CN111730604B - Mechanical clamping jaw control method and device based on human body electromyographic signals and electronic equipment - Google Patents
Mechanical clamping jaw control method and device based on human body electromyographic signals and electronic equipment Download PDFInfo
- Publication number
- CN111730604B CN111730604B CN202010770968.8A CN202010770968A CN111730604B CN 111730604 B CN111730604 B CN 111730604B CN 202010770968 A CN202010770968 A CN 202010770968A CN 111730604 B CN111730604 B CN 111730604B
- Authority
- CN
- China
- Prior art keywords
- recognition result
- hand
- hand action
- occurrence frequency
- action type
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
- B25J9/161—Hardware, e.g. neural networks, fuzzy logic, interfaces, processor
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J13/00—Controls for manipulators
- B25J13/08—Controls for manipulators by means of sensing devices, e.g. viewing or touching devices
- B25J13/081—Touching devices, e.g. pressure-sensitive
- B25J13/084—Tactile sensors
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
Abstract
The invention provides a mechanical clamping jaw control method, a device and electronic equipment based on human body electromyographic signals, wherein the electromyographic signals of the forearm of an operator are acquired; carrying out filtering processing on the electromyographic signals; inputting the filtered electromyographic signals into a support vector machine classifier to perform action classification so as to identify the hand action type of an operator; taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result to generate a control instruction; sending the control instruction to the robot to control the mechanical clamping jaw to make corresponding action; thereby improving the control accuracy of the mechanical clamping jaw.
Description
Technical Field
The invention relates to the technical field of robot control, in particular to a mechanical clamping jaw control method and device based on human body electromyographic signals and electronic equipment.
Background
In some dangerous occasions, such as explosive disposal and disaster relief in disaster areas with dangerous cases, in order to protect the life safety of disaster relief personnel, a robot is often used to enter a field for operation, and the disaster relief personnel remotely control the robot. At present, when a mechanical clamping jaw of a robot is controlled, the robot is generally operated by a key, a control lever and the like, and flexible action is difficult to perform. Therefore, the mechanical clamping jaw can be controlled based on the myoelectric signal of the human body, the myoelectric signal of the arm of an operator is collected, filtered and input into the classifier to identify the hand action of the operator, and finally the mechanical clamping jaw is controlled to make the same action according to the identification result, so that the mechanical clamping jaw can make flexible action.
However, when the mechanical clamping jaw is controlled according to the identification result, the control command is generally generated according to the identification result at a certain moment before the control command is sent, and if the identification result at the moment is wrong, the control command is wrong, so that the mechanical clamping jaw makes a wrong action, and the control accuracy of the mechanical clamping jaw is low.
Disclosure of Invention
In view of the defects of the prior art, an object of the embodiments of the present application is to provide a mechanical clamping jaw control method and apparatus based on human body electromyographic signals, and an electronic device, which can improve the control accuracy of the mechanical clamping jaw.
In a first aspect, an embodiment of the present application provides a mechanical clamping jaw control method based on a human body electromyographic signal, which is applied to a robot control device, and includes the steps of:
acquiring electromyographic signals of the forearm of an operator;
carrying out filtering processing on the electromyographic signals;
inputting the filtered electromyographic signals into a support vector machine classifier to perform action classification so as to identify the hand action type of an operator;
taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result to generate a control instruction;
and sending the control instruction to the robot to control the mechanical clamping jaw to perform corresponding action.
In the mechanical clamping jaw control method based on human body electromyographic signals, the step of acquiring the electromyographic signals of the forearm of an operator comprises the following steps:
acquiring electrical signals of at least 6 paths of electrode sensors;
amplifying the electric signal;
and carrying out A/D conversion on the amplified electric signal to obtain an electromyographic signal.
Further, the at least 6 electrode sensors are equally spaced around the lower arm of the operator.
In the mechanical clamping jaw control method based on human body electromyographic signals, the step of filtering the electromyographic signals comprises:
and performing Kalman filtering processing on the electromyographic signals.
In the mechanical clamping jaw control method based on the human body electromyographic signal, the support vector machine classifier is obtained by training in the following way:
collecting myoelectric signals of a preset number of preset hand actions as sample data;
carrying out standardization processing on the sample data to obtain training data;
training an initial support vector machine classifier using the training data to generate a final support vector machine classifier.
In the mechanical clamping jaw control method based on the human body electromyographic signal, the step of generating the control command by taking the hand action type with the largest occurrence frequency in a preset period before the current time as a final recognition result comprises the following steps:
acquiring hand motion recognition result information in a preset period before the current moment;
counting the occurrence frequency of various hand actions according to the hand action recognition result information;
taking the hand action type with the largest occurrence frequency as a final recognition result of the hand action;
and generating a control instruction of the next control cycle according to the final identification result.
In a second aspect, an embodiment of the present application provides a mechanical clamping jaw control device based on a human body myoelectric signal, including:
the first acquisition module is used for acquiring electromyographic signals of the forearm of an operator;
the first execution module is used for carrying out filtering processing on the electromyographic signals;
the second execution module is used for inputting the filtered electromyographic signals into a support vector machine classifier for action classification so as to identify the hand action type of an operator;
the third execution module is used for generating a control instruction by taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result;
and the fourth execution module is used for sending the control instruction to the robot so as to control the mechanical clamping jaw to make corresponding action.
In the mechanical clamping jaw control device based on the human body electromyographic signal, the first execution module carries out filtering processing on the electromyographic signal, wherein the filtering processing comprises Kalman filtering processing.
In the mechanical clamping jaw control device based on the human body electromyographic signal, when the third execution module generates the control instruction by taking the hand action type with the largest occurrence frequency in the preset period before the current time as the final recognition result,
firstly, acquiring hand motion recognition result information in a preset period before the current moment;
counting the occurrence frequency of various hand actions according to the hand action recognition result information;
then, taking the hand action type with the largest occurrence frequency as a final recognition result of the hand action;
and finally, generating a control instruction of the next control cycle according to the final recognition result.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor and a memory, where the memory stores a computer program, and the processor is configured to execute the mechanical jaw control method based on a human body myoelectric signal by calling the computer program stored in the memory.
Has the advantages that:
according to the mechanical clamping jaw control method, the mechanical clamping jaw control device and the electronic equipment based on the human body electromyographic signals, the electromyographic signals of the forearm of an operator are obtained; carrying out filtering processing on the electromyographic signals; inputting the filtered electromyographic signals into a support vector machine classifier to perform action classification so as to identify the hand action type of an operator; taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result to generate a control instruction; sending the control instruction to the robot to control the mechanical clamping jaw to make corresponding action; the hand action with the largest occurrence frequency in the preset period is the action type with the largest occurrence probability in the preset period, and the accuracy rate of taking the hand action as the final recognition result is higher than that of directly taking the recognition result at the current moment as the final recognition result, so that the control accuracy rate of the mechanical clamping jaw can be improved.
Drawings
Fig. 1 is a flowchart of a mechanical jaw control method based on a human body electromyogram signal according to an embodiment of the present application.
Fig. 2 is a block diagram of a mechanical jaw control device based on a human body myoelectric signal according to an embodiment of the present application.
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Referring to fig. 1, a mechanical clamping jaw control method based on human body electromyogram signals provided in an embodiment of the present application is applied to a robot control device (a mechanical clamping jaw is disposed at a tail end of a mechanical arm of a robot), and includes the steps of:
A1. acquiring electromyographic signals of the forearm of an operator;
A2. carrying out filtering processing on the electromyographic signals;
A3. inputting the filtered electromyographic signals into a support vector machine classifier to perform action classification so as to identify the hand action type of an operator;
A4. taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result to generate a control instruction;
A5. and sending a control command to the robot to control the mechanical clamping jaw to perform corresponding action.
In step a4, the hand motion with the largest occurrence frequency in the preset period is the motion type with the largest occurrence probability in the preset period, and the accuracy rate of taking the hand motion as the final recognition result is higher than the accuracy rate of directly taking the recognition result at the current moment as the final recognition result, so that the control accuracy rate of the mechanical clamping jaw can be improved.
In some preferred embodiments, a1. the step of acquiring electromyographic signals of the operator's lower arm comprises:
A101. acquiring electrical signals of at least 6 paths of electrode sensors;
A102. amplifying the electric signal;
A103. and carrying out A/D conversion on the amplified electric signal to obtain an electromyographic signal.
The electric signal obtained by the electrode sensor is a relatively weak analog signal, the electric signal can be amplified to obtain a relatively strong electric signal, so that the change condition of the electric signal is more obvious, and the amplified electric signal is subjected to A/D conversion to convert the analog signal into a digital signal so as to be analyzed and processed.
In practical applications, the more electrode sensors, the higher the accuracy of motion recognition, but the larger the data processing amount, if the electrode sensors are too few, the accuracy of motion recognition is too low, and if the electrode sensors are too many, the processing speed is too slow to be applied to practical operations. The number of road electrode sensors is preferably 6-9.
Furthermore, the at least 6 electrode sensors can be attached to the skin of the forearm of the operator in a linear arrangement mode, a surrounding arrangement mode, a matrix arrangement mode and the like; preferably, the electrode sensors are distributed around the forearm of the operator at equal intervals, and the electrode sensors are arranged in a surrounding and equal distribution mode under the same condition through tests, so that the accuracy of motion recognition is highest.
In some preferred embodiments, a2. the step of performing a filtering process on the electromyogram signal includes:
and performing Kalman filtering processing on the electromyographic signals.
Because the electromyographic signal is a physiological signal which is aperiodic, nonlinear and easy to be influenced and contains a large amount of noise, filtering processing is needed to remove the noise; in the prior art, filtering processing is generally performed by adopting wavelet transform algorithm, notch filtering, low-pass filtering and other modes, and the processing methods have an unsatisfactory effect on processing the electromyographic signals containing a large amount of aperiodic Gaussian noise and white noise, so that the Gaussian noise and the white noise can be effectively removed by using Kalman filtering, more accurate input data is provided for motion recognition, and the accuracy of a recognition result is improved.
When Kalman filtering processing is carried out, the covariance Q of system noise and the covariance R of measurement noise can be determined firstly, then a Kalman state equation is constructed, and the true value of the current state is calculated according to the last calculated value and the current measured value by using a recursive optimization method; the method specifically comprises the following steps:
(1) estimating the state of the system at the current time k by the following formula:
wherein the content of the first and second substances,
for the electromyographic signal estimate for the current time k,
for the electromyographic signal estimate at the previous time k-1,
systematic transmission of electromyographic signals for the current time kIn, A is the state transition matrix and B is the input matrix. In the present embodiment, it is preferred that,
since the mathematical model corresponding to the hand movement and the myoelectricity is not known, model-free estimation is used, that is, the original state is assumed to be kept unchanged, so that the input is input
Is set to 0.
(2) Calculating an estimated error covariance matrix by the following equation:
wherein the content of the first and second substances,
for the estimated error covariance matrix for the current time k,
for the estimated error covariance matrix at the last time instant k-1,
in order to be the covariance of the system noise,
is the inverse of the a matrix.
(3) Calculating a kalman filter gain coefficient by the following formula:
wherein the content of the first and second substances,
in order to be the kalman filter gain coefficient,
in order to output the coefficients of the output,
in order to measure the covariance of the noise,
is the inverse of the C matrix.
(4) Updating the state variables according to the measured values:
wherein the content of the first and second substances,
are measured values.
(5) Updating the estimation error covariance matrix:
in this embodiment, the support vector machine classifier used in step a3 is trained by:
s1, collecting myoelectric signals with preset number of preset hand actions as sample data, wherein the hand actions are used as result labels;
s2, carrying out standardization processing on the sample data to obtain training data;
and S3, training the initial support vector machine classifier by using the training data to generate a final support vector machine classifier.
Take the case of having 6 paths of electromyographic signals and 4 types of preset hand movements of fist making, fist showing, wrist upturning and wrist downturning as an example;
in step S1, 6 electromyographic signals of each action may be collected, and 200 data are collected as sample data corresponding to each signal;
in step S2, by performing normalization processing on the sample data, the training speed and the recognition accuracy can be improved; in this embodiment, Z-score normalization processing is performed on sample data:
first, an average value of sample data is required
And standard deviation of
；
Then by the formula
The normalization treatment is carried out, and the normalization treatment is carried out,
for the purpose of the normalized sample data,
is the input sample data;
the data after the Z-score standardization treatment are all concentrated around 0 and become normal distribution with the mean value of 0 and the variance of 1; the comparison is suitable for normalization of data where the maximum or minimum is unknown, or where there is an excess over the sample.
The Support Vector Machine (SVM) is a small sample machine learning method based on statistics, can solve the problem of nonlinear sample classification, can process a multi-feature high-latitude sample data set, has no local minimum value problem, and has the advantage of strong generalization capability; here, the multi-class identification problem using the support vector machine is solved by solving an optimization problem:
to realize multi-class classification, wherein
Is a vector of the input of the features,
is a vector of the weights that is,
is biased, is a relaxation variable, is and
the corresponding output classification label is the number of the sample, which is the number of samples in the sample data set. In this embodiment, the SVM uses a radial basis kernel function (RBF):
In step S3, the specific training process is as follows: pressing a training start button, waiting for 10s, then the system starts to prompt the first action of training, at this time, making a pre-thought hand action and keeping the hand action unchanged, waiting for 30s, then the system prompts the action to be trained, at this time, the system prompts to rest for 5s, and then trains the next action in the same way. When the training is required to be stopped, a training stopping button can be pressed at rest, when the training stopping button is pressed, the system starts to learn and establish a prediction model through a Support Vector Machine (SVM), and after waiting for about 1 minute, the system prompts that the training is finished. After training is finished, the system can predict hand movement according to the input of the hand electromyographic signals by pressing a start recognition button.
Further, A4, the step of generating the control command by taking the hand action type with the largest occurrence frequency in a preset period before the current time as the final recognition result comprises:
A401. acquiring hand motion recognition result information in a preset period T1 before the current moment;
A402. counting the occurrence frequency of various hand actions according to the hand action recognition result information;
A403. taking the hand action type with the largest occurrence frequency as a final recognition result of the hand action;
A404. the control instruction of the next control cycle T2 is generated based on the final recognition result.
Because the sampling frequency is generally high, a plurality of electromyographic signal sampling values can be obtained within one second, for example, the sampling frequency is 50Hz, 50 electromyographic signal sampling values can be obtained within one second, the electromyographic signal sampling values are input into a support vector machine classifier, 50 action recognition results per second are obtained, when a mechanical clamping jaw acts, a certain reaction time exists, if a control command is generated according to each action recognition result and sent to a robot, the mechanical clamping jaw cannot respond in time, therefore, a control command is generally generated periodically and sent to the robot, and the generation period of the control command is the control period T2.
Here, with all the hand motion recognition results within the preset period T1 before the control command is generated each time as the analysis target, the hand motion type with the largest occurrence frequency is selected as the final recognition result of the hand motion, and the hand motion corresponding to the final recognition result is the hand motion with the largest occurrence probability within the preset period T1, which is more accurate than the case where the recognition result at a single moment is the final recognition result. And after sending the control command to the robot, the mechanical gripper will make and maintain the corresponding action in the next control period T2.
The preset period T1 is generally equal to the control period T2, but not limited thereto.
In the actual working process, the hand motion types with the largest occurrence frequency in the preset period T1 may be more than 1, and at this time, if the type of one hand motion is the same as the final recognition result of the previous control period T2, the hand motion type is selected as the final recognition result of the control period T2; otherwise, the hand action type with the largest occurrence frequency in the second half period in the preset period T1 is used as the final recognition result of the hand action, if the hand action type with the largest occurrence frequency in the second half period is still more than 1, the hand action type with the largest occurrence frequency in the second half period is selected as the target, until the hand action type with the largest occurrence frequency is only 1, and the finally selected hand action type is used as the final recognition result. If the hand motion types appearing most frequently in the preset period T1 may be more than 1, the following processing may be performed: the hand motion recognition result information of the preset period T1 is successively reduced from front to back (only the hand motion recognition result information is removed from the hand motion recognition result information set of the preset period T1, not deleted from the storage device), each time the hand motion recognition result information is reduced according to a preset number (the preset number is generally 1, but not limited thereto), for the hand motion recognition result information in the preset period T1, the hand motion type with the largest occurrence frequency in the remaining hand motion recognition result information is inquired after each deletion until the hand motion type with the largest occurrence frequency is only 1, and the finally screened hand motion type is used as the final recognition result. The former method has a faster screening speed, while the latter method has a relatively slower screening speed, but is more accurate because of the smaller amount of data per reduction.
According to the mechanical clamping jaw control method based on the human body electromyographic signals, the electromyographic signals of the lower arm of an operator are obtained; carrying out filtering processing on the electromyographic signals; inputting the filtered electromyographic signals into a support vector machine classifier to perform action classification so as to identify the hand action type of an operator; taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result to generate a control instruction; sending the control instruction to the robot to control the mechanical clamping jaw to make corresponding action; the hand action with the largest occurrence frequency in the preset period is the action type with the largest occurrence probability in the preset period, and the accuracy rate of taking the hand action as the final recognition result is higher than that of directly taking the recognition result at the current moment as the final recognition result, so that the control accuracy rate of the mechanical clamping jaw can be improved.
Referring to fig. 2, an embodiment of the present application further provides a mechanical clamping jaw control device based on a human body electromyographic signal, including a first obtaining module 1, a first executing module 2, a second executing module 3, a third executing module 4, and a fourth executing module 5;
the first acquisition module 1 is used for acquiring electromyographic signals of an operator forearm;
the first execution module 2 is used for performing filtering processing on the electromyographic signals;
the second execution module 3 is used for inputting the filtered electromyographic signals into a support vector machine classifier for action classification so as to identify hand action types of operators;
the third execution module 4 is configured to generate a control instruction by using the hand action type with the largest occurrence frequency in a preset period before the current time as a final recognition result;
and the fourth execution module 5 is used for sending a control command to the robot so as to control the mechanical clamping jaw to make a corresponding action.
In some embodiments, when acquiring the electromyographic signal of the forearm of the operator, the first acquiring module 1 first acquires the electrical signal of at least 6 paths of electrode sensors, then amplifies the electrical signal, and finally performs a/D conversion on the amplified electrical signal to obtain the electromyographic signal.
In some embodiments, the first execution module 2 performs a filtering process on the electromyographic signals, including a kalman filtering process.
In some embodiments, the support vector machine classifier used by the second execution module 3 is trained by:
firstly, collecting myoelectric signals with preset number of preset hand motions as sample data;
then, carrying out standardization processing (for example, Z-score standardization processing) on the sample data to obtain training data;
and finally, training the initial support vector machine classifier by using the training data to generate a final support vector machine classifier.
In some embodiments, when the third execution module 4 generates the control command by using the hand motion type with the largest occurrence frequency in the preset period before the current time as the final recognition result,
firstly, acquiring hand motion recognition result information in a preset period T1 before the current moment;
counting the occurrence times of various hand actions according to the hand action recognition result information;
then, taking the hand action type with the largest occurrence frequency as a final recognition result of the hand action;
and finally, generating a control instruction of the next control period T2 according to the final identification result.
Further, when the hand motion types with the largest occurrence frequency in the preset period T1 are more than 1, if the type of one hand motion is the same as the final recognition result of the previous control period T2, the hand motion type is selected as the final recognition result of the control period T2; otherwise, the hand action type with the largest occurrence frequency in the second half period in the preset period T1 is used as the final recognition result of the hand action, if the hand action type with the largest occurrence frequency in the second half period is still more than 1, the hand action type with the largest occurrence frequency in the second half period is selected as the target, until the hand action type with the largest occurrence frequency is only 1, and the finally selected hand action type is used as the final recognition result. If the hand motion types appearing most frequently in the preset period T1 may be more than 1, the following processing may be performed: the hand motion recognition result information of the preset period T1 is successively reduced from front to back (only the hand motion recognition result information is removed from the hand motion recognition result information set of the preset period T1, not deleted from the storage device), each time the hand motion recognition result information is reduced according to a preset number (the preset number is generally 1, but not limited thereto), for the hand motion recognition result information in the preset period T1, the hand motion type with the largest occurrence frequency in the remaining hand motion recognition result information is inquired after each deletion until the hand motion type with the largest occurrence frequency is only 1, and the finally screened hand motion type is used as the final recognition result. The former method has a faster screening speed, while the latter method has a relatively slower screening speed, but is more accurate because of the smaller amount of data per reduction.
According to the above, the mechanical clamping jaw control device based on the human body electromyographic signal obtains the electromyographic signal of the forearm of the operator; carrying out filtering processing on the electromyographic signals; inputting the filtered electromyographic signals into a support vector machine classifier to perform action classification so as to identify the hand action type of an operator; taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result to generate a control instruction; sending the control instruction to the robot to control the mechanical clamping jaw to make corresponding action; the hand action with the largest occurrence frequency in the preset period is the action type with the largest occurrence probability in the preset period, and the accuracy rate of taking the hand action as the final recognition result is higher than that of directly taking the recognition result at the current moment as the final recognition result, so that the control accuracy rate of the mechanical clamping jaw can be improved.
Referring to fig. 3, an electronic device 100 is further provided in an embodiment of the present application, and includes a processor 101 and a memory 102, where the memory 102 stores a computer program, and the processor 101 is configured to execute the above-mentioned mechanical jaw control method based on a human body electromyogram signal by calling the computer program stored in the memory 102.
The processor 101 is electrically connected to the memory 102. The processor 101 is a control center of the electronic device 100, connects various parts of the entire electronic device using various interfaces and lines, and performs various functions of the electronic device and processes data by running or calling a computer program stored in the memory 102 and calling data stored in the memory 102, thereby performing overall monitoring of the electronic device.
The memory 102 may be used to store computer programs and data. The memory 102 stores computer programs containing instructions executable in the processor. The computer program may constitute various functional modules. The processor 101 executes various functional applications and data processing by calling a computer program stored in the memory 102.
In this embodiment, the processor 101 in the electronic device 100 loads instructions corresponding to one or more processes of the computer program into the memory 102, and the processor 101 runs the computer program stored in the memory 102 according to the following steps, so as to implement various functions: acquiring electromyographic signals of the forearm of an operator; carrying out filtering processing on the electromyographic signals; inputting the filtered electromyographic signals into a support vector machine classifier to perform action classification so as to identify the hand action type of an operator; taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result to generate a control instruction; and sending the control instruction to the robot to control the mechanical clamping jaw to perform corresponding action.
According to the above, the electronic device acquires the electromyographic signals of the forearm of the operator; carrying out filtering processing on the electromyographic signals; inputting the filtered electromyographic signals into a support vector machine classifier to perform action classification so as to identify the hand action type of an operator; taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result to generate a control instruction; sending the control instruction to the robot to control the mechanical clamping jaw to make corresponding action; the hand action with the largest occurrence frequency in the preset period is the action type with the largest occurrence probability in the preset period, and the accuracy rate of taking the hand action as the final recognition result is higher than that of directly taking the recognition result at the current moment as the final recognition result, so that the control accuracy rate of the mechanical clamping jaw can be improved.
In summary, although the present invention has been described with reference to the preferred embodiments, the above-described preferred embodiments are not intended to limit the present invention, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, which are substantially the same as the present invention.
Claims (8)
1. A mechanical clamping jaw control method based on human body electromyographic signals is applied to a robot control device and is characterized by comprising the following steps:
acquiring electromyographic signals of the forearm of an operator;
carrying out filtering processing on the electromyographic signals;
inputting the filtered electromyographic signals into a support vector machine classifier to perform action classification so as to identify the hand action type of an operator;
taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result to generate a control instruction;
sending the control instruction to the robot to control the mechanical clamping jaw to make corresponding action;
the step of generating the control instruction by taking the hand action type with the largest occurrence frequency in a preset period before the current time as the final recognition result comprises the following steps of: acquiring hand motion recognition result information in a preset period before the current moment; counting the occurrence frequency of various hand actions according to the hand action recognition result information;
taking the hand action type with the largest occurrence frequency as a final recognition result of the hand action; generating a control instruction of the next control cycle according to the final identification result;
if the hand motion types with the largest occurrence frequency in the preset period are more than 1, the step of taking the hand motion types with the largest occurrence frequency as the final recognition result of the hand motion comprises the following steps:
if the type of one hand action in the hand action types with the largest occurrence frequency is the same as the final recognition result of the previous control cycle, selecting the hand action type as the final recognition result of the control cycle; otherwise, taking the hand action type with the maximum occurrence frequency in the second half period of the preset period as a final recognition result of the hand action, if the hand action type with the maximum occurrence frequency in the second half period is still more than 1, then taking the second half period of the second half period as an object for screening until the hand action type with the maximum occurrence frequency is only 1, and taking the finally screened hand action type as the final recognition result;
alternatively, the first and second electrodes may be,
and successively reducing the hand action recognition result information of the preset period from front to back, inquiring the hand action type with the largest occurrence frequency in the rest hand action recognition result information after each reduction until the hand action type with the largest occurrence frequency is only 1, and using the finally screened hand action type as a final recognition result.
2. The mechanical jaw control method based on human body electromyographic signals according to claim 1, wherein the step of acquiring electromyographic signals of an operator's forearm comprises:
acquiring electrical signals of at least 6 paths of electrode sensors;
amplifying the electric signal;
and carrying out A/D conversion on the amplified electric signal to obtain an electromyographic signal.
3. The mechanical jaw control method based on human electromyographic signals according to claim 2, wherein the at least 6 ways electrode sensors are equally spaced around the operator's lower arm.
4. The mechanical clamping jaw control method based on human electromyographic signals according to claim 1, wherein the step of filtering the electromyographic signals comprises:
and performing Kalman filtering processing on the electromyographic signals.
5. The mechanical clamping jaw control method based on the human body electromyographic signal according to claim 1, wherein the support vector machine classifier is trained by:
collecting myoelectric signals of a preset number of preset hand actions as sample data;
carrying out standardization processing on the sample data to obtain training data;
training an initial support vector machine classifier using the training data to generate a final support vector machine classifier.
6. The utility model provides a mechanical clamping jaw controlling means based on human flesh electrical signal which characterized in that includes:
the first acquisition module is used for acquiring electromyographic signals of the forearm of an operator;
the first execution module is used for carrying out filtering processing on the electromyographic signals;
the second execution module is used for inputting the filtered electromyographic signals into a support vector machine classifier for action classification so as to identify the hand action type of an operator;
the third execution module is used for generating a control instruction by taking the hand action type with the largest occurrence frequency in a preset period before the current moment as a final recognition result;
the fourth execution module is used for sending the control instruction to the robot so as to control the mechanical clamping jaw to make corresponding action;
when the third execution module takes the hand action type with the largest occurrence frequency in the preset period before the current time as the final recognition result to generate the control instruction,
firstly, acquiring hand motion recognition result information in a preset period before the current moment;
counting the occurrence frequency of various hand actions according to the hand action recognition result information;
then, taking the hand action type with the largest occurrence frequency as a final recognition result of the hand action;
finally, generating a control instruction of the next control cycle according to the final recognition result;
if the hand motion types with the largest occurrence frequency in the preset period are more than 1, the third execution module determines the final recognition result of the hand motion in the following mode:
if the type of one hand action in the hand action types with the largest occurrence frequency is the same as the final recognition result of the previous control cycle, selecting the hand action type as the final recognition result of the control cycle; otherwise, taking the hand action type with the maximum occurrence frequency in the second half period of the preset period as a final recognition result of the hand action, if the hand action type with the maximum occurrence frequency in the second half period is still more than 1, then taking the second half period of the second half period as an object for screening until the hand action type with the maximum occurrence frequency is only 1, and taking the finally screened hand action type as the final recognition result;
alternatively, the first and second electrodes may be,
and successively reducing the hand action recognition result information of the preset period from front to back, inquiring the hand action type with the largest occurrence frequency in the rest hand action recognition result information after each reduction until the hand action type with the largest occurrence frequency is only 1, and using the finally screened hand action type as a final recognition result.
7. The mechanical clamping jaw control device based on the human body electromyographic signal according to claim 6, wherein the first execution module performs filtering processing on the electromyographic signal, wherein the filtering processing includes Kalman filtering processing.
8. An electronic device, characterized by comprising a processor and a memory, wherein the memory stores a computer program, and the processor is used for executing the mechanical clamping jaw control method based on human body electromyography signals according to any one of claims 1 to 5 by calling the computer program stored in the memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010770968.8A CN111730604B (en) | 2020-08-04 | 2020-08-04 | Mechanical clamping jaw control method and device based on human body electromyographic signals and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010770968.8A CN111730604B (en) | 2020-08-04 | 2020-08-04 | Mechanical clamping jaw control method and device based on human body electromyographic signals and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111730604A CN111730604A (en) | 2020-10-02 |
| CN111730604B true CN111730604B (en) | 2020-12-04 |
Family
ID=72657090
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010770968.8A Active CN111730604B (en) | 2020-08-04 | 2020-08-04 | Mechanical clamping jaw control method and device based on human body electromyographic signals and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111730604B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114442798A (en) * | 2020-11-06 | 2022-05-06 | 复旦大学附属妇产科医院 | Portable control system and control method |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07248873A (en) * | 1994-03-08 | 1995-09-26 | Sharp Corp | Controller using myoelectric signal |
| CN102426651B (en) * | 2011-08-25 | 2014-04-16 | 武汉理工大学 | Human body forearm surface electromyogram signal acquisition and pattern recognition system |
| CN105205436B (en) * | 2014-06-03 | 2019-09-06 | 北京创思博德科技有限公司 | A kind of gesture recognition system based on forearm bioelectricity multisensor |
| US11596346B2 (en) * | 2016-08-01 | 2023-03-07 | University Of Utah Research Foundation | Signal processing for decoding intended movements from electromyographic signals |
| CN107553499A (en) * | 2017-10-23 | 2018-01-09 | 上海交通大学 | Natural the gesture motion control system and method for a kind of Multi-shaft mechanical arm |
| CN110413107B (en) * | 2019-06-21 | 2023-04-25 | 浙江科技学院 | Bionic manipulator interaction control method based on electromyographic signal pattern recognition and particle swarm optimization |
| CN111241982B (en) * | 2020-01-07 | 2023-04-28 | 金陵科技学院 | Robot hand recognition method based on CAE-SVM |
-
2020
- 2020-08-04 CN CN202010770968.8A patent/CN111730604B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111730604A (en) | 2020-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107378944B (en) | Multidimensional surface electromyographic signal artificial hand control method based on principal component analysis method | |
| KR102463764B1 (en) | Arrhythmia detection method, device, electronic device and computer storage medium | |
| CN107516075B (en) | Electrocardiosignal detection method and device and electronic equipment | |
| US10061891B2 (en) | System and method for the analysis of biopotential signals using motion artifact removal techniques | |
| EP3367897A1 (en) | An automatic method to delineate or categorize an electrocardiogram | |
| Ahsan et al. | Hand motion detection from EMG signals by using ANN based classifier for human computer interaction | |
| CN115736902B (en) | Constructor management system based on intelligent wearable equipment | |
| CN116602642B (en) | Heart rate monitoring method, device and equipment | |
| CN111844032A (en) | Electromyographic signal processing and exoskeleton robot control method and device | |
| CN111730604B (en) | Mechanical clamping jaw control method and device based on human body electromyographic signals and electronic equipment | |
| CN112426160A (en) | Electrocardiosignal type identification method and device | |
| Kunapipat et al. | Sensor-assisted EMG data recording system | |
| CN111134637B (en) | Signal processing method, signal processing device, electronic equipment and storage medium | |
| JP3816762B2 (en) | Neural network, neural network system, and neural network processing program | |
| JP2000506039A (en) | Method for classifying time-series data having a predetermined number of sample values, for example, electric signals by a computer | |
| CN115407872A (en) | Intelligent man-machine cooperative system evaluation method and device and storage medium | |
| CN111743668B (en) | Prosthesis control method, device, electronic equipment and storage medium | |
| Xue et al. | SEMG based intention identification of complex hand motion using nonlinear time series analysis | |
| Dai et al. | Comparison of sEMG-based feature extraction and hand motion classification methods | |
| CN113057655A (en) | Recognition method, recognition system and detection system for electroencephalogram signal interference | |
| Mukaeda et al. | Development of an anomaly detection method with a novel hidden semi-Markov model incorporating unlearned states | |
| CN111401304A (en) | Calibration method and system of electroencephalogram application system, terminal device and storage medium | |
| Rodriguez Jorge et al. | Adaptive Threshold, Wavelet and Hilbert Transform for QRS Detection in Electrocardiogram Signals | |
| CN115721275B (en) | Pulse wave data acquisition and blood pressure measurement method, device, equipment and medium | |
| CN111449652B (en) | Construction safety monitoring method and device based on brain wave analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |