CN114767463A

Movatterモバイル変換

Info

Publication number: CN114767463A
Application number: CN202210241198.7A
Authority: CN
Inventors: 张迪; 周志勇; 康威
Original assignee: Shanghai Dianji University
Current assignee: Shanghai Dianji University
Priority date: 2022-03-11
Filing date: 2022-03-11
Publication date: 2022-07-22

Abstract

Translated fromChinese

本发明公开了一种基于表面肌电的意识控制运动康复系统及其方法，解决了传统的康复治疗往往借助人工辅助或简单的康复设备来实现，难以保证训练强度及精度，不利于提高康复疗效的问题，其技术方案要点是通过肌电采集模块采集表面肌电信号、康复机器人进行辅助康复和关节角度检测反馈、运动意图识别模块进行运动轨迹记录，并通过运动交互控制模块根据检测反馈数据控制康复机器人辅助运动，本发明的一种基于表面肌电的意识控制运动康复系统及其方法，能实现对康复机器人的控制，可实时准确进行康复辅助运动，有效减轻对人工康复的依赖度。

The invention discloses a consciousness-controlled exercise rehabilitation system based on surface electromyography and a method thereof, which solves the problem that traditional rehabilitation therapy is often realized by manual assistance or simple rehabilitation equipment, and it is difficult to ensure training intensity and accuracy, which is not conducive to improving the rehabilitation effect. The main point of the technical solution is to collect surface EMG signals through the EMG acquisition module, the rehabilitation robot to perform assisted rehabilitation and joint angle detection feedback, the motion intention recognition module to record the motion trajectory, and the motion interaction control module to control according to the detection feedback data. Rehabilitation robot assisted motion, a conscious control motion rehabilitation system and method based on surface electromyography of the present invention, can realize control of the rehabilitation robot, can accurately perform rehabilitation assisted motion in real time, and effectively reduce the dependence on manual rehabilitation.

Description

Translated fromChinese

一种基于表面肌电的意识控制运动康复系统及其方法A consciousness-controlled motor rehabilitation system and method based on surface electromyography

技术领域technical field

本发明涉及运动康复系统，特别涉及一种基于表面肌电的意识控制运动康复系统及其方法。The invention relates to a sports rehabilitation system, in particular to a consciousness-controlled sports rehabilitation system based on surface electromyography and a method thereof.

背景技术Background technique

目前导致中枢神经系统损伤的主要原因包括脑卒中和脊髓损伤，它们通常会造成患者偏瘫或截瘫，进而引发内脏器官功能障碍以及一系列的并发症。大量临床研究表明，通过康复训练可使中枢神经系统实现一定程度的结构重组或功能代偿。目前临床康复机器人的康复方式单一，以被动康复训练为主，康复机器人通过预设的行为模式，带动患者进行康复训练。传统的康复治疗往往借助人工辅助或简单的康复设备来实现，难以保证训练强度及精度，不利于提高康复疗效。现有这种被动康复训练只适用于软瘫期的患者，且训练时间长会降低患者训练的积极性，甚至造成二次伤害。同时，这种方法也不能根据不同患者的不同病情具有个体适应性，人机交互性差，康复效果具有局限性。At present, the main causes of central nervous system damage include stroke and spinal cord injury, which usually cause hemiplegia or paraplegia in patients, which in turn leads to dysfunction of internal organs and a series of complications. A large number of clinical studies have shown that the central nervous system can achieve a certain degree of structural reorganization or functional compensation through rehabilitation training. At present, the rehabilitation methods of clinical rehabilitation robots are single, mainly passive rehabilitation training, and the rehabilitation robot drives patients to carry out rehabilitation training through preset behavior patterns. Traditional rehabilitation therapy is often realized with artificial assistance or simple rehabilitation equipment, which is difficult to ensure the training intensity and accuracy, which is not conducive to improving the rehabilitation effect. The existing passive rehabilitation training is only suitable for patients with flaccid paralysis, and long training time will reduce the enthusiasm of patients for training, and even cause secondary injury. At the same time, this method cannot have individual adaptability according to the different conditions of different patients, the interaction between human and machine is poor, and the rehabilitation effect is limited.

发明内容SUMMARY OF THE INVENTION

本发明的目的是提供一种基于表面肌电的意识控制运动康复系统及其方法，能实现对康复机器人的控制，可实时准确进行康复辅助运动，有效减轻对人工康复的依赖度。The purpose of the present invention is to provide a consciously controlled exercise rehabilitation system and method based on surface electromyography, which can realize the control of the rehabilitation robot, can accurately perform rehabilitation auxiliary exercise in real time, and effectively reduce the dependence on artificial rehabilitation.

本发明的上述技术目的是通过以下技术方案得以实现的：The above-mentioned technical purpose of the present invention is achieved through the following technical solutions:

一种基于表面肌电的意识控制运动康复系统及其方法，包括有：A consciously controlled movement rehabilitation system and method based on surface electromyography, comprising:

肌电采集模块，设置有可穿戴生物传感器，用于采集获取被测者下肢的表面肌电信号；The EMG acquisition module is provided with a wearable biosensor, which is used to acquire the surface EMG signals of the lower limbs of the subject;

康复机器人，供被测者装配，记录被测者的设定关节角度，辅助被测者进行康复运动；Rehabilitation robot, for the test subject to assemble, record the test subject's set joint angle, and assist the test subject to perform rehabilitation exercises;

运动意图识别模块，接收表面肌电信号及关节角度，构建下肢骨骼运动学模型以记录运动轨迹；The motion intention recognition module receives surface EMG signals and joint angles, and builds a lower extremity skeletal kinematics model to record motion trajectories;

运动交互控制模块，获取采集的表面肌电信号，并记录同时序下的运动轨迹，并基于改进归零神经网络模型的人工智能算法建立运动轨迹-表面肌电的交互映射；The motion interactive control module obtains the collected surface EMG signals, records the motion trajectory under the same sequence, and establishes the motion trajectory-surface EMG interactive mapping based on the artificial intelligence algorithm of the improved zeroing neural network model;

采集患者实时表面肌电信号，通过交互映射获取跟踪期望轨迹，运动交互控制模块输出控制信号，康复机器人响应于控制信号以进行辅助康复运动。The real-time surface EMG signal of the patient is collected, and the desired trajectory is obtained and tracked through interactive mapping. The motion interaction control module outputs a control signal, and the rehabilitation robot responds to the control signal to perform auxiliary rehabilitation motion.

一种基于表面肌电的意识控制运动康复方法，包括有以下步骤：A consciously controlled movement rehabilitation method based on surface electromyography includes the following steps:

采集患者的表面肌电信号数据，并采集获取患者的设定关节数据，建立下肢骨骼运动学模型；Collect the surface electromyographic signal data of the patient, and collect and obtain the set joint data of the patient, and establish a lower extremity skeletal kinematics model;

基于改进归零神经网络模型的人工智能算法，建立运动轨迹-表面肌电的交互映射；Based on the artificial intelligence algorithm of the improved zeroing neural network model, the interactive mapping between the motion trajectory and the surface EMG is established;

可穿戴生物传感器对患者进行检测获取对应表面肌电信号数据，并通过康复机器人获取患者实时同步的关节角度数据；The wearable biosensor detects the patient to obtain the corresponding surface EMG signal data, and obtains the patient's real-time synchronized joint angle data through the rehabilitation robot;

基于采集的肌电信号数据通过交互映射获取对应的期望轨迹；Based on the collected EMG signal data, the corresponding desired trajectory is obtained through interactive mapping;

康复机器人基于期望轨迹进行辅助运动控制。The rehabilitation robot performs auxiliary motion control based on the desired trajectory.

综上所述，本发明具有以下有益效果：To sum up, the present invention has the following beneficial effects:

通过对患者的肌电信号数据及关节角度数据进行采集，并建立下肢骨骼模型，通过迭代学习控制器的设计，能建立运动轨迹和表面肌电的交互映射，进而能通过采集的肌电信号对运动轨迹进行预测获取期望轨迹，通过控制器输出对应的控制信号可实现对康复机器人的控制，进而可实现康复机器人能在康复训练过程中实时准确的进行康复辅助运动，能有效减轻对人工康复的依赖度。By collecting the EMG signal data and joint angle data of the patient, and establishing the lower limb skeleton model, through the design of the iterative learning controller, the interactive mapping between the motion trajectory and the surface EMG can be established, and then the collected EMG signals can be used for The motion trajectory is predicted to obtain the desired trajectory, and the control of the rehabilitation robot can be realized by outputting the corresponding control signal from the controller, so that the rehabilitation robot can perform the rehabilitation-assisted motion in real time and accurately during the rehabilitation training process, which can effectively reduce the need for manual rehabilitation. Dependency.

附图说明Description of drawings

图1为本系统的模块构成示意图。Figure 1 is a schematic diagram of the module structure of the system.

具体实施方式Detailed ways

以下结合附图对本发明作进一步详细说明。The present invention will be further described in detail below with reference to the accompanying drawings.

根据一个或多个实施例，公开了一种基于表面肌电的意识控制运动康复系统，如图1所示，包括有肌电采集模块、康复机器人、运动意图识别模块及运动交互控制模块。According to one or more embodiments, a consciousness-controlled exercise rehabilitation system based on surface electromyography is disclosed, as shown in FIG. 1 , including an electromyography acquisition module, a rehabilitation robot, a movement intention recognition module, and a movement interaction control module.

肌电采集模块设置有可穿戴生物传感器，用于供被测者穿戴以进行设定的表面肌电信号的采集。肌电信号根据肌肉位置分别进行采集，肌肉力对关节的转动效应是通过当量力臂反映，通过比较髋关节、膝关节和踝关节对应肌肉的额状轴当量力臂和真实力臂的大小，最终筛选出8种肌肉，分别为股直肌、股外侧肌、胫骨前肌、股内侧肌、股二头肌、外侧腓肠肌、半腱肌和内侧腓肠肌，采集这八种肌肉的肌电信号。通过可穿戴式生物传感器设备采集被测者下肢的表面肌电信息，为后续意图识别算法和控制系统提供数据支持。同时，采集患者的表面肌电数据对比正常数据可以准确获取患者的肌肉收缩状态，从而客观评估患者的运动状态和康复情况，为智能化交互控制系统提供基础。The electromyography acquisition module is provided with a wearable biosensor, which is used for the subject to wear to collect the set surface electromyography signal. The EMG signals are collected according to the position of the muscle, and the rotational effect of the muscle force on the joint is reflected by the equivalent moment arm. Finally, 8 kinds of muscles were screened, namely rectus femoris, vastus lateralis, tibialis anterior, vastus medialis, biceps femoris, lateral gastrocnemius, semitendinosus and medial gastrocnemius, and the EMG signals of these eight muscles were collected. The surface EMG information of the lower limbs of the subject is collected through the wearable biosensor device, which provides data support for the subsequent intention recognition algorithm and control system. At the same time, collecting the patient's surface EMG data and comparing the normal data can accurately obtain the patient's muscle contraction state, so as to objectively evaluate the patient's exercise state and rehabilitation situation, and provide the basis for the intelligent interactive control system.

肌电采集模块的整体电路中包括有检测电极、放大电路、带通滤波电路、陷波电路、MCU模块、WIFI模块，以对采集的肌电信号进行处理。The overall circuit of the EMG acquisition module includes detection electrodes, amplifier circuits, band-pass filter circuits, notch circuits, MCU modules, and WIFI modules to process the collected EMG signals.

康复机器人用于装配在被测患者身上，康复机器人对应的记录被测者设定关节的角度，记录的设定关节为膝关节和髋关节，同时康复机器人响应于运动交互控制模块对被测者进行运动辅助。The rehabilitation robot is used to be assembled on the patient under test. The rehabilitation robot correspondingly records the angle of the joint set by the test subject, and the recorded set joints are the knee joint and the hip joint. At the same time, the rehabilitation robot responds to the motion interaction control module to the test subject. Exercise assistance.

运动意图识别模块接收采集的表面肌电信号及关节角度相关数据，用于构建下肢骨骼运动学模型。人体下肢运动，主要依靠矢状面的关节驱动，因此以下肢关节为节点建立人体下肢骨骼动力学模型。通过采集的数据对模型进行训练建立，通过正常肢体和运动障碍肢体实验，验证和改进算法，为人体下肢主动运动意图识别奠定算法框架，使得下肢骨骼运动学模型能跟随被测者的运动，下肢骨骼运动学模型以膝关节和髋关节节点构建，用以描绘运动轨迹。在一个完整康复周期内，建立一类具有非线性、强耦合、不确定、时变等特性的动力学模型，为后续识别和控制器的研究奠定基础。The motion intention recognition module receives the collected surface EMG signals and joint angle-related data, and is used to construct the lower limb skeletal kinematics model. The motion of the lower limbs of the human body is mainly driven by the joints in the sagittal plane, so the lower limb joints are used as nodes to establish the skeletal dynamics model of the lower limbs of the human body. The model is trained and established through the collected data, and the algorithm is verified and improved through experiments on normal limbs and limbs with movement disorders, and an algorithm framework is established for the recognition of the active motion intention of the lower limbs of the human body, so that the lower limb skeletal kinematics model can follow the movement of the subject and the lower limbs. The skeletal kinematics model is constructed with knee joint and hip joint nodes to describe the motion trajectory. In a complete recovery cycle, a class of dynamic models with nonlinear, strong coupling, uncertainty, and time-varying characteristics is established to lay the foundation for subsequent identification and controller research.

运动交互控制模块获取采集的肌电信号数据及关节角度数据，并根据下肢骨骼运动学模型记录的同时序下的运动轨迹，基于改进归零神经网络模型的人工智能算法建立运动轨迹-表面肌电的交互映射。The motion interaction control module obtains the collected EMG signal data and joint angle data, and establishes the motion trajectory-surface EMG based on the motion trajectory under the same sequence recorded by the lower limb skeletal kinematics model, and the artificial intelligence algorithm based on the improved zeroing neural network model. interactive mapping.

采集被测者的实时表面肌电信号，通过建立的交互映射，可对应的获得期望轨迹，运动交互控制模块根据期望轨迹输出对应的控制信号，康复机器人接收并响应于控制信号，参与期望轨迹的辅助动作，建立的交互映射结合迭代学习控制器实现闭环系统稳定性，运动交互控制模块可使患者的关节康复轨迹能快速地跟踪期望轨迹，跟踪误差迅速地收敛到零，以完成对被测者的辅助康复运动，主动训练与被动训练结合，增加康复的多样化。The real-time surface EMG signal of the subject is collected, and the desired trajectory can be correspondingly obtained through the established interactive mapping. The motion interactive control module outputs the corresponding control signal according to the desired trajectory. The rehabilitation robot receives and responds to the control signal and participates in the desired trajectory. Auxiliary action, the established interactive map combined with iterative learning controller realizes the stability of the closed-loop system, and the motion interactive control module enables the patient's joint rehabilitation trajectory to quickly track the desired trajectory, and the tracking error quickly converges to zero, so as to complete the detection of the subject. The combination of active training and passive training increases the diversity of rehabilitation.

根据一个或多个实施例，公开了一种基于表面肌电的意识控制运动康复方法，包括有以下步骤：According to one or more embodiments, a surface electromyography-based consciously controlled exercise rehabilitation method is disclosed, comprising the following steps:

通过可穿戴生物传感器采集患者的表面肌电信号数据，通过康复机器人获取患者关节角度数据，下肢骨骼运动学模型基于表面肌电信号数据及关节角度数据进行训练建立，以记录实时运动轨迹。The patient's surface EMG signal data is collected through wearable biosensors, and the patient's joint angle data is obtained through the rehabilitation robot. The lower limb skeletal kinematics model is trained and established based on the surface EMG signal data and joint angle data to record real-time motion trajectories.

通过可穿戴生物传感器采集的表面肌电信号数据，Surface EMG signal data collected by wearable biosensors,

对于下肢康复训练，恢复患者下肢力量有极为重要的意义，在这个过程中，需要评估患者的运动能力，并使患者主动参与康复训练。因此，本发明的目的是：设计能够根据患者运动能力并根据患者的病情程度和运动状态自适应调整的上肢康复机器人控制策略；通过肌电信号和机械运动康复进行生物反馈结合，并对其康复过程进行多参数智能化分析评估，满足对患者个体差异的适应性，减轻对人工康复的依赖度。For lower extremity rehabilitation training, it is of great significance to restore the strength of the lower limbs of patients. In this process, it is necessary to evaluate the patient's exercise ability and make the patient actively participate in the rehabilitation training. Therefore, the purpose of the present invention is to: design an upper limb rehabilitation robot control strategy that can be adaptively adjusted according to the patient's exercise ability and according to the patient's disease degree and exercise state; combine biofeedback through myoelectric signal and mechanical exercise rehabilitation, and restore its rehabilitation Multi-parameter intelligent analysis and evaluation are carried out in the process to meet the adaptability to individual differences of patients and reduce the dependence on artificial rehabilitation.

本具体实施例仅仅是对本发明的解释，其并不是对本发明的限制，本领域技术人员在阅读完本说明书后可以根据需要对本实施例做出没有创造性贡献的修改，但只要在本发明的权利要求范围内都受到专利法的保护。This specific embodiment is only an explanation of the present invention, and it does not limit the present invention. Those skilled in the art can make modifications without creative contribution to the present embodiment as required after reading this specification, but as long as the rights of the present invention are used All claims are protected by patent law.

Claims

1. A consciousness control exercise rehabilitation system based on surface myoelectricity is characterized in that: comprises that

The myoelectricity acquisition module is provided with a wearable biosensor and is used for acquiring surface myoelectricity signals of the lower limbs of the detected person;

the rehabilitation robot is assembled by the testee, records the set joint angle of the testee and assists the testee to perform rehabilitation movement;

the movement intention identification module is used for receiving the surface electromyographic signals and the joint angles and constructing a lower limb skeleton kinematics model to record a movement track;

the motion interaction control module is used for acquiring the acquired surface electromyogram signals, recording motion tracks in the same sequence, and establishing motion track-surface electromyogram interaction mapping based on an artificial intelligence algorithm of an improved return-to-zero neural network model;

the real-time surface electromyographic signals of the patient are collected, the expected tracking track is obtained through interactive mapping, the motion interactive control module outputs control signals, and the rehabilitation robot responds to the control signals to perform auxiliary rehabilitation motion.

2. The surface myoelectricity-based consciousness-controlled motor rehabilitation system according to claim 1, wherein: the surface electromyographic signals collected by the electromyographic collecting module comprise electromyographic signals of rectus femoris, vastus lateralis, tibialis anterior, vastus medialis, biceps femoris, lateral gastrocnemius, semitendinosus and medial gastrocnemius.

3. The surface myoelectricity-based consciousness-controlled motor rehabilitation system according to claim 1, wherein: the recorded set joints are knee joints and hip joints; the lower limb skeleton kinematics model is constructed by knee joint and hip joint nodes to describe a motion trail.

4. The surface myoelectricity-based consciousness-controlled motor rehabilitation system according to claim 1, wherein: the motion interaction control module is based on an iterative learning controller, and iterative training is carried out through the collected surface electromyogram signals and the motion trail recorded in the same time sequence to obtain the motion trail-surface electromyogram interaction mapping.

5. A consciousness controlled exercise rehabilitation method based on surface myoelectricity is characterized by comprising the following steps:

collecting surface electromyogram signal data of a patient, collecting and acquiring set joint data of the patient, and establishing a lower limb skeleton kinematics model;

establishing a motion trail-surface myoelectric interactive mapping based on an artificial intelligence algorithm of an improved return-to-zero neural network model;

the wearable biosensor detects a patient to obtain corresponding surface electromyogram signal data, and obtains real-time synchronous joint angle data of the patient through the rehabilitation robot;

acquiring a corresponding expected track through interactive mapping based on the acquired electromyographic signal data;

the rehabilitation robot performs auxiliary motion control based on the desired trajectory.

6. The surface myoelectricity-based consciousness-controlled motor rehabilitation method according to claim 5, wherein: the wearable biological sensor is used for collecting surface electromyographic signal data of a patient, the rehabilitation robot is used for obtaining joint angle data of the patient, and the lower limb skeleton kinematics model is trained and established based on the surface electromyographic signal data and the joint angle data so as to record a real-time motion track.