Upper limb cranial nerve rehabilitation training system and training method

Abstract

The invention relates to the field of medical rehabilitation, and particularly discloses an upper limb cranial nerve rehabilitation training system and method, and the system comprises an information collection module which collects the upper limb movement track information and upper limb training parameters of a historical trainer according to a multi-dimensional sensor; the virtual reality and man-machine interaction module displays a training virtual reality scene according to the head-mounted display of the trainee, controls the robot to drive the upper limbs of the trainee to move according to the training virtual reality scene, and controls the head-mounted loudspeaker of the trainee to feed back robot training information; the trajectory tracking module obtains upper limb training feedback of the trainee after man-machine interaction in real time and tracks and follows an upper limb training action trajectory in real time; the early warning analysis module performs analysis according to the tracking and following result of the upper limb training action trajectory and receives upper limb training feedback and early warning of the trainee; the feedback module adjusts the upper limb movement track information of the trainee according to the upper limb training feedback and updates the three-dimensional track model.

Description

Technical Field

[0001] The present invention relates to the field of medical rehabilitation, and in particular to an upper limb cranial nerve rehabilitation training system and a training method. Background Art

[0002] Stroke is one of the main causes of adult disability, especially upper limb dysfunction, which seriously affects patients' daily living ability and quality of life. With the aging of the population and the increasing incidence of related diseases, the demand for upper limb neurological rehabilitation is growing.

[0003] Although traditional rehabilitation treatment methods are effective, they are constrained by the insufficient number of rehabilitation therapists and limited treatment efficiency. In recent years, the development of rehabilitation robot technology has provided a new way to solve this problem. Rehabilitation robots can provide standardized and highly repetitive training, and at the same time, intelligent control systems can adjust training parameters according to the specific conditions of patients to improve rehabilitation effects.

[0004] In existing technologies, the combination of non-invasive brain stimulation technology (such as transcranial direct current stimulation, tDCS) and rehabilitation robot training has been shown in clinical studies to further promote the recovery of upper limb motor function in stroke patients. The combination of these technologies utilizes the principle of neuroplasticity and enhances the reorganization of neural circuits and functional recovery by stimulating the cerebral cortex.

[0005] Although clinical sports rehabilitation training robot control technology has achieved certain upper limb rehabilitation treatment effects, there are still some problems in the process of applying it to actual operations to restore patients' upper limb abilities: due to the influence of the rehabilitation model required by the trainee, the severity of the trainee's condition, and their physical fitness factors, in actual rehabilitation training, it still relies on the rehabilitation techniques and practical operation experience of medical staff, and the effect of using robots to achieve rehabilitation training is less than ideal. Summary of the invention

[0006] The purpose of the present invention is to provide an upper limb cranial nerve rehabilitation training system and training method to solve the following technical problems:

[0007] How to improve the effect of intelligent interactive training in upper limb cranial nerve rehabilitation training and optimize the rehabilitation experience of trainees.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] An upper limb cranial nerve rehabilitation training system includes a robot capable of driving a trainee's upper limbs to move, a trainee's head-mounted display, an upper limb training control system and a trainee's head-mounted speaker;

[0010] The upper limb training control system includes:

[0011] An information collection module is used to collect historical trainees' upper limb motion trajectory information and upper limb training parameters based on multi-dimensional sensors;

[0012] The upper limb motion trajectory information is realized by simulating a preset three-dimensional trajectory model through machine training, and the upper limb motion trajectory information includes a three-dimensional motion trajectory image and trajectory image data information;

[0013] Upper limb training parameters include upper limb training sites, upper limb training intensity levels, and upper limb training frequency;

[0014] A virtual reality and human-computer interaction module, which is used to display the training virtual reality scene according to the trainee's head-mounted display, and to control the robot to drive the trainee's upper limbs to perform movements according to the training virtual reality scene and control the trainee's head-mounted speaker to feedback the robot training information;

[0015] The trajectory tracking module is used to obtain the trainee's upper limb training feedback in real time after human-computer interaction, and to track the upper limb training movement trajectory in real time;

[0016] The early warning analysis module is used to perform standard analysis based on the tracking results of the upper limb training movement trajectory:

[0017] When the standard analysis meets the requirements, early warning analysis is performed;

[0018] When the standard analysis does not meet the requirements, the first warning signal is issued;

[0019] When the early warning analysis meets the requirements, receive upper limb training feedback from the trainer;

[0020] When the early warning analysis does not meet the requirements, a second early warning signal is issued;

[0021] The feedback module is used to adjust the trainee's upper limb motion trajectory information according to the upper limb training feedback and update the three-dimensional trajectory model.

[0022] Preferably, the process of standard analysis is:

[0023] Design a tracking controller to obtain the dynamic equations of multiple motion joints of the robot;

[0024] The position tracking error range threshold is set according to the actual position, motion speed and acceleration of each motion joint in the dynamic equation;

[0025] The historical trajectory image data information and dynamic equations are input into the neural network model for training to output the actual position tracking error.

[0026] Preferably, the actual position tracking error is compared with a position tracking error range threshold:

[0027] If the actual position tracking error is less than the position tracking error range threshold, it is determined that the current position standard analysis meets the requirements; otherwise, the current position standard analysis does not meet the requirements.

[0028] Preferably, the process of early warning analysis is:

[0029] Get the curve E of the actual position tracking error of each point of the trainee's upper limb training changing with time i ;

[0030] Collect the change curve E of the same upper limb training frequency at all upper limb training sites i The set A∈{E1, E2, …, E n}, n is the total number of upper limb training sites, and i∈[1,n];

[0031] The standard change curve E of the same upper limb training frequency at the same upper limb training site in the history thr The set X of

[0032] Determine the number of elements x in set A that fall into set X:

[0033] If B is greater than the preset standard quantity threshold B ith , then it is judged that the current early warning analysis meets the requirements;

[0034] If B is less than the preset standard quantity threshold B ith , then it is judged that the current early warning analysis does not meet the requirements.

[0035] Preferably, upper limb training feedback includes:

[0036] The trainer receives the robot training information and selects the training mode; the training modes include: passive training, active training, and power training;

[0037] According to the training mode selected by the trainer, the upper limb training parameters are collected to perform robot training stability analysis, and whether the training stability analysis results meet the requirements:

[0038] If yes, continue with the current mode and generate the control strategy for the current mode;

[0039] If not, switch to other modes automatically.

[0040] Preferably, the training stability analysis method is:

[0041] By formula Calculate and obtain the stability coefficient Sta;

[0042] Among them, V i is the robot joint position parameter of the i-th upper limb training site; is the robot moving speed parameter of the i-th upper limb training site; is the robot movement acceleration parameter of the i-th upper limb training position; V i0 is the standard joint position parameter of the robot at the i-th upper limb training site; is the standard moving speed parameter of the robot at the i-th upper limb training site; is the robot standard movement acceleration parameter of the i-th upper limb training site; ΔV is the preset joint position parameter deviation value; is the preset moving speed parameter deviation value; is the preset movement acceleration parameter deviation value; F is the upper limb training frequency; S is the upper limb training intensity; θ(F, S) is the upper limb training mode factor influence function of the joint position parameter; ∈(F, S) is the movement speed parameter upper limb training mode factor influence function; δ(F, S) is the movement acceleration parameter upper limb training mode factor influence function;

[0043] The stability coefficient Sta is compared with the preset stability coefficient threshold interval [Sta X , Sta Y ] for comparison:

[0044] If Sta∈[Sta X , Sta Y ], it is judged that the robot is in a stable state when executing the current training mode;

[0045] like It is judged that the robot is in an unstable state when executing the current training mode.

[0046] Preferably, the system further comprises:

[0047] The trainee evaluation module is used to obtain the evaluation of the rehabilitation training effect based on the trainee's comprehensive sensitivity to each upper limb training point; the sensitivity is a parameter selected by the trainee's continuous pain, soreness, and tolerance time level of the current upper limb training point; within the set tolerance time level range, the greater the sensitivity, the better the patient's rehabilitation effect.

[0048] Preferably, the trainer evaluation module includes:

[0049] By formula Calculate the recovery coefficient Re;

[0050] Among them, Per i is the sensitivity parameter of the i-th upper limb training site; Per i0 is the standard sensitivity parameter of the i-th upper limb training site; σ i Set the rehabilitation influence function of the tolerance time level for the i-th upper limb training site.

[0051] Preferably, the method for obtaining the evaluation of the effect of this rehabilitation training is:

[0052] The recovery coefficient Re and the standard recovery coefficient Re ith To compare:

[0053] If Re≥Re ith , then it is judged that the current rehabilitation effect is good.

[0054] An upper limb cranial nerve rehabilitation training method is applied to an upper limb cranial nerve rehabilitation training system. The method is implemented by a robot capable of driving a trainee's upper limb movement, a trainee's head-mounted display, an upper limb training control system, and a trainee's head-mounted speaker, and includes:

[0055] S1. Collecting upper limb motion trajectory information and upper limb training parameters of historical trainees according to multidimensional sensors; the upper limb motion trajectory information is simulated by a preset three-dimensional trajectory model of machine training, and the upper limb motion trajectory information includes a three-dimensional motion trajectory image and trajectory image data information; the upper limb training parameters include upper limb training sites, upper limb training intensity levels, and upper limb training frequencies;

[0056] S2, displaying the training virtual reality scene according to the trainee's head-mounted display, and controlling the robot to drive the trainee's upper limbs to perform movements according to the training virtual reality scene and controlling the trainee's head-mounted speaker to feed back the robot training information;

[0057] S3, real-time acquisition of upper limb training feedback from the trainee after human-computer interaction, and real-time tracking of upper limb training movement trajectory;

[0058] S4. Perform standard analysis based on the tracking results of the upper limb training movement trajectory:

[0059] When the standard analysis meets the requirements, early warning analysis is performed;

[0060] When the standard analysis does not meet the requirements, the first warning signal is issued;

[0061] When the early warning analysis meets the requirements, receive upper limb training feedback from the trainer;

[0062] When the early warning analysis does not meet the requirements, a second early warning signal is issued;

[0063] S5. Adjust the trainee's upper limb motion trajectory information according to upper limb training feedback and update the three-dimensional trajectory model.

[0064] Beneficial effects of the present invention:

[0065] (1) The present invention realizes the systematized process of rehabilitation training of the trainee's upper limb cranial nerves by setting up four parts: a robot that can drive the trainee's upper limb movement, a trainee's head-mounted display, an upper limb training control system and a trainee's head-mounted speaker. The training items are displayed on the trainee's head-mounted display, the trainee selects the training item mode, and the robot starts the corresponding upper limb movement training instructions according to the training item mode; the training results are also fed back through the trainee's head-mounted speaker, and the trainee can also issue corresponding feedback on the rehabilitation training effect, so that medical staff and management personnel can directly receive the trainee's rehabilitation training information, and the entire rehabilitation training is completed through interactive training between the trainee and the robot.

[0066] (2) The present invention uses an upper limb training control system to issue control instructions and transmit upper limb movement training instructions; specifically, by setting a trajectory tracking module, it is ensured that the trainee's upper limb training feedback after human-computer interaction is obtained in real time, and the upper limb training movement trajectory is tracked in real time, so as to ensure the monitoring of the trajectory data during the robot's training and rehabilitation process, and ensure the robot's good trajectory tracking ability; by setting an early warning analysis module, it is ensured that a progressive analysis process of standard analysis and early warning analysis is performed on the acquired upper limb training data, and finally feedback on the trainee's upper limb training situation is achieved and the trainee's trajectory information is adjusted in time according to the analysis results; by setting a feedback module, the trainee's upper limb movement trajectory information is adjusted according to the upper limb training feedback to achieve an update process of a three-dimensional data model such as a three-dimensional trajectory model, so as to ensure further optimization of the robot's human-computer interaction rehabilitation training effect.

[0067] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0069] Figure 1 This is a structural diagram of an upper limb cranial nerve rehabilitation training system of the present invention;

[0070] Figure 2 The present invention is a step diagram of an upper limb cranial nerve rehabilitation training method. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0072] Although clinical sports rehabilitation training robot control technology has achieved certain upper limb rehabilitation treatment effects, there are still some problems in the process of applying it to actual operations to restore patients' upper limb abilities: due to the influence of the rehabilitation model required by the trainee, the severity of the trainee's condition, and their physical fitness factors, in actual rehabilitation training, it still relies on the rehabilitation techniques and practical operation experience of medical staff, and the effect of using robots to achieve rehabilitation training is less than ideal.

[0073] To change this situation, see Figure 1-2 As shown, by designing an upper limb cranial nerve rehabilitation training system and training method, real-time feedback of rehabilitation training effects can be achieved, the interaction effect between trainers and machines can be enriched, and the rehabilitation experience of trainers can be optimized.

[0074] The present invention is an upper limb cranial nerve rehabilitation training system, comprising a robot capable of driving a trainee's upper limbs to move, a trainee's head-mounted display, an upper limb training control system and a trainee's head-mounted loudspeaker;

[0075] The upper limb training control system includes:

[0076] An information collection module is used to collect historical trainees' upper limb motion trajectory information and upper limb training parameters based on multi-dimensional sensors;

[0077] The upper limb motion trajectory information is realized by simulating a preset three-dimensional trajectory model through machine training, and the upper limb motion trajectory information includes a three-dimensional motion trajectory image and trajectory image data information;

[0078] Upper limb training parameters include upper limb training sites, upper limb training intensity levels, and upper limb training frequency;

[0079] A virtual reality and human-computer interaction module, which is used to display the training virtual reality scene according to the trainee's head-mounted display, and to control the robot to drive the trainee's upper limbs to perform movements according to the training virtual reality scene and control the trainee's head-mounted speaker to feedback the robot training information;

[0080] The trajectory tracking module is used to obtain the trainee's upper limb training feedback in real time after human-computer interaction, and to track the upper limb training movement trajectory in real time;

[0081] The early warning analysis module is used to perform standard analysis based on the tracking results of the upper limb training movement trajectory:

[0082] When the standard analysis meets the requirements, early warning analysis is performed;

[0083] When the standard analysis does not meet the requirements, the first warning signal is issued;

[0084] When the early warning analysis meets the requirements, receive upper limb training feedback from the trainer;

[0085] When the early warning analysis does not meet the requirements, a second early warning signal is issued;

[0086] The feedback module is used to adjust the trainee's upper limb motion trajectory information according to the upper limb training feedback and update the three-dimensional trajectory model.

[0087] In the above technical scheme, in this embodiment, the systematization process of the rehabilitation training of the upper limb cranial nerves of the trainee is realized by setting a robot that can drive the trainee's upper limb movement, a trainee's head-mounted display, an upper limb training control system and a trainee's head-mounted speaker. The training project is displayed through the trainee's head-mounted display, the trainee selects the training project mode, and the robot starts the corresponding upper limb movement training instruction according to the training project mode; and the acquisition of the instruction is controlled by the upper limb training control system to issue and transmit the control instruction. In addition, the training results are also fed back through the trainee's head-mounted speaker, and the trainee can also issue corresponding feedback on the rehabilitation training effect, so that medical care and management personnel can directly receive the trainee's rehabilitation training information, and the entire rehabilitation training is completed through the interactive training of the trainer and the robot. Among them, the specific software and hardware interaction process of this design also uses big data processing technology, intelligent model analysis training and verification process, so as to make real-time adjustments to the trajectory of the robot's rehabilitation training to ensure the improvement of the trainee's rehabilitation experience.

[0088] Specifically, the software design modules of the upper limb training control system mainly involved in this design mainly include information acquisition module, virtual reality and human-computer interaction module, trajectory tracking module, early warning analysis module, feedback module. The above five modules realize the optimization process of the trainer's upper limb training data, specifically:

[0089] Firstly, by setting up an information collection module, the upper limb motion trajectory information and upper limb training parameters of historical trainees are collected according to multi-dimensional sensors, thereby ensuring the acquisition of real-time training data and the collection process of historical data.

[0090] Among them, the acquisition of upper limb motion trajectory information is realized by simulating a preset three-dimensional trajectory model through machine training, and the upper limb motion trajectory information mainly includes a three-dimensional motion trajectory image and trajectory image data information; the upper limb motion trajectory data can intuitively and digitally reflect the training method of the current trainee; the upper limb training parameters set in this embodiment include upper limb training sites, upper limb training intensity levels, and upper limb training frequencies; through the analysis of different training sites of the patient's upper limbs, the selection of key training sites, upper limb training intensity levels, and upper limb training frequencies for rehabilitation training of different trainees is realized.

[0091] Secondly, by setting up a virtual reality and human-computer interaction module, the robot and the trainer can realize the interaction process through the interactive device according to the virtual reality and human-computer interaction module, ensure the accurate transmission of rehabilitation training information, and improve the interaction effect. Specifically, this design mainly includes a head-mounted display and a head-mounted speaker; the training virtual reality scene is displayed according to the trainer's head-mounted display, and the robot can be controlled in real time to drive the trainer's upper limbs to perform movements according to the training virtual reality scene; the trainer's head-mounted speaker is controlled to feedback the robot training information.

[0092] Thirdly, by setting up a trajectory tracking module, it is ensured that the trainee's upper limb training feedback is obtained in real time after human-computer interaction, and the upper limb training movement trajectory is tracked in real time, so as to ensure the monitoring of trajectory data during further training and rehabilitation of the robot and ensure the robot's good trajectory tracking capability. The main equipment involved includes a trajectory tracking controller.

[0093] Fourthly, by setting up the early warning analysis module, a progressive analysis process of standard analysis and early warning analysis of the acquired upper limb training data is ensured, ultimately achieving feedback on the trainee's upper limb training situation and further timely adjusting the trainee's trajectory information based on the analysis results.

[0094] Fifth, by setting up a feedback module to adjust the trainee's upper limb motion trajectory information according to the upper limb training feedback, the three-dimensional data model, such as the three-dimensional trajectory model, is updated to ensure further optimization of the robot's human-computer interaction rehabilitation training effect.

[0095] As an embodiment of the present invention, the process of standard analysis is:

[0096] Design a tracking controller to obtain the dynamic equations of multiple motion joints of the robot;

[0097] The position tracking error range threshold is set according to the actual position, motion speed and acceleration of each motion joint in the dynamic equation;

[0098] The historical trajectory image data information and dynamic equations are input into the neural network model for training to output the actual position tracking error.

[0099] In the above technical scheme, the standard analysis process in this embodiment ensures the acquisition of actual data errors in position tracking, and ensures that the dynamic equations of multiple motion joints of the robot are first obtained according to the designed trajectory tracking controller; then the position tracking error range threshold is set according to the actual position, motion speed and acceleration of each motion joint in the dynamic equation, and the accurate acquisition of the robot rehabilitation training trajectory is ensured by obtaining data information on the robot's motion conditions; finally, by introducing models related to machine algorithms, such as a neural network model, the historical trajectory image data information and the dynamic equation are taken as input, and the actual position tracking error is output through multiple machine training.

[0100] As an implementation of the present invention, the actual position tracking error is compared with the position tracking error range threshold:

[0101] If the actual position tracking error is less than the position tracking error range threshold, it is determined that the current position standard analysis meets the requirements; otherwise, the current position standard analysis does not meet the requirements.

[0102] In the above technical scheme, this embodiment also determines whether the corresponding position meets the normal deviation range based on the output actual position tracking error, and compares the actual position tracking error with the position tracking error range threshold. For objects that meet the standard analysis, the tracking error range threshold is used as the maximum critical value standard. If it is less than the current standard, it meets the standard analysis requirements. Otherwise, it is judged that it does not meet the standard analysis requirements.

[0103] As an implementation mode of the present invention, the process of early warning analysis is:

[0104] Get the curve E of the actual position tracking error of each point of the trainee's upper limb training changing with time i ;

[0105] Collect the change curve E of the same upper limb training frequency at all upper limb training sites i The set A∈{E1, E2, …, E n}, n is the total number of upper limb training sites, and i∈[1,n];

[0106] The standard change curve E of the same upper limb training frequency at the same upper limb training site in the history thr The set X of

[0107] Determine the number of elements x in set A that fall into set X:

[0108] If B is greater than the preset standard quantity threshold B ith , then it is judged that the current early warning analysis meets the requirements;

[0109] If B is less than the preset standard quantity threshold Bith , then it is judged that the current early warning analysis does not meet the requirements.

[0110] In the above technical solution, in this embodiment, a warning analysis is performed for the situation that meets the preliminary standard analysis. Since the frequency and intensity of upper limb training required for different upper limb training points are different, in this embodiment, the change amplitude of the upper limb training frequency is used as a reference for the warning analysis. Of course, in actual operation, the training intensity of different upper limb training points can also be used as a judgment, and the decision is made according to the actual specific operation; therefore, the judgment process of the warning analysis includes: first, obtaining the curve E of the actual position tracking error of each upper limb training point of the trainer changing with time i ; Then, collect the change curve E of the same upper limb training frequency of all upper limb training sites i The set A∈{E1, E2, …, E n}, n is the total number of upper limb training sites, and i∈[1,n]; then, collect the standard change curve E of the same upper limb training frequency at the same upper limb training site in history thr Finally, determine the number of elements x in set A that fall into set X: if B is greater than the preset standard number threshold B ith , then the current warning analysis meets the requirements; if B is less than the preset standard quantity threshold B ith , then it is judged that the current early warning analysis does not meet the requirements.

[0111] As an embodiment of the present invention, upper limb training feedback includes:

[0112] The trainer receives the robot training information and selects the training mode; the training modes include: passive training, active training, and power training;

[0113] According to the training mode selected by the trainer, the upper limb training parameters are collected to perform robot training stability analysis, and whether the training stability analysis results meet the requirements:

[0114] If yes, continue with the current mode and generate the control strategy for the current mode;

[0115] If not, switch to other modes automatically.

[0116] In the above technical scheme, the main content of the upper limb training feedback in this embodiment is to select the robot training information mode received by the trainee and further judge whether the current mode meets the stability requirements of the training based on the selected training mode. For the trainee's upper limb rehabilitation training, a relatively stable training mode is more in line with the current upper limb rehabilitation status. Therefore, the trainee shortens the upper limb rehabilitation training cycle and speeds up the rehabilitation training process through multiple training controls under the stable mode; and for training situations that do not meet the current mode, the mode is automatically or manually changed. The automatic change method is adopted in this embodiment to achieve the intelligent needs.

[0117] It should also be noted that the training modes in this embodiment mainly include three modes: passive training, active training, and assisted training. Different modes have different requirements for patients' upper limb training. In actual training, the current training mode can be further refined to ensure that the diverse needs of upper limb rehabilitation of a wider range of trainees are met.

[0118] As an implementation mode of the present invention, the training stability analysis method is:

[0119] By formula Calculate and obtain the stability coefficient Sta;

[0120] Among them, V i is the robot joint position parameter of the i-th upper limb training site; is the robot moving speed parameter of the i-th upper limb training site; is the robot movement acceleration parameter of the i-th upper limb training position; V i0 is the standard joint position parameter of the robot at the i-th upper limb training site; is the standard moving speed parameter of the robot at the i-th upper limb training site; is the robot standard movement acceleration parameter of the i-th upper limb training site; ΔV is the preset joint position parameter deviation value; is the preset moving speed parameter deviation value; is the preset movement acceleration parameter deviation value; F is the upper limb training frequency; S is the upper limb training intensity; θ(F, S) is the upper limb training mode factor influence function of the joint position parameter; ∈(F, S) is the movement speed parameter upper limb training mode factor influence function; δ(F, S) is the movement acceleration parameter upper limb training mode factor influence function;

[0121] The stability coefficient Sta is compared with the preset stability coefficient threshold interval [Sta X , Sta Y ] for comparison:

[0122] If Sta∈[Sta X , Sta Y ], it is judged that the robot is in a stable state when executing the current training mode;

[0123] like It is judged that the robot is in an unstable state when executing the current training mode.

[0124] In the above technical solution, the specific method of training stability analysis in this embodiment is realized by a calculation formula, mainly through the formula The stability coefficient Sta is calculated and obtained; by obtaining the stability coefficient and according to the influence of its size, the actual position and movement speed of each moving joint of the robot, the acceleration value and the setting of its change influence function are obtained to judge the stability of the current training mode of the robot.

[0125] The stability coefficient needs further explanation. Since the robot needs to design joint movements during the rehabilitation training of the trainee, the change in the amplitude of joint movement further reflects whether the patient is adapted to this mode. When the patient is not adapted, he will struggle and work hard. At this time, the stability will be affected by the tension of the human body's struggle. Another situation is that the patient's rehabilitation part is not properly trained. The movement amplitude of each joint of the machine will also show continuous abnormal changes, such as the actual position of the joint will be offset.

[0126] It should be noted that V i0 , All of them are obtained by simulation based on historical data and will not be described in detail here; ΔV, All of them are selected and set based on empirical data and will not be described in detail here. The relationship between θ(F, S), ∈(F, S), δ(F, S) and the upper limb training frequency F and the upper limb training intensity S is selected and obtained based on the average values ​​of various parameters of the robot under different upper limb training modes. θ(F, S), ∈(F, S), and δ(F, S) will be adjusted according to the change of the current upper limb training mode, and then under different upper limb training modes, the stability coefficient Sta will be adjusted accordingly to improve the accuracy of judgment under different upper limb training modes.

[0127] As an embodiment of the present invention, the system further includes:

[0128] The trainee evaluation module is used to obtain the evaluation of the rehabilitation training effect based on the trainee's comprehensive sensitivity to each upper limb training point; the sensitivity is a parameter selected by the trainee's continuous pain, soreness, and tolerance time level of the current upper limb training point; within the set tolerance time level range, the greater the sensitivity, the better the patient's rehabilitation effect.

[0129] In the above technical scheme, in addition to expressing the patient's reflections during the interaction process through intuitive language information and expressions, the system also detects the patient's sensitivity during the interaction process, reflects the comprehensive sensitivity of each upper limb training point to obtain the evaluation of the effectiveness of this rehabilitation training, including the parameters selected by the current upper limb training point's continuous pain, soreness, and tolerance time level; this parameter is obtained by selecting and setting the evaluation option during the patient's training interaction process. Of course, in the actual implementation process, it still needs to be accurately confirmed by medical staff.

[0130] As an embodiment of the present invention, the trainer evaluation module includes:

[0131] By formula Calculate the recovery coefficient Re;

[0132] Among them, Per i is the sensitivity parameter of the i-th upper limb training site; Per i0 is the standard sensitivity parameter of the i-th upper limb training site; σ i Set the rehabilitation influence function of the tolerance time level for the i-th upper limb training site.

[0133] In the above technical solution, the trainee evaluation module in this embodiment obtains the rehabilitation coefficient Re through a formula calculation. The rehabilitation coefficient can reflect the effect of the current rehabilitation training, which is specifically determined by the trainee's perception and the influence of the trainee's current tolerance time level on the rehabilitation effect. And Per i0 It is obtained by machine simulation based on historical data; the rehabilitation influence function σ i It is an influence function set according to historical data to ensure that the result of the rehabilitation coefficient is within a specific reasonable range.

[0134] As an implementation of the present invention, a method for obtaining the evaluation of the effect of this rehabilitation training is as follows:

[0135] The recovery coefficient Re and the standard recovery coefficient Re ith To compare:

[0136] If Re≥Re ith , then it is judged that the current rehabilitation effect is good.

[0137] In the above technical solution, this embodiment determines the rehabilitation training effect by comparing the obtained rehabilitation coefficient with the set standard threshold, mainly by judging whether the rehabilitation coefficient Re is greater than the standard rehabilitation coefficient Re ith , if it is greater than or equal to the current standard rehabilitation coefficient Re ith , it is judged that the current rehabilitation training meets the expected effect.

[0138] See also Figure 2 As shown, an upper limb cranial nerve rehabilitation training method is also designed and applied to an upper limb cranial nerve rehabilitation training system. The method is implemented by a robot that can drive the trainee's upper limb movement, a trainee's head-mounted display, an upper limb training control system and a trainee's head-mounted speaker, including:

[0139] S1. Collecting upper limb motion trajectory information and upper limb training parameters of historical trainees according to multidimensional sensors; the upper limb motion trajectory information is simulated by a preset three-dimensional trajectory model of machine training, and the upper limb motion trajectory information includes a three-dimensional motion trajectory image and trajectory image data information; the upper limb training parameters include upper limb training sites, upper limb training intensity levels, and upper limb training frequencies;

[0140] S2, displaying the training virtual reality scene according to the trainee's head-mounted display, and controlling the robot to drive the trainee's upper limbs to perform movements according to the training virtual reality scene and controlling the trainee's head-mounted speaker to feed back the robot training information;

[0141] S3, real-time acquisition of upper limb training feedback from the trainee after human-computer interaction, and real-time tracking of upper limb training movement trajectory;

[0142] S4. Perform standard analysis based on the tracking results of the upper limb training movement trajectory:

[0143] When the standard analysis meets the requirements, early warning analysis is performed;

[0144] When the standard analysis does not meet the requirements, the first warning signal is issued;

[0145] When the early warning analysis meets the requirements, receive upper limb training feedback from the trainer;

[0146] When the early warning analysis does not meet the requirements, a second early warning signal is issued;

[0147] S5. Adjust the trainee's upper limb motion trajectory information according to upper limb training feedback and update the three-dimensional trajectory model.

[0148] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0149] The above specific embodiments of this specification are described. Other embodiments are within the scope of the attached documents. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0150] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this application, they shall all fall within the protection scope of the present invention.

Claims

1. An upper limb cranial nerve rehabilitation training system, characterized in that: It includes a robot capable of driving the trainee's upper limbs to move, a head-mounted display for the trainee, an upper limb training control system and a head-mounted speaker for the trainee; The upper limb training control system comprises: An information collection module is used to collect historical trainees' upper limb motion trajectory information and upper limb training parameters based on multi-dimensional sensors; The upper limb motion trajectory information is realized by simulating a preset three-dimensional trajectory model through machine training, and the upper limb motion trajectory information includes a three-dimensional motion trajectory image and trajectory image data information; The upper limb training parameters include upper limb training sites, upper limb training intensity levels, and upper limb training frequencies; A virtual reality and human-computer interaction module, used to display a training virtual reality scene according to a trainer's head-mounted display, and to control the robot to drive the trainer's upper limbs to perform movements according to the training virtual reality scene and to control the trainer's head-mounted speaker to feed back robot training information; The trajectory tracking module is used to obtain the trainee's upper limb training feedback in real time after human-computer interaction, and to track the upper limb training movement trajectory in real time; The early warning analysis module is used to perform standard analysis based on the tracking results of the upper limb training movement trajectory: When the standard analysis meets the requirements, early warning analysis is performed; When the standard analysis does not meet the requirements, the first warning signal is issued; When the early warning analysis meets the requirements, receive upper limb training feedback from the trainer; When the early warning analysis does not meet the requirements, a second early warning signal is issued; The feedback module is used to adjust the trainee's upper limb motion trajectory information according to the upper limb training feedback and update the three-dimensional trajectory model.

2. The upper limb cranial nerve rehabilitation training system according to claim 1, characterized in that: The process of the standard analysis is: Designing a tracking controller to obtain dynamic equations of multiple motion joints of the robot; The position tracking error range threshold is set according to the actual position, motion speed and acceleration of each motion joint in the dynamic equation; The historical trajectory image data information and dynamic equations are input into the neural network model for training to output the actual position tracking error.

3. An upper limb cranial nerve rehabilitation training system according to claim 2, characterized in that: Compare the actual position tracking error to the position tracking error range threshold: If the actual position tracking error is less than the position tracking error range threshold, it is determined that the current position standard analysis meets the requirements; otherwise, the current position standard analysis does not meet the requirements.

4. The upper limb cranial nerve rehabilitation training system according to claim 3, characterized in that: The process of early warning analysis is as follows: Get the curve E of the actual position tracking error of each point of the trainee's upper limb training changing with time i ; Collect the change curve E of the same upper limb training frequency at all upper limb training sites i The set A∈{E1, E2, ..., E n }, n is the total number of upper limb training sites, and i∈[1,n]; The standard change curve E of the same upper limb training frequency at the same upper limb training site in the history thr The set X of Determine the number of elements x in set A that fall into set X: If B is greater than the preset standard quantity threshold B ith , then it is judged that the current early warning analysis meets the requirements; If B is less than the preset standard quantity threshold B ith , then it is judged that the current early warning analysis does not meet the requirements.

5. The upper limb cranial nerve rehabilitation training system according to claim 1, characterized in that: The upper limb training feedback includes: The trainer receives the robot training information and selects a training mode; the training modes include: passive training, active training, and power training; Collect upper limb training parameters according to the training mode selected by the trainer to perform robot training stability analysis and determine whether the training stability analysis results meet the requirements: If yes, continue with the current mode and generate the control strategy for the current mode; If not, switch to other modes automatically.

6. An upper limb cranial nerve rehabilitation training system according to claim 5, characterized in that: The training stability analysis method is: By formula Calculate and obtain the stability coefficient Sta; Among them, V i is the robot joint position parameter of the i-th upper limb training site; is the robot moving speed parameter of the i-th upper limb training site; is the robot movement acceleration parameter of the i-th upper limb training position; V i0 is the standard joint position parameter of the robot at the i-th upper limb training site; is the standard moving speed parameter of the robot at the i-th upper limb training site; is the robot standard movement acceleration parameter of the i-th upper limb training site; ΔV is the preset joint position parameter deviation value; is the preset moving speed parameter deviation value; is the preset movement acceleration parameter deviation value; F is the upper limb training frequency; S is the upper limb training intensity; θ(F, S) is the upper limb training mode factor influence function of the joint position parameter; ∈(F, S) is the movement speed parameter upper limb training mode factor influence function; δ(F, S) is the movement acceleration parameter upper limb training mode factor influence function; The stability coefficient Sta is compared with the preset stability coefficient threshold interval [Sta X , Sta Y ] for comparison: If Sta∈[Sta X , Sta Y ], it is judged that the robot is in a stable state when executing the current training mode; like It is judged that the robot is in an unstable state when executing the current training mode.

7. The upper limb cranial nerve rehabilitation training system according to claim 1, characterized in that: The system further comprises: The trainee evaluation module is used to obtain the evaluation of the rehabilitation training effect based on the trainee's comprehensive sensitivity to each upper limb training point; the sensitivity is a parameter selected by the trainee's continuous pain, soreness, and tolerance time level of the current upper limb training point; within the set tolerance time level range, the greater the sensitivity, the better the patient's rehabilitation effect.

8. An upper limb cranial nerve rehabilitation training system according to claim 7, characterized in that: The trainer evaluation module includes: By formula Calculate the recovery coefficient Re; Among them, Per i is the sensitivity parameter of the i-th upper limb training site; Per i0 is the standard sensitivity parameter of the i-th upper limb training site; σ i Set the rehabilitation influence function of the tolerance time level for the i-th upper limb training site.

9. An upper limb cranial nerve rehabilitation training system according to claim 8, characterized in that: Method for obtaining the evaluation of the effect of the rehabilitation training: The recovery coefficient Re and the standard recovery coefficient Re ith To compare: If Re≥Re ith , then it is judged that the current rehabilitation effect is good.

10. An upper limb cranial nerve rehabilitation training method, applied to an upper limb cranial nerve rehabilitation training system as claimed in any one of claims 1 to 9, the method is implemented by a robot capable of driving the trainee's upper limb movements, a trainee's head-mounted display, an upper limb training control system and a trainee's head-mounted speaker, characterized in that: The method comprises: S1. Collecting upper limb motion trajectory information and upper limb training parameters of historical trainees according to multidimensional sensors; the upper limb motion trajectory information is simulated by a preset three-dimensional trajectory model of machine training, and the upper limb motion trajectory information includes a three-dimensional motion trajectory image and trajectory image data information; the upper limb training parameters include upper limb training sites, upper limb training intensity levels, and upper limb training frequencies; S2, displaying a training virtual reality scene according to the trainer's head-mounted display, and controlling the robot to drive the trainer's upper limbs to perform movements according to the training virtual reality scene and controlling the trainer's head-mounted speaker to feed back robot training information; S3, real-time acquisition of upper limb training feedback from the trainee after human-computer interaction, and real-time tracking of the upper limb training movement trajectory; S4. Perform standard analysis based on the tracking results of the upper limb training movement trajectory: When the standard analysis meets the requirements, early warning analysis is performed; When the standard analysis does not meet the requirements, the first warning signal is issued; When the early warning analysis meets the requirements, receive upper limb training feedback from the trainer; When the early warning analysis does not meet the requirements, a second early warning signal is issued; S5. Adjust the trainee's upper limb motion trajectory information according to upper limb training feedback and update the three-dimensional trajectory model.

Images