intention perception, characterized in that: The following steps are involved: By arranging a flexible piezoresistive sensor on the joint of the patient, collecting the patient's real-time muscle force signal characteristic data through the flexible piezoresistive sensor, and processing the patient's real-time muscle force signal characteristic data through the Kalman filter method; Acquiring processed real-time muscle force signal characteristic data of the patient, and constructing a patient-specific biomechanical intention model according to the processed real-time muscle force signal characteristic data of the patient; Initializing the motion parameters of the rehabilitation robot, performing a comfort preview and analysis of the rehabilitation movement based on the motion parameters of the rehabilitation robot and the patient-specific biomechanical intention model, and obtaining the patient's rehabilitation movement comfort; The motion parameters of the rehabilitation robot are analyzed and configured based on the rehabilitation motion comfort of the patient, the final motion parameters of the rehabilitation robot are obtained, and the rehabilitation robot is controlled and optimized according to the final motion parameters of the rehabilitation robot.

2. The method for data collection of a rehabilitation robot based on intention perception according to claim 1, characterized in that: By setting a flexible piezoresistive sensor on the patient's joint, the patient's real-time muscle force signal characteristic data is collected by the flexible piezoresistive sensor, and the patient's real-time muscle force signal characteristic data is processed by the Kalman filter method, specifically: A flexible piezoresistive sensor is arranged on the joint of the patient to form a flexible piezoresistive sensor sequence, and the real-time muscle force signal characteristic data of the patient is collected through the flexible piezoresistive sensor to determine the effective muscle force signal characteristic data range; Setting a system state vector based on the effective muscle force signal characteristic data range by the Kalman filtering method, setting the patient's real-time muscle force signal characteristic data as an observation vector, and setting a system model based on the system state vector; An observation model is set based on the observation vector, a state estimate and an error covariance matrix at a current moment are predicted based on the system model and the observation model, and a Kalman gain is calculated based on the state estimate and the error covariance matrix at the current moment; Based on the Kalman gain update state estimation and covariance matrix, adaptive filtering is performed in combination with the weights updated by the Kalman filter to complete the denoising of the patient's real-time muscle force signal feature data.

3. The method for data collection of a rehabilitation robot based on intention perception according to claim 1, characterized in that: The processed real-time muscle force signal characteristic data of the patient is obtained, and a patient-specific biomechanical intention model is constructed according to the processed real-time muscle force signal characteristic data of the patient, specifically: Acquire the processed real-time muscle force signal characteristic data of the patient, construct a biological joint model of the patient through three-dimensional modeling software, and preview the processed real-time muscle force signal characteristic data of the patient in the biological joint model of the patient; Based on the processed real-time muscle force signal characteristic data of the patient, the finite element analysis method is used to preview the force distribution of the tissue, obtain a schematic diagram of the force distribution of the tissue, and construct a patient-specific biomechanical intention model based on the schematic diagram of the force distribution of the tissue.

4. The method for data collection of a rehabilitation robot based on intention perception according to claim 1, characterized in that: Initialize the motion parameters of the rehabilitation robot, perform a comfort preview and analysis of the rehabilitation movement based on the motion parameters of the rehabilitation robot and the patient-specific biomechanical intention model, and obtain the patient's rehabilitation movement comfort, specifically: Setting an exercise time threshold, obtaining the patient's tissue force threshold range in each rehabilitation state through big data, and setting comfort evaluation index characteristics according to the patient's tissue force threshold range and exercise time threshold in each rehabilitation state; Initializing and configuring the motion parameters of the rehabilitation robot including motion speed, motion torque information, and motion angle information according to the patient-specific biomechanical intention model; Perform motion preview according to the motion parameters of the rehabilitation robot, including motion speed, motion torque information, and motion angle information; By performing exercise preview within the exercise time threshold, the comfort level of the patient-specific biomechanical intention model is evaluated based on the comfort level evaluation index characteristics to obtain the patient's rehabilitation exercise comfort level.

5. The method for data collection of a rehabilitation robot based on intention perception according to claim 4, characterized in that: Set the motion time threshold, specifically: The best exercise time for patients in each rehabilitation state is obtained through big data, and a knowledge graph is constructed. A graph neural network is introduced to input the best exercise time for patients in each rehabilitation state into the graph neural network for processing; Taking the rehabilitation status and the optimal exercise time as graph nodes of the graph neural network, constructing a directed description relationship, connecting the graph nodes based on the directed description relationship, constructing a topological structure graph, and inputting the topological structure graph into the knowledge graph for node representation; Obtain the rehabilitation status information of the current patient, input the rehabilitation status information of the current patient into the knowledge graph for analysis, obtain the best exercise time for the patient during the current rehabilitation state, and set the exercise time threshold according to the best exercise time for the patient during the current rehabilitation state.

6. The method for data collection of a rehabilitation robot based on intention perception according to claim 1, characterized in that: The motion parameters of the rehabilitation robot are analyzed and configured based on the patient's rehabilitation motion comfort level to obtain the final motion parameters of the rehabilitation robot, specifically: Introducing a particle swarm algorithm, setting the number of iterations based on the particle swarm algorithm, and judging whether the patient's rehabilitation exercise comfort is not lower than a preset rehabilitation exercise comfort evaluation index; When the patient's rehabilitation exercise comfort level is not lower than a preset rehabilitation exercise comfort level evaluation index, exercising according to the current exercise parameters of the rehabilitation robot; When the patient's rehabilitation exercise comfort is lower than a preset rehabilitation exercise comfort evaluation index, iteration is performed based on the number of iterations to reset the motion parameters of the rehabilitation robot until the patient's rehabilitation exercise comfort is no lower than the preset rehabilitation exercise comfort evaluation index.

7. The rehabilitation robot data acquisition system based on intention perception is characterized by: It comprises a memory and a processor, wherein the memory comprises a rehabilitation robot data collection method program based on intention perception, and when the rehabilitation robot data collection method program based on intention perception is executed by the processor, the steps of the rehabilitation robot data collection method based on intention perception as described in any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium, characterized in that: It includes a rehabilitation robot data collection method program based on intention perception. When the rehabilitation robot data collection method program based on intention perception is executed by a processor, the steps of the rehabilitation robot data collection method based on intention perception as described in any one of claims 1 to 6 are implemented.

Images