Bionic limb action control system and method thereof

Abstract

The invention discloses a bionic limb movement control system and method, and belongs to the technical field of bionic systems.The bionic limb movement control system comprises a central processing unit, a motor controller and a data collector, and the motor controller is electrically connected with a first motor and a second motor; the first motor is used for controlling bending of the bionic fingers. The second motor is used for controlling the bionic fingers to swing. The data collector is electrically connected with a data sensing assembly; the data sensing assembly is used for collecting abnormal myoelectricity data of a severed finger patient and image data of gestures of the severed finger patient and sending the collected data to the data collector, and the data collector sends the collected data to the central processing unit. In this way, the bionic finger can quickly adapt to the basic gesture action of the severed finger patient.

Description

Technical Field

[0001] The present invention relates to the technical field of bionic systems, and in particular to a bionic limb motion control system and method thereof. Background Art

[0002] For patients with amputated fingers (one finger missing), when they perform basic gestures such as making a fist or extending their palm, their bionic fingers need to operate according to the patient's electromyographic signals, so they need to be customized.

[0003] Currently, bionic fingers have low control accuracy for complex movements and require high customization costs, while basic gestures can meet the daily needs of patients with amputated fingers.

[0004] For example, among many existing technologies, Chinese patent application CN108703824B discloses a bionic hand control system and control method based on an electromyographic bracelet, which uses an acquisition module to collect the user's electromyographic signals and the direction, speed, acceleration and other information of the built-in gyroscope, and sends the collected information to the control module wirelessly; the control module is used to receive the signals sent by the acquisition module and process them to extract characteristic values.

[0005] However, this patent application requires collecting and analyzing the user's electromyographic signals, which requires a large amount of calculation and makes it impossible to quickly adapt to the user's control needs for basic gesture bionic fingers.

[0006] Based on this, the present invention designs a bionic limb motion control system and method thereof to solve the above problems. Summary of the Invention

[0007] In view of the above-mentioned shortcomings of the prior art, the present invention provides a bionic limb motion control system and method thereof.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A bionic limb motion control system includes a central processing unit, a motor controller and a data collector, wherein the motor controller is electrically connected to a first motor and a second motor;

[0010] The motor 1 is used to control the bending of the bionic finger;

[0011] The second motor is used to control the swing of the bionic finger;

[0012] The data collector is electrically connected to a data sensing component;

[0013] The data sensing component is used to collect abnormal electromyographic data of the finger amputation patient and image data of the finger amputation patient's gesture, and sends the collected data to the data collector, which sends the data to the central processing unit;

[0014] The central processing unit is electrically connected to a storage unit for storing data collected by the data sensing component;

[0015] The storage unit is also used to store normal myoelectric data of normal human hand gestures and bionic finger control signals corresponding to the normal myoelectric data;

[0016] Also included is a data processing unit, said data processing unit being electrically connected to the central processing unit;

[0017] The data processing unit is used to read the data in the storage unit through the central processing unit;

[0018] The data processing unit reads the image data of the current finger amputee's gesture, identifies the type of the finger amputee's gesture and matches it with the corresponding normal person's gesture electromyographic data;

[0019] At the same time, the data processing unit reads the electromyographic data of the finger amputation patient's gesture;

[0020] The data processing unit compares the abnormal electromyographic data of the finger amputee with the normal electromyographic data of a normal person and calculates the error value of the two data;

[0021] The data processing unit identifies the size of the error value and the set value;

[0022] If the error value is less than or equal to the set value, it indicates that the error between the two data is small. At this time, the data processing unit directly binds the abnormal myoelectric data of the finger amputation patient with the gesture with the normal myoelectric data.

[0023] If the error value is greater than the set value, it indicates that the error between the two signals is large. At this time, the data processing unit collects the abnormal electromyographic data multiple times and takes the average electromyographic signal after averaging it and binds it to the normal electromyographic data;

[0024] The central processing unit is used for reading the abnormal myoelectric data sent by the data collector and identifying the normal myoelectric data bound thereto when the data processing unit completes the binding of the myoelectric signals;

[0025] If there is normal myoelectric data bound to the abnormal myoelectric data, the central processing unit sends the bionic finger control signal corresponding to the normal myoelectric data to the motor controller;

[0026] If there is no normal myoelectric data bound to the abnormal myoelectric data, the central processing unit analyzes the short-term feature V of the abnormal myoelectric signal through a sliding window algorithm, matches the average myoelectric signal related to the short-term feature V, and then identifies the normal myoelectric data bound to it. The central processing unit sends the bionic finger control signal corresponding to the normal myoelectric data to the motor controller;

[0027] The motor controller is used to control the operation of motor 1 and motor 2 according to the bionic finger control signal.

[0028] Furthermore, the collected gestures of the finger amputation patient include clenching a fist, extending the palm, and an OK gesture.

[0029] Furthermore, the data sensing component includes a visual sensor and an electromyographic sensor;

[0030] The visual sensor and the electromyographic sensor are both electrically connected to the data collector;

[0031] The visual sensor is used to collect image data of the hand gestures of the patient with amputated fingers;

[0032] The electromyographic sensor is used to collect abnormal electromyographic data of patients with amputated fingers;

[0033] The visual sensor and electromyographic sensor send the collected data to the data collector.

[0034] Furthermore, the data processing unit includes a gesture data processing unit and a visual data processing unit;

[0035] The gesture data processing unit and the visual data processing unit are both electrically connected to the central processing unit.

[0036] Furthermore, the gesture data processing unit is used to read the image data of the gesture of the finger amputation patient stored in the storage unit through the central processing unit. The gesture data processing unit identifies the gesture type through the image recognition algorithm, and sends the gesture type to the central processing unit, which then sends it to the visual data processing unit.

[0037] Furthermore, the visual data processing unit is used to read the normal electromyographic data corresponding to the gesture stored in the storage unit through the central processing unit according to the gesture type, and at the same time the visual data processing unit reads the abnormal electromyographic data corresponding to the gesture, and the visual data processing unit calculates the error between the normal electromyographic data and the abnormal electromyographic data through the mean square error MSE algorithm, and adopts the following calculation formula:

[0038]

[0039] x represents normal EMG data, y represents abnormal EMG data, n represents the number of sampling points, i represents the sample point, and (x(i)-y(i)) represents the error of each sample point.

[0040] Furthermore, the visual data processing unit compares the calculated MSE value with the set value K;

[0041] When MSE is less than or equal to K, the error between the two surface electromyographic data is small, and the visual data processing unit directly binds the abnormal electromyographic data with the normal electromyographic data.

[0042] When MSE is greater than K, the error between the two surface electromyographic data is large. At this time, the visual data processing unit collects multiple groups of y and calculates the average signal, and uses the following calculation formula:

[0043]

[0044] Among them, v represents the number of y signal groups, N represents the number of y signal groups, A j represents the average electromyographic data of the y signal at the jth time point, y vj Represents the value of the v-th group y signal at the j-th time point.

[0045] Furthermore, the short-term feature V is a standard deviation feature.

[0046] Furthermore, the CPU calculates V and A j The relative standard deviation CV; and the formula is:

[0047] CV=V1 / A j *100%;

[0048] When the CV value is greater than 20%, the abnormal EMG signal is correlated with the average EMG signal;

[0049] When the CV value is less than 20%, the abnormal myoelectric signal is not matched.

[0050] In order to better achieve the purpose of the present invention, the present invention also provides a bionic limb motion control method, comprising the following steps:

[0051] Step 1: The visual sensor collects image data of the finger amputation patient's gestures, and the electromyography sensor collects abnormal electromyography data of the finger amputation patient;

[0052] Step 2: The central processing unit reads the image data of the hand gesture of the finger amputee stored in the storage unit, and the hand gesture data processing unit identifies the hand gesture type through the image recognition algorithm;

[0053] Step 3: The visual data processing unit reads the normal electromyographic data corresponding to the gesture stored in the storage unit through the central processing unit according to the gesture type. At the same time, the visual data processing unit reads the abnormal electromyographic data corresponding to the gesture. The visual data processing unit calculates the error between the normal electromyographic data and the abnormal electromyographic data through the mean square error (MSE) algorithm and uses the following calculation formula:

[0054]

[0055] x represents normal EMG data, y represents abnormal EMG data, n represents the number of sampling points, i represents the sample point, and (x(i)-y(i)) represents the error of each sample point;

[0056] Step 4: The visual data processing unit compares the calculated MSE value with the set value K;

[0057] When MSE is less than or equal to K, the error between the two surface electromyographic data is small, and the visual data processing unit directly binds the abnormal electromyographic data with the normal electromyographic data.

[0058] When MSE is greater than K, the error between the two surface electromyographic data is large. At this time, the visual data processing unit collects multiple groups of y and calculates the average signal, and uses the following calculation formula:

[0059]

[0060] Among them, v represents the number of y signal groups, N represents the number of y signal groups, A j represents the average electromyographic data of the y signal at the jth time point, y vj represents the value of the y signal of the vth group at the jth time point;

[0061] Step 5: The CPU calculates V and A j The relative standard deviation CV; and the formula is:

[0062] CV=V1 / A j *100%;

[0063] When the CV value is greater than 20%, the abnormal EMG signal is correlated with the average EMG signal;

[0064] When the CV value is less than 20%, no gesture is matched to the abnormal myoelectric signal.

[0065] Compared with the prior art, the present invention has the following beneficial effects: 1. When the present invention is used, the data sensing component collects the gesture data of the finger amputee and identifies the gesture type, and the visual data processing unit uses the image recognition algorithm to compare the image of the normal finger of the finger amputee with the standard gesture image to identify the gesture type;

[0066] By binding abnormal electromyographic signals with normal electromyographic signals through data sensing components, it is possible to directly use the bionic finger control signals of normal electromyographic signals as the control signals of motors 1 and 2 at the bionic finger control ends of patients with amputated fingers, thereby reducing development costs.

[0067] 2. Since there are differences between the electromyographic signals of the gestures made by patients with amputated fingers and those of normal people, the present invention directly binds the ones with small differences, and first collects the abnormal electromyographic signals of the gestures with large differences multiple times and calculates the average value, and then binds the average electromyographic signal with the normal electromyographic signal to reduce the error. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] 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 use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0069] Figure 1 This is a block diagram of a bionic limb motion control system of the present invention.

[0070] The numbers in the figure represent:

[0071] 1. Central processing unit 2. Motor controller 3. Motor 1 4. Motor 2 5. Data collector 6. Vision sensor 7. Myoelectric sensor 8. Storage unit 9. Gesture data processing unit 10. Vision data processing unit DETAILED DESCRIPTION

[0072] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely 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 the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0073] Example 1: In some embodiments, please refer to the accompanying drawings of the specification. Figure 1 A bionic limb motion control system includes a central processing unit 1, a motor controller 2 and a data collector 5, wherein the motor controller 2 is electrically connected to a motor 1 3 and a motor 2 4;

[0074] Motor 1 3 is used to control the bending of the bionic finger;

[0075] Motor 2 4 is used to control the swing of the bionic finger;

[0076] The data collector 5 is electrically connected to the data sensing component;

[0077] The data sensing component is used to collect abnormal electromyographic data of the finger amputation patient and image data of the finger amputation patient's gestures, and sends the collected data to the data collector 5, which sends it to the central processing unit 1;

[0078] The central processing unit 1 is electrically connected to a storage unit 8 for storing data collected by the above-mentioned data sensing component;

[0079] The storage unit 8 is also used to store normal myoelectric data of normal human hand gestures and bionic finger control signals corresponding to the normal myoelectric data;

[0080] It also includes a data processing unit, which is electrically connected to the central processing unit 1;

[0081] The data processing unit is used to read the data in the storage unit 8 through the central processing unit 1;

[0082] The data processing unit reads the image data of the hand gesture of the current finger amputee, identifies the type of the hand gesture of the finger amputee and matches it with the corresponding hand gesture electromyographic data of a normal person;

[0083] At the same time, the data processing unit reads the electromyographic data of the finger amputation patient's gesture;

[0084] The data processing unit compares the abnormal electromyographic data of the finger amputee with the normal electromyographic data of a normal person and calculates the error value of the two data;

[0085] The data processing unit identifies the size of the error value and the set value;

[0086] If the error value is less than or equal to the set value, it indicates that the error between the two data is small. At this time, the data processing unit directly binds the abnormal myoelectric data of the finger amputation patient with the gesture with the normal myoelectric data.

[0087] If the error value is greater than the set value, it indicates that the error between the two signals is large. At this time, the data processing unit collects the abnormal electromyographic data multiple times and takes the average electromyographic signal after averaging it and binds it to the normal electromyographic data;

[0088] The central processing unit 1 is used to read the abnormal myoelectric data sent by the data collector 5 and identify the normal myoelectric data bound thereto when the data processing unit completes the binding of the myoelectric signals;

[0089] If there is normal myoelectric data bound to the abnormal myoelectric data, the central processing unit 1 sends the bionic finger control signal corresponding to the normal myoelectric data to the motor controller 2;

[0090] If there is no normal myoelectric data bound to the abnormal myoelectric data, the central processing unit 1 analyzes the short-term feature V of the abnormal myoelectric signal through a sliding window algorithm, matches the average myoelectric signal related to the short-term feature V, and then identifies the normal myoelectric data bound to it. The central processing unit 1 sends the bionic finger control signal corresponding to the normal myoelectric data to the motor controller 2;

[0091] The motor controller 2 is used to control the operation of the motor 1 3 and the motor 2 4 according to the bionic finger control signal.

[0092] When the present invention is used, the data sensing component collects the gesture data of the finger amputee and identifies the gesture type. The visual data processing unit 10 uses the image recognition algorithm to compare the image of the normal finger of the finger amputee with the standard gesture image to identify the gesture type.

[0093] Abnormal electromyographic signals are bound to normal electromyographic signals through data sensing components, thereby enabling the bionic finger control signal to directly use the normal electromyographic signal as the control signal of the bionic finger control end motor 1 3 and motor 2 4 of the finger amputee patient, thereby reducing development costs and enabling the bionic finger to quickly adapt to the basic gesture movements of the finger amputee patient.

[0094] When the present invention is used, since there are differences between the electromyographic signals of the gestures made by the finger amputation patient and the electromyographic signals of the normal person's gestures, the present invention directly binds the ones with small differences, and first collects the abnormal electromyographic signals of the gesture with large differences multiple times and calculates the average value, and then binds the average electromyographic signal with the normal electromyographic signal, thereby reducing the error.

[0095] Embodiment 2: In some embodiments, as Figure 1 As shown, as a preferred embodiment of the present invention, the collected finger amputation patient's gestures include fist clenching, palm stretching, and OK gesture;

[0096] The data sensing component includes a visual sensor 6 and an electromyographic sensor 7;

[0097] The visual sensor 6 and the myoelectric sensor 7 are both electrically connected to the data collector 5;

[0098] The visual sensor 6 is used to collect image data of the hand gestures of the finger amputation patient;

[0099] The myoelectric sensor 7 is used to collect abnormal myoelectric data of the finger amputation patient;

[0100] The visual sensor 6 and the myoelectric sensor 7 send the collected data to the data collector 5 .

[0101] The data processing unit includes a gesture data processing unit 9 and a visual data processing unit 10;

[0102] The gesture data processing unit 9 and the visual data processing unit 10 are both electrically connected to the central processing unit 1 .

[0103] The gesture data processing unit 9 is used to read the image data of the gesture of the finger amputation patient stored in the storage unit 8 through the central processing unit 1. The gesture data processing unit 9 identifies the gesture type through the image recognition algorithm, and sends the gesture type to the central processing unit 1, which then sends it to the visual data processing unit 10;

[0104] The visual data processing unit 10 is used to read the normal electromyographic data corresponding to the gesture stored in the storage unit 8 through the central processing unit 1 according to the gesture type. At the same time, the visual data processing unit 10 reads the abnormal electromyographic data corresponding to the gesture. The visual data processing unit 10 calculates the error between the normal electromyographic data and the abnormal electromyographic data using the mean square error (MSE) algorithm and adopts the following calculation formula:

[0105]

[0106] x represents normal EMG data, y represents abnormal EMG data, n represents the number of sampling points, i represents the sample point, and (x(i)-y(i)) represents the error of each sample point;

[0107] The visual data processing unit 10 compares the calculated MSE value with the set value K;

[0108] When MSE is less than or equal to K, the error between the two surface electromyographic data is small, and the visual data processing unit 10 directly binds the abnormal electromyographic data with the normal electromyographic data;

[0109] When MSE is greater than K, the error between the two surface electromyographic data is large. At this time, the visual data processing unit 10 collects multiple groups of y and calculates the average signal, and uses the following calculation formula:

[0110]

[0111] Among them, v represents the number of y signal groups, N represents the number of y signal groups, A j represents the average electromyographic data of the y signal at the jth time point, y vj represents the value of the y signal of the vth group at the jth time point;

[0112] The short-term characteristic V is a standard deviation characteristic;

[0113] The CPU 1 calculates V and A j The relative standard deviation CV; and the formula is:

[0114] CV=V1 / A j *100%;

[0115] When the CV value is greater than 20%, the abnormal EMG signal is correlated with the average EMG signal.

[0116] When the CV value is less than 20%, the abnormal myoelectric signal is not matched.

[0117] When the present invention is used, when the central processing unit 1 identifies an abnormal myoelectric signal, the central processing unit 1 can directly output a bionic finger control signal to the motor controller 2 for the abnormal myoelectric signal that has been bound to a normal myoelectric signal;

[0118] For the binding of the averaged electromyographic signal with the normal electromyographic signal, the abnormal electromyographic signal is first correlated with the averaged electromyographic signal. If the two signals are correlated, the central processing unit 1 matches the normal electromyographic signal bound to the averaged electromyographic signal, and the central processing unit 1 then outputs the bionic finger control signal to the motor controller 2;

[0119] When the two signals are uncorrelated, it means that the gesture corresponding to the abnormal electromyographic signal is not among the types of gestures recognized by the visual data processing unit 10 .

[0120] The bionic finger of a patient with amputated finger can be controlled based on the existing bionic finger control signal and complete relatively basic gesture movements.

[0121] Embodiment 3: In some embodiments, as Figure 1 As shown, as a preferred embodiment of the present invention, a bionic limb motion control method includes the following steps:

[0122] Step 1: The visual sensor 6 collects image data of the finger amputation patient's gesture, and the electromyographic sensor 7 collects abnormal electromyographic data of the finger amputation patient;

[0123] Step 2: The central processing unit 1 reads the image data of the hand gesture of the finger amputee stored in the storage unit 8, and the hand gesture data processing unit 9 identifies the hand gesture type through the image recognition algorithm;

[0124] Step 3: The visual data processing unit 10 reads the normal electromyographic data corresponding to the gesture stored in the storage unit 8 through the central processing unit 1 according to the gesture type. At the same time, the visual data processing unit 10 reads the abnormal electromyographic data corresponding to the gesture. The visual data processing unit 10 calculates the error between the normal electromyographic data and the abnormal electromyographic data through the mean square error (MSE) algorithm and uses the following calculation formula:

[0125]

[0126] x represents normal EMG data, y represents abnormal EMG data, n represents the number of sampling points, i represents the sample point, and (x(i)-y(i)) represents the error of each sample point;

[0127] Step 4: The visual data processing unit 10 compares the calculated MSE value with the set value K;

[0128] When MSE is less than or equal to K, the error between the two surface electromyographic data is small, and the visual data processing unit 10 directly binds the abnormal electromyographic data with the normal electromyographic data;

[0129] When MSE is greater than K, the error between the two surface electromyographic data is large. At this time, the visual data processing unit 10 collects multiple groups of y and calculates the average signal, and uses the following calculation formula:

[0130]

[0131] Among them, v represents the number of y signal groups, N represents the number of y signal groups, A j represents the average electromyographic data of the y signal at the jth time point, y vj represents the value of the y signal of the vth group at the jth time point;

[0132] Step 5: CPU 1 calculates V and A j The relative standard deviation CV; and the formula is:

[0133] CV=V1 / A j *100%;

[0134] When the CV value is greater than 20%, the abnormal EMG signal is correlated with the average EMG signal.

[0135] When the CV value is less than 20%, no gesture is matched to the abnormal myoelectric signal.

[0136] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. 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 will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A bionic limb motion control system, characterized in that: It comprises a central processing unit (1), a motor controller (2) and a data acquisition device (5), wherein the motor controller (2) is electrically connected to a motor 1 (3) and a motor 2 (4); The motor 1 (3) is used to control the bending of the bionic finger; The second motor (4) is used to control the swing of the bionic finger; The data collector (5) is electrically connected to a data sensor component; The data sensing component is used to collect abnormal electromyographic data of the finger amputation patient and image data of the finger amputation patient's gestures, and sends the collected data to the data collector (5), and the data collector (5) sends it to the central processing unit (1); The central processing unit (1) is electrically connected to a storage unit (8) for storing data collected by the data sensing component; The storage unit (8) is also used to store normal electromyographic data of normal human hand gestures and bionic finger control signals corresponding to the normal electromyographic data; It also includes a data processing unit, which is electrically connected to the central processing unit (1); The data processing unit is used to read data in the storage unit (8) through the central processing unit (1); The data processing unit reads the image data of the current hand gesture of the finger amputee, identifies the type of the hand gesture of the finger amputee and matches it with the corresponding hand gesture electromyographic data of a normal person; At the same time, the data processing unit reads the electromyographic data of the finger amputation patient's gesture; The data processing unit compares the abnormal electromyographic data of the finger amputee with the normal electromyographic data of a normal person and calculates the error value of the two data; The data processing unit identifies the magnitude of the error value and the set value; If the error value is less than or equal to the set value, it indicates that the error between the two data is small. At this time, the data processing unit directly binds the abnormal electromyographic data of the finger amputee with the normal electromyographic data of the gesture; If the error value is greater than the set value, it indicates that the error between the two signals is large. At this time, the data processing unit collects the abnormal electromyographic data multiple times and takes the average electromyographic signal after averaging, and binds it to the normal electromyographic data; The central processor (1) is used for reading abnormal electromyographic data sent by the data collector (5) and identifying normal electromyographic data bound thereto when the data processing unit completes the binding of electromyographic signals; If there is normal electromyographic data bound to the abnormal electromyographic data, the central processing unit (1) sends a bionic finger control signal corresponding to the normal electromyographic data to the motor controller (2); If there is no normal electromyographic data bound to the abnormal electromyographic data, the central processing unit (1) analyzes the short-term feature V of the abnormal electromyographic signal through a sliding window algorithm, matches the average electromyographic signal related to the short-term feature V, and then identifies the normal electromyographic data bound to it. The central processing unit (1) sends the bionic finger control signal corresponding to the normal electromyographic data to the motor controller (2); The motor controller (2) is used to control the operation of motor one (3) and motor two (4) according to the bionic finger control signal.

2. The bionic limb motion control system according to claim 1, characterized in that: The collected gestures of the finger amputation patient include clenching a fist, extending the palm, and OK gesture.

3. The bionic limb motion control system according to claim 2, characterized in that: The data sensing component comprises a visual sensor (6) and an electromyographic sensor (7); The visual sensor (6) and the electromyographic sensor (7) are both electrically connected to the data acquisition device (5); The visual sensor (6) is used to collect image data of the gestures of the patient with amputated fingers; The electromyographic sensor (7) is used to collect abnormal electromyographic data of patients with amputated fingers; The visual sensor (6) and the electromyographic sensor (7) send the collected data to the data collector (5).

4. The bionic limb motion control system according to claim 3, characterized in that: The data processing unit comprises a gesture data processing unit (9) and a visual data processing unit (10); The gesture data processing unit (9) and the visual data processing unit (10) are both electrically connected to the central processing unit (1).

5. The bionic limb motion control system according to claim 4, characterized in that: The gesture data processing unit (9) is used to read the image data of the gesture of the finger amputee stored in the storage unit (8) through the central processing unit (1). The gesture data processing unit (9) identifies the gesture type through an image recognition algorithm, and sends the gesture type to the central processing unit (1), which then sends it to the visual data processing unit (10).

6. The bionic limb motion control system according to claim 5, characterized in that: The visual data processing unit (10) is used to read the normal electromyographic data corresponding to the gesture stored in the storage unit (8) through the central processing unit (1) according to the gesture type, and the visual data processing unit (10) reads the abnormal electromyographic data corresponding to the gesture. The visual data processing unit (10) calculates the error between the normal electromyographic data and the abnormal electromyographic data through the mean square error (MSE) algorithm, and adopts the following calculation formula: x represents normal EMG data, y represents abnormal EMG data, n represents the number of sampling points, i represents the sample point, and (x(i)-y(i)) represents the error of each sample point.

7. The bionic limb motion control system according to claim 6, characterized in that: The visual data processing unit (10) compares the calculated MSE value with a set value K; When MSE is less than or equal to K, the error between the two surface electromyographic data is small, and at this time the visual data processing unit (10) directly binds the abnormal electromyographic data with the normal electromyographic data; When MSE is greater than K, the error between the two surface electromyographic data is large. At this time, the visual data processing unit (10) collects multiple groups of y and calculates the average signal, and uses the following calculation formula: Where v represents the number of y signal groups, N represents the number of y signal groups, and A j represents the average electromyographic data of the y signal at the jth time point, y vj Represents the value of the vth group of y signals at the jth time point.

8. The bionic limb motion control system according to claim 7, characterized in that: The short-term characteristic V is a standard deviation characteristic.

9. The bionic limb motion control system according to claim 8, characterized in that: The central processing unit (1) calculates V and A j The relative standard deviation CV; and the formula is: CV=V1 / A j *100%; When the CV value is greater than 20%, the abnormal EMG signal is related to the average EMG signal; When the CV value is less than 20%, the abnormal electromyographic signal is not matched.

10. A bionic limb motion control method, using the bionic limb motion control system according to claim 9, characterized in that: The following steps are involved: Step 1: The visual sensor (6) collects image data of the hand gestures of the finger amputee, and the electromyographic sensor (7) collects abnormal electromyographic data of the finger amputee; Step 2: The central processor (1) reads the image data of the hand gesture of the finger amputee stored in the storage unit (8), and the hand gesture data processing unit (9) identifies the hand gesture type through an image recognition algorithm; Step 3: The visual data processing unit (10) reads the normal electromyographic data corresponding to the gesture stored in the storage unit (8) through the central processing unit (1) according to the gesture type, and at the same time, the visual data processing unit (10) reads the abnormal electromyographic data corresponding to the gesture. The visual data processing unit (10) calculates the error between the normal electromyographic data and the abnormal electromyographic data through the mean square error (MSE) algorithm, and adopts the following calculation formula: x represents normal EMG data, y represents abnormal EMG data, n represents the number of sampling points, i represents the sample point, and (x(i)-y(i)) represents the error of each sample point; Step 4, the visual data processing unit (10) compares the calculated MSE value with the set value K; When MSE is less than or equal to K, the error between the two surface electromyographic data is small, and at this time the visual data processing unit (10) directly binds the abnormal electromyographic data with the normal electromyographic data; When MSE is greater than K, the error between the two surface electromyographic data is large. At this time, the visual data processing unit (10) collects multiple groups of y and calculates the average signal, and uses the following calculation formula: Where v represents the number of y signal groups, N represents the number of y signal groups, and A j represents the average electromyographic data of the y signal at the jth time point, y vj represents the value of the vth group y signal at the jth time point; Step 5: The CPU (1) calculates V and A j The relative standard deviation CV is: CV=V1 / A j *100%; When the CV value is greater than 20%, the abnormal EMG signal is related to the average EMG signal; When the CV value is less than 20%, no gesture is matched to the abnormal electromyographic signal.

Images