Active upper limb rehabilitation training method and system based on electroencephalogram-robot cooperation

Abstract

The invention provides an upper limb active rehabilitation training method and system based on electroencephalogram-robot cooperation. The method comprises the steps that four-classification motor imagery EEG data of single-side upper limb shoulder, elbow, wrist and hand joints are collected through an electroencephalogram signal collecting system; carrying out preprocessing on the EEG (electroencephalogram) data; fusing the SE module with the EEGNet, and performing model training by using the preprocessed EEG data to obtain an EEG signal recognition model; the electroencephalogram signal recognition model is used for classifying and recognizing electroencephalogram EEG data collected in real time to obtain a control instruction; and the exoskeleton control module receives the control instruction, and drives the upper limbs to complete rehabilitation actions according to an exoskeleton control command with seven degrees of freedom of shoulders, elbows, wrists and hands. According to the method disclosed by the invention, through combination of accurate electroencephalogram signal processing, stable data transmission and exoskeleton control, efficient unilateral upper limb rehabilitation training is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of medical rehabilitation equipment, and in particular to an upper limb active rehabilitation training method and system based on EEG-robot collaboration. Background Art

[0002] With the aging of the population and the increase in neurological diseases and accidental injuries, the number of patients with upper limb dysfunction continues to rise. Traditional upper limb rehabilitation training mainly relies on manual assisted training by rehabilitation therapists and simple rehabilitation equipment training, which has problems such as low training efficiency, lack of personalization, and low patient participation. In recent years, brain-computer interface technology has gradually been applied to the field of rehabilitation. By collecting electrical activity signals from the patient's brain to control rehabilitation equipment, it has brought new ideas to upper limb rehabilitation training. However, most existing upper limb rehabilitation training systems based on brain-computer interfaces can only achieve simple two-category or three-category movement control, which cannot meet the needs of unilateral upper limb shoulder, elbow, wrist, and hand multi-joint complex movement rehabilitation training. In addition, there is room for improvement in EEG signal processing, classification algorithms, and rehabilitation exoskeleton design, making it difficult to achieve efficient and accurate rehabilitation training results. Therefore, a more advanced rehabilitation training system is urgently needed. Summary of the Invention

[0003] In view of this, the present invention provides an upper limb active rehabilitation training method and system based on EEG-robot collaboration to solve the above problems.

[0004] The present invention provides an upper limb active rehabilitation training method based on EEG-robot collaboration, comprising: collecting four-category motor imagery EEG data of the shoulder, elbow, wrist, and hand joints of a unilateral upper limb through an EEG signal acquisition system; preprocessing the EEG data; fusing an SE module with EEGNet, and using the preprocessed EEG data for model training to obtain an EEG signal recognition model; the EEG signal recognition model classifies and recognizes the real-time collected EEG data to obtain control instructions; an exoskeleton control module receives the control instructions, and controls an exoskeleton with seven degrees of freedom of the shoulder, elbow, wrist, and hand, thereby driving the upper limb to complete rehabilitation movements.

[0005] In another implementation of the present invention, the four-category motor imagery EEG data includes motor imagery EEG signals corresponding to the movement of extending the shoulder to the horizontal and then falling to the vertical, bending the elbow 90 degrees and then straightening it, internally rotating the wrist 90 degrees and then returning to the straight position, and grasping the hand and then stretching it.

[0006] In another implementation of the present invention, the preprocessing includes: basic filtering, ICA artifact removal, segmentation, baseline correction, and manual inspection.

[0007] In another implementation of the present invention, the EEGNet module includes a temporal convolution layer, a depth-wise separable convolution layer, a first random dropout layer, a first average pooling layer, a separable convolution layer, a second average pooling layer, and a second random dropout layer, which are connected in sequence.

[0008] In another implementation of the present invention, the temporal convolution layer is used to extract the frequency and spatial features of the original EEG signal; the random inactivation layer is used to prevent network overfitting; and the average pooling layer is used to reduce the feature dimension.

[0009] In another implementation of the present invention, the SE module includes a global pooling layer, a first fully connected layer, a second fully connected layer, and a normalized exponential function layer.

[0010] In another implementation of the present invention, the global pooling layer is used to compress the channel to form a compressed vector; the fully connected layer is used for dimensionality reduction and dimensionality increase.

[0011] Another aspect of the present invention provides an upper limb active rehabilitation training system based on EEG-robot collaboration, including: an EEG signal acquisition system: used to collect four-category motor imagery EEG data of the shoulder, elbow, wrist, and hand joints of a unilateral upper limb; a data processing and analysis module: used to preprocess the EEG data; the SE module is integrated with EEGNet, and the preprocessed EEG data is used for model training to obtain an EEG signal recognition model; the EEG signal recognition model classifies and recognizes the real-time collected EEG data to obtain control instructions; an exoskeleton control module: used to receive control instructions and control commands for the exoskeleton with 7 degrees of freedom of the shoulder, elbow, wrist, and hand, so as to drive the upper limb to complete rehabilitation movements.

[0012] Another aspect of the present invention provides an electronic device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of an upper limb active rehabilitation training method based on EEG-robot collaboration as described in any one of the above items are implemented.

[0013] Another aspect of the present invention provides a computer storage medium, characterized in that a computer program is stored on the computer storage medium, and when the computer program is executed by a processor, the steps of the upper limb active rehabilitation training method based on EEG-robot collaboration as described in any one of the above items are implemented.

[0014] The present invention's active upper limb rehabilitation training method based on EEG-robot collaboration decodes the subject's motor imagination intentions and converts the decoding results into control instructions for the upper limb rehabilitation exoskeleton system, thereby enabling the subject to actually control the rehabilitation exoskeleton through imagination, helping the subject to train and rehabilitate upper limb functions, and through the combination of precise EEG signal processing, stable data transmission and exoskeleton control, achieving efficient unilateral upper limb rehabilitation training, providing patients with more targeted rehabilitation plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. By reading the detailed description of the embodiments below, the advantages and benefits of the solutions will become clear to those skilled in the art. The drawings are only for the purpose of illustrating preferred embodiments and are not to be considered as limiting the present invention. In the drawings:

[0016] Figure 1 This is a flow chart of an upper limb active rehabilitation training method based on EEG-robot collaboration according to an embodiment of the present invention.

[0017] Figure 2 This is a control block diagram of an upper limb active rehabilitation training system based on EEG-robot collaboration according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and detailedly described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments in the embodiments of the present invention should fall within the scope of protection of the embodiments of the present invention.

[0019] Figure 1 A flowchart of an upper limb active rehabilitation training method based on EEG-robot collaboration is provided in an embodiment of the present invention. Figure 1 As shown, this embodiment mainly includes:

[0020] S101. Collect four-category motor imagery EEG data of the shoulder, elbow, wrist, and hand joints of the unilateral upper limb through an EEG signal acquisition system.

[0021] For example, the EEG signal acquisition system uses highly sensitive EEG electrodes, arranged according to the international 10-20 system electrode placement standard, at corresponding locations on the patient's scalp. This ensures accurate acquisition of EEG signals related to imagined movement of the shoulder, elbow, wrist, and hand joints of the unilateral upper limb. The acquisition equipment uses a professional EEG acquisition device with a sampling frequency of 500Hz to ensure sufficient temporal resolution of the acquired EEG signals.

[0022] Before data acquisition begins, the patient's scalp must be cleaned to remove grease and dirt to reduce skin resistance and improve signal acquisition quality. At the same time, the patient is fitted with a suitable electrode cap to ensure close contact between the electrodes and the scalp, and electrode gel is used to further enhance the conductivity between the electrodes and the scalp.

[0023] During the acquisition process, the patient is guided to perform four motor imagery tasks: extending the shoulder forward to a horizontal position and then lowering it to a vertical position; bending the elbow 90 degrees and then straightening it; internally rotating the wrist 90 degrees and then returning it to its original position; and grasping the hand and then extending it. To enable patients to better perform motor imagery, the specific requirements and sensations of each motor imagery task can be explained to the patient in detail before training, and video demonstrations can be used to assist the patient's understanding. Each motor imagery task lasts 3-5 seconds, with a 2-3 second rest interval between adjacent tasks to prevent patient fatigue from affecting signal quality. Each task is repeated 10-15 times to obtain sufficient data samples.

[0024] S102: Preprocess the EEG data.

[0025] S103 , integrating the SE (Squeeze-and-Excitation) module with EEGNet, and using the preprocessed EEG data to perform model training to obtain an EEG signal recognition model.

[0026] For example, the data processing and analysis module builds a neural network by integrating the SE module and EEGNet to realize EEG signal classification. During training, the processed EEG data is divided into a training set and a test set. The network parameters are repeatedly trained using an optimizer and a cross-entropy loss function as the target. Finally, the test data is used to test the network's recognition accuracy of the four types of motor imagery EEG signals.

[0027] The data processing and analysis module is the system's core processing unit, utilizing a combination of offline and online control methods. Offline, it collects the subject's motor imagery EEG data, deeply mines and analyzes this data using specific algorithms, and constructs and trains an EEG signal recognition model. In the online phase, leveraging a well-trained offline model, it rapidly analyzes and accurately judges the real-time EEG signals, decomposing the subject's motor intentions. Furthermore, this module supports the choice of active or passive rehabilitation modes. In active training mode, the subject uses the trained offline model to actively control the rehabilitation exoskeleton through self-imagery. Passive training mode relies on the intervention of external assistants, coordinated with the subject's motor imagery, to implement stimulation training.

[0028] S104: The EEG signal recognition model classifies and recognizes the EEG data collected in real time to obtain control instructions.

[0029] S105. The exoskeleton control module receives the control instruction and controls the exoskeleton with 7 degrees of freedom of the shoulder, elbow, wrist, and hand to drive the upper limbs to complete the rehabilitation movement.

[0030] For example, the exoskeleton module is designed using Solidworks modeling software and has freedom of movement at four joints of the upper limb: shoulder, elbow, wrist, and hand. After the design is completed, 3D printing technology is used to print and assemble the model. The exoskeleton module is designed to conform to the human upper limb structure and can perform rehabilitation movements such as arm raising, elbow flexion, wrist rotation, and fist clenching based on the subject's motor imagery. The shoulder joint is designed with three rotational structures, providing three degrees of freedom, enabling flexion and extension (swinging forward and backward, such as raising the arm forward), abduction and adduction (swinging left and right, such as raising the arm sideways), and internal and external rotation (rotation around the long axis of the arm, such as turning the palm upward). The elbow joint is designed as a rotational joint with one degree of freedom, enabling flexion and extension (folding the forearm and upper arm, such as bending the arm). The wrist joint is designed as a rotational joint with two degrees of freedom, enabling flexion and extension (bending the palm forward, such as moving the wrist inward toward the arm), and pronation and supination (rotation around the long axis of the forearm, such as turning the palm from upward to downward). The hand is designed as a biomimetic structure with one degree of freedom, enabling four-finger grasping and releasing movements. It receives control commands from the EEG signal processing module, converting EEG signals into mechanical power to drive the exoskeleton to perform the corresponding movements, thereby stimulating the damaged area of ​​the brain and promoting the recovery process.

[0031] Specifically, during the design process, the exoskeleton's strength, weight, and comfort were considered, and lightweight, high-strength materials were selected. Finite element analysis was used to analyze the mechanical properties of the design model and optimize the structural design to ensure the exoskeleton's safety and reliability during use.

[0032] After the design is complete, the model is printed using 3D printing technology. Based on the structural characteristics of the exoskeleton, the model is rationally divided and each component is printed separately.

[0033] After printing, the parts undergo post-processing, such as grinding and polishing, to remove surface burrs and imperfections. Finally, they are assembled according to the design requirements, installing components such as the drive motor, transmission mechanism, and sensors to complete the exoskeleton.

[0034] The present invention's active upper limb rehabilitation training method based on EEG-robot collaboration decodes the subject's motor imagination intentions and converts the decoding results into control instructions for the upper limb rehabilitation exoskeleton system, thereby enabling the subject to actually control the rehabilitation exoskeleton through imagination, helping the subject to train and rehabilitate upper limb functions, and through the combination of precise EEG signal processing, stable data transmission and exoskeleton control, achieving efficient unilateral upper limb rehabilitation training, providing patients with more targeted rehabilitation plans.

[0035] In another implementation of the present invention, the four-category motor imagery EEG data includes motor imagery EEG signals corresponding to the movement of extending the shoulder to the horizontal and then falling to the vertical, bending the elbow 90 degrees and then straightening it, internally rotating the wrist 90 degrees and then returning to the straight position, and grasping the hand and then stretching it.

[0036] For example, the collected original EEG signals contain various noises and artifacts, and thus need to be preprocessed.

[0037] First, basic filtering is performed, using a bandpass filter to remove low-frequency drift and high-frequency noise interference; at the same time, a notch filter is used to remove 50Hz power frequency interference.

[0038] Then, ICA artifact removal is performed, and the original EEG signal is decomposed into multiple independent components using the independent component analysis algorithm. By observing the waveform and spectral characteristics of each component, components related to artifacts such as electrooculography and electromyography are identified and removed.

[0039] During segmentation and baseline correction, the EEG signal is segmented according to the time window of the motor imagery task. The length of each segment is the duration of the motor imagery task plus a certain lead and delay time (e.g., 1 second lead, 1 second delay). For each signal segment, the signal 1 second before the motor imagery task begins is used as the baseline for baseline correction to remove DC offset and baseline drift.

[0040] Finally, manual review is conducted, and professionals check the pre-processed data one by one to ensure that the data quality meets the requirements, and mark or re-collect any abnormal data segments.

[0041] In another implementation of the present invention, the preprocessing includes: basic filtering, ICA artifact removal, segmentation, baseline correction, and manual inspection.

[0042] In another implementation of the present invention, the EEGNet module includes a temporal convolution layer, a depth-wise separable convolution layer, a first random dropout layer, a first average pooling layer, a separable convolution layer, a second average pooling layer, and a second random dropout layer, which are connected in sequence.

[0043] In another implementation of the present invention, the temporal convolution layer is used to extract the frequency and spatial features of the original EEG signal; the random inactivation layer is used to prevent network overfitting; and the average pooling layer is used to reduce the feature dimension.

[0044] In another implementation of the present invention, the SE module includes a global pooling layer, a first fully connected layer, a second fully connected layer, and a normalized exponential function layer.

[0045] In another implementation of the present invention, the global pooling layer is used to compress the channel to form a compressed vector; the fully connected layer is used for dimensionality reduction and dimensionality increase.

[0046] For example, the SE module aggregates feature information, compresses, excites and adds weights to time-frequency features. It first compresses the channel to form a compressed vector through global pooling, then reduces the dimension through the first fully connected layer, increases the dimension through the second fully connected layer and generates a weight vector through the Sigmoid activation function. Finally, the weight vector is multiplied element-by-element with the original feature map to assign weights to the original features, thereby obtaining valuable features and suppressing unimportant features.

[0047] After the above processing, all features are flattened and passed through the fully connected layer and the normalized exponential function (Softmax) layer in sequence to output the four classification results corresponding to the four types of movement imagination intentions: the shoulder extends forward to the horizontal and then falls to the vertical, the elbow bends 90 degrees and then straightens, the wrist rotates 90 degrees and then returns to the straight position, and the hand grasps and then stretches.

[0048] Another aspect of the present invention, as Figure 2 As shown, an upper limb active rehabilitation training system based on EEG-robot collaboration is provided, comprising:

[0049] EEG signal acquisition system: used to collect four-category motor imagery EEG data of the shoulder, elbow, wrist, and hand joints of the unilateral upper limb.

[0050] Data processing and analysis module: used to preprocess the EEG data; fuse the SE module with EEGNet, use the preprocessed EEG data to perform model training, and obtain an EEG signal recognition model; the EEG signal recognition model classifies and recognizes the real-time collected EEG data to obtain control instructions.

[0051] Exoskeleton control module: used to receive control instructions and control the exoskeleton with 7 degrees of freedom of the shoulder, elbow, wrist and hand to drive the upper limbs to complete rehabilitation movements.

[0052] For example, the data communication module uses the TCP / IP protocol to connect the data processing and analysis module to the exoskeleton control module. The data processing and analysis module is responsible for monitoring and receiving connection requests from the exoskeleton control module (client). After receiving and processing the EEG data, the data processing and analysis module sends instructions to the exoskeleton control module.

[0053] During the system integration process, the EEG signal acquisition system, data processing and analysis module, and exoskeleton control module were jointly debugged. First, a hardware connection test was conducted to ensure proper electrical connectivity between modules. Next, a software communication test was conducted to verify accurate data transmission and reception. Finally, a system functionality test was conducted, with patients performing actual motor imagery exercises to observe whether the exoskeleton could accurately perform rehabilitation movements according to the patient's motor imagery. Based on the test results, the system was optimized and adjusted to ensure stable and efficient operation of the entire system.

[0054] The specific implementation process of the system is as follows: First, the EEG signal acquisition system is correctly worn on the subject's head, and the exoskeleton is worn on their upper limbs; then, the EEG signal processing module trains the offline motor imagery EEG signals to generate a reliable recognition model, and then analyzes the EEG data collected online in real time, converts it into control instructions, and selects active or passive rehabilitation mode according to the subject's needs; finally, the exoskeleton module receives the control instructions and executes the corresponding rehabilitation movements to stimulate the damaged area of ​​the brain and assist in rehabilitation.

[0055] In another aspect of the present invention, an electronic device includes a processor, a memory, a communication bus, and a communication interface.

[0056] in:

[0057] The processor, memory and communication interface communicate with each other through a communication bus.

[0058] Communication interface, used to communicate with other electronic devices or servers.

[0059] The processor is used to execute the program, and specifically can execute the steps of any one of the upper limb active rehabilitation training methods based on EEG-robot collaboration in the above embodiments.

[0060] Specifically, the program may include program codes including computer operation instructions.

[0061] The processor may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the smart device may be processors of the same type, such as one or more CPUs; or different types of processors, such as one or more CPUs and one or more ASICs.

[0062] Memory is used to store programs. The memory may include high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk storage.

[0063] The program can be specifically used to enable the processor to execute the steps of any one of the upper limb active rehabilitation training methods based on EEG-robot collaboration described in the embodiments. The specific implementation of each step in the program can refer to the corresponding descriptions in the steps and units executed in any one of the above steps of the upper limb active rehabilitation training method based on EEG-robot collaboration, which will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described equipment and modules can refer to the corresponding process description in the aforementioned method embodiment.

[0064] The exemplary embodiments of the present application further provide a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the methods of the various embodiments of the present application.

[0065] The method according to the embodiment of the present invention described above can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk or magneto-optical disk), or as computer code that is originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded via a network and will be stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor or programmable or dedicated hardware (such as an ASIC or FPGA). It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, a processor or hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown here, the execution of the code converts the general-purpose computer into a dedicated computer for executing the method shown here.

[0066] Thus far, specific embodiments of the present invention have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve the desired results. Additionally, the processes depicted in the accompanying drawings do not necessarily require the specific order shown, or sequential order, to achieve the desired results.

[0067] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, back, etc.) are only used to explain the relative position relationship between the components in a certain specific order (as shown in the accompanying drawings). If the specific order changes, the directional indication will also change accordingly.

[0068] In the description of the present invention, the terms "first" and "second" are used solely to facilitate description of different components or names and should not be construed as indicating or implying a sequential relationship, relative importance, or implicitly specifying the quantity of the technical features being described. Therefore, features specified as "first" or "second" may explicitly or implicitly include at least one of such features.

[0069] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0070] It should be noted that although the specific embodiments of the present invention are described in detail in conjunction with the accompanying drawings, this should not be construed as limiting the scope of protection of the present invention. Within the scope described by the claims, various modifications and variations that can be made by those skilled in the art without creative effort still fall within the scope of protection of the present invention.

[0071] The examples of the embodiments of the present invention are intended to briefly illustrate the technical features of the embodiments of the present invention so that those skilled in the art can intuitively understand the technical features of the embodiments of the present invention, and are not intended to improperly limit the embodiments of the present invention.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An upper limb active rehabilitation training method based on EEG-robot collaboration, characterized in that: include: The EEG data of four categories of motor imagery of the shoulder, elbow, wrist and hand joints of the unilateral upper limb were collected through the EEG signal acquisition system; Preprocessing the EEG data; The SE module is integrated with EEGNet, and the pre-processed EEG data is used for model training to obtain an EEG signal recognition model; The EEG signal recognition model classifies and identifies the real-time collected EEG data to obtain control instructions; The exoskeleton control module receives control instructions and controls the exoskeleton with 7 degrees of freedom of the shoulder, elbow, wrist, and hand, driving the upper limbs to complete rehabilitation movements.

2. The method according to claim 1, characterized in that The four-category motor imagery EEG data includes motor imagery EEG signals corresponding to the movement of extending the shoulder to a horizontal position and then dropping it to a vertical position, bending the elbow 90 degrees and then straightening it, internally rotating the wrist 90 degrees and then returning to the normal position, and grasping the hand and then stretching it.

3. The method according to claim 1, characterized in that The preprocessing includes: basic filtering, ICA artifact removal, segmentation, baseline correction, and manual inspection.

4. The method according to claim 1, wherein The EEGNet module includes a temporal convolution layer, a depth-wise separable convolution layer, a first random dropout layer, a first average pooling layer, a separable convolution layer, a second average pooling layer, and a second random dropout layer, which are connected in sequence.

5. The method according to claim 4, characterized in that The temporal convolution layer is used to extract the frequency and spatial features of the original EEG signal; The random dropout layer is used to prevent the network from overfitting; The average pooling layer is used to reduce the feature dimension.

6. The method according to claim 1, characterized in that The SE module includes a global pooling layer, a first fully connected layer, a second fully connected layer, and a normalized exponential function layer.

7. The method according to claim 6, characterized in that The global pooling layer is used to compress the channel to form a compressed vector; The fully connected layer is used for dimensionality reduction and dimensionality increase.

8. An upper limb active rehabilitation training system based on EEG-robot collaboration, characterized in that: include: EEG signal acquisition system: used to collect four-category motor imagery EEG data of the shoulder, elbow, wrist, and hand joints of the unilateral upper limb; Data processing and analysis module: used for preprocessing the EEG data; The SE module is integrated with EEGNet, and the pre-processed EEG data is used for model training to obtain an EEG signal recognition model; The EEG signal recognition model classifies and identifies the real-time collected EEG data to obtain control instructions; Exoskeleton control module: used to receive control instructions and control the exoskeleton with 7 degrees of freedom of the shoulder, elbow, wrist and hand to drive the upper limbs to complete rehabilitation movements.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the upper limb active rehabilitation training method based on EEG-robot collaboration as described in any one of claims 1 to 7 are implemented.

10. A computer storage medium, characterized in that The computer storage medium stores a computer program, which, when executed by a processor, implements the steps of the upper limb active rehabilitation training method based on EEG-robot collaboration as described in any one of claims 1 to 7.

Images