Method, device and system for controlling bionic pneumatic soft hand based on electromyographic signals

Abstract

The embodiment of the invention provides a control method, device and system for a bionic pneumatic soft hand based on an electromyographic signal, and relates to human bionics, the control method for the pneumatic soft hand based on the electromyographic signal comprises the following steps: acquiring a surface electromyographic signal of a corresponding gesture of the pneumatic soft hand; the method comprises the following steps: performing preprocessing, feature extraction and decoding on a surface electromyogram signal to obtain a motion instruction, inputting the motion instruction into a neural mimicry muscle model, calculating a tension signal, inputting the tension signal into a tension-pressure conversion model, calculating a pressure signal, and proportionally controlling the input pressure of the pneumatic soft hand according to the pressure signal. The tension-pressure conversion model is obtained by combining a Gaussian simplified model of a muscle tension and length relation with a muscle volume conservation model. According to the control method provided by the embodiment of the invention, the control accuracy can be improved, and the user experience is improved.

Description

Technical Field

[0001] This application relates to the field of human bionics technology, and in particular to a control method, device and system for a bionic pneumatic soft hand based on electromyography signals. Background Technology

[0002] Surface electromyography (SEMG) signals are the combined effect of superficial muscle electromyographic signals and electrical activity on nerve trunks on the skin surface, and can reflect neuromuscular activity to a certain extent. Therefore, SEMG signals have significant practical value in clinical medicine, ergonomics, rehabilitation medicine, and sports science. For example, for people with disabilities, prosthetic systems based on SEMG signals can bring numerous benefits.

[0003] Currently, the pneumatic soft hand is a type of prosthesis driven by air pressure, characterized by safety, flexibility, lightweight, and high biomimicry, making it suitable for use by people with disabilities. It utilizes elastic materials (such as silicone) combined with an internal air cavity structure, achieving adaptive deformation similar to biological tissue through air pressure changes, naturally conforming to the shape of objects without complex control. Combined with an air pressure control device based on surface electromyography signals, it can essentially simulate human hand movements, enabling grasping and other operations.

[0004] However, most existing soft hand systems based on electromyography (EMG) signals use the collected EMG signals to perform simple on / off control of the pneumatic soft hand, which fails to reflect the actual dynamic force regulation needs of the human body, resulting in poor accuracy and affecting the user experience. Summary of the Invention

[0005] Based on this, this application provides a control method, device, system, readable storage medium, and product for a pneumatic soft hand based on electromyography signals, which can improve the accuracy of pneumatic soft hand control to a certain extent, thereby enhancing the user experience.

[0006] In a first aspect, this application provides a control method for a biomimetic pneumatic soft hand based on electromyographic signals, comprising:

[0007] Acquire surface electromyography signals corresponding to gestures of the pneumatic soft hand;

[0008] The surface electromyography (EMG) signal is preprocessed, features are extracted and decoded to obtain the estimated motion command; the motion command is input into the neuromimetic muscle model to calculate the tension signal; the tension signal is input into the tension-pressure conversion model to calculate the pressure signal.

[0009] The input pressure of the pneumatic soft hand is controlled proportionally according to the pressure signal to drive the corresponding action of the pneumatic soft hand;

[0010] The tension-pressure conversion model is obtained by combining the simplified Gaussian model of the relationship between muscle tension and length with the muscle volume conservation model.

[0011] In one possible implementation, the tension-pressure conversion model is obtained by combining a simplified Gaussian model of the relationship between muscle tension and length with a model of muscle volume conservation, satisfying the following relationship:

[0012]

[0013] In the formula: T represents the instantaneous tension of the muscle, in N; T max is the maximum tension of the muscle at length L0, in N; L0 is the length of the muscle when it is naturally relaxed, in m; L is the instantaneous length of the muscle, in m; k is a constant coefficient.

[0014] The muscle volume conservation model satisfies the following relationship:

[0015]

[0016] In the formula: V is the muscle volume, in meters. 3 S represents the cross-sectional area of ​​the muscle, in meters. 2 (Units for length, area, and volume are m, m², and m³.)

[0017] The tension-pressure conversion model is obtained, which satisfies the following relationship:

[0018]

[0019] In the formula: P in This is a pressure signal, measured in Pa.

[0020] In one possible implementation, the input pressure of the pneumatic soft hand is controlled proportionally based on a pressure signal, including:

[0021] The pressure signal is converted into a voltage signal via digital-to-analog conversion;

[0022] The input pressure of the pneumatic soft hand is controlled proportionally by an electronically controlled proportional valve based on the voltage signal.

[0023] In one possible implementation, the pressure signal is converted from a digital-to-analog converter to a voltage signal, satisfying the following relationship:

[0024]

[0025] In the formula: V out The voltage signal output by the digital-to-analog converter, in volts (V). ref is the reference voltage of the digital-to-analog converter, in volts (V); n is the resolution of the digital-to-analog converter.

[0026] The input pressure of the pneumatic soft hand is controlled proportionally through an electronically controlled proportional valve based on the voltage signal, satisfying the following relationship:

[0027]

[0028] In the formula: P act P represents the input pressure of the pneumatic soft hand, measured in Pa. max P represents the internal gas pressure of the pneumatic soft hand in its fully bent state, expressed in Pa. min V represents the internal gas pressure of the pneumatic soft hand in its unbent state, expressed in Pa. max The maximum nominal voltage of the electronically controlled proportional valve, in volts (V). min This is the minimum nominal voltage of the electronically controlled proportional valve, in volts (V).

[0029] In one possible implementation, motion commands are input into a neuromimetic muscle model to calculate tension signals, including:

[0030] Receive the pulse sequence in the motion command;

[0031] Generates simulated muscle tension signals based on movement commands;

[0032] Among them, the change in length converted from the fingertip force of the pneumatic soft hand is fed back in real time, and the firing frequency of the pulse sequence is dynamically adjusted according to the tension signal.

[0033] In one possible implementation, the surface electromyography signal is preprocessed, its features extracted, and then decoded, including:

[0034] The surface electromyography signal is amplified and filtered.

[0035] Extract motor neural features;

[0036] Generate a pulse sequence.

[0037] Secondly, this application provides a control device for a biomimetic pneumatic soft hand based on electromyographic signals, comprising:

[0038] The acquisition module acquires surface electromyography signals corresponding to the gestures of the pneumatic soft hand;

[0039] The processing module preprocesses, extracts, and decodes the surface electromyography signals to obtain estimated motion commands; it inputs the motion commands into the neuromimetic muscle model to calculate the tension signal; and it inputs the tension signal into the tension-pressure conversion model to calculate the pressure signal.

[0040] The drive module proportionally controls the input pressure of the pneumatic soft hand according to the pressure signal, so as to drive the corresponding action of the pneumatic soft hand;

[0041] The tension-pressure conversion model is obtained by combining the simplified Gaussian model of the relationship between muscle tension and length with the muscle volume conservation model.

[0042] Thirdly, this application provides a prosthetic system, comprising:

[0043] Pneumatic soft hand body;

[0044] Electromyography (EMG) sensors used to fit the limbs corresponding to the pneumatic soft hand body;

[0045] Main controller connected to the electromyography (EMG) signal sensor:

[0046] The main controller is used to acquire surface electromyography (EMG) signals collected by the EMG signal sensor and execute the steps of the control method for the bionic pneumatic soft hand based on EMG signals provided in the first aspect to control the movement of the pneumatic soft hand body.

[0047] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the control method for the bionic pneumatic soft hand based on electromyographic signals provided in the first aspect.

[0048] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the control method for a bionic pneumatic soft hand based on electromyographic signals provided in the first aspect.

[0049] The control method, device, system, readable storage medium, and product of the biomyograph pneumatic soft hand based on electromyography (EMG) signals provided in this application include: acquiring surface EMG signals corresponding to the gestures of the pneumatic soft hand; obtaining estimated motion commands by preprocessing, feature extraction, and decoding the surface EMG signals; inputting the motion commands into a neuromimetic muscle model to calculate the tension signal; inputting the tension signal into a tension-pressure conversion model to calculate the pressure signal; and controlling the input pressure of the pneumatic soft hand proportionally according to the pressure signal to drive the corresponding action of the pneumatic soft hand. The tension-pressure conversion model is obtained by combining a simplified Gaussian model of the relationship between muscle tension and length with a muscle volume conservation model. Therefore, the control method for the biomimetic pneumatic soft hand based on electromyography (EMG) signals provided in this application can accurately obtain the underlying instructions of the movement intention based on EMG activation when extracting motion commands from surface EMG signals. When calculating the tension signal from the movement intention, it can simulate tetanic contraction, force-frequency relationship, muscle fatigue and other situations from a deeper physiological mechanism to obtain a more accurate and refined tension estimate. When converting the tension signal into a pressure signal, it can accurately convert the pressure signal for controlling the pneumatic soft hand by combining the mechanism of muscle length change and muscle volume conservation in the Hill model. This allows for real-time and proportional control of the pneumatic soft hand to complete adaptive grasping actions, which improves the accuracy of control to a certain extent, making it more human-like and refined, and thus improving the user experience. Attached Figure Description

[0050] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0051] Figure 1 A flowchart of the control method for the biomyographed pneumatic soft hand based on electromyography signals provided in this application;

[0052] Figure 2 The signal processing steps of each part in the control method of the biomyographed pneumatic soft hand based on electromyography signals provided in this application. Figure 1 ;

[0053] Figure 3 The signal processing steps of each part in the control method of the biomyographed pneumatic soft hand based on electromyography signals provided in this application. Figure 2 ;

[0054] Figure 4 The signal processing steps of each part in the control method of the biomyographed pneumatic soft hand based on electromyography signals provided in this application. Figure 3 ;

[0055] Figure 5 A schematic diagram of the connection relationship of the control device for the biomyograph pneumatic soft hand based on electromyography signals provided in this application;

[0056] Figure 6 A schematic diagram of the connection relationship of the prosthetic system provided in this application.

[0057] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0058] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0059] First, let me explain the terms used in this application:

[0060] Electromyographic signal: refers to the signal formed by the temporal and spatial superposition of action potentials of motor units in numerous muscle fibers;

[0061] Surface electromyography (EMG) signals refer to the combined effect of superficial muscle EMG signals and electrical activity on nerve trunks on the skin surface, which can reflect neuromuscular activity to a certain extent.

[0062] Currently, the inherent limitations of traditional rigid robotic hands in terms of compliance and human-computer interaction safety severely restrict their application in rehabilitation medicine, precision operations, and other fields. This not only causes numerous inconveniences to the daily lives of people with disabilities but also hinders the adaptive development of service robots in complex environments and restricts the in-depth application of human-machine collaboration. By adopting novel biomimetic devices such as pneumatic soft hands, the naturalness and safety of human-computer interaction can be significantly improved, bringing revolutionary changes to related fields. The lack of biomimetic control capabilities in traditional rigid robotic hands makes it difficult to meet the needs of precise operations, and this problem has become a key technological bottleneck that urgently needs to be solved in the fields of rehabilitation medicine and human-computer interaction.

[0063] A pneumatic soft hand is a flexible robotic device driven by air pressure, characterized by safety, compliance, lightweight design, and high biomimicry. Its core advantage lies in using elastic materials (such as silicone) combined with an internal air cavity structure. Through changes in air pressure, it achieves adaptive deformation similar to biological tissue, naturally conforming to the shape of objects without complex control. This design makes it particularly suitable for medical rehabilitation (such as hand function training), precision grasping (of fragile items), and human-computer interaction. Compared to traditional rigid robotic hands, it can more naturally simulate human hand movements, achieving continuous adjustment from gentle touch to strong grip while ensuring operational safety, representing an important development direction in the field of flexible robotics.

[0064] The pneumatic soft hand control system is the core technology for realizing human-like operation. Currently, most electromyography (EMG) controlled soft hand systems only realize a simple mapping of "EMG signal → on / off control", which cannot truly reflect the dynamic force regulation needs of the human body, resulting in stiff pneumatic soft hand movements.

[0065] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0066] Firstly, referring to Figures 1 to 4 As shown, this application provides a control method for a biomimetic pneumatic soft hand based on electromyographic signals, including:

[0067] S100: Acquire surface electromyographic signals corresponding to the gestures of the pneumatic soft hand.

[0068] Specifically, surface electromyography (EMG) signals of corresponding gestures (i.e., limb gestures) can be obtained through EMG sensors, which can be attached to the surface of the corresponding limb.

[0069] S200. Preprocess, extract features, and decode the surface electromyography signal to obtain the estimated motion command; input the motion command into the neuromimetic muscle model to calculate the tension signal; input the tension signal into the tension-pressure conversion model to calculate the pressure signal; the tension-pressure conversion model is obtained by combining the Gaussian simplified model of the relationship between muscle tension and length with the muscle volume conservation model.

[0070] First, since the original surface electromyography (EMG) signal is very weak and contains a lot of noise, preprocessing can purify the signal and lay the foundation for subsequent analysis. Preprocessing can include noise reduction, rectification, and smoothing. During feature extraction, features that effectively represent the intention to move can be extracted, such as time-domain features, frequency-domain features, time-frequency-domain features, and higher-order features. The neuromimetic muscle model includes a motor neuron model, a muscle model, a muscle spindle model, a Golgi tendon organ model, and a synapse model. During decoding, the neuromimetic muscle model analyzes the received motion commands in real time, simulating the real characteristics of human spinal cord neuromuscular reflexes. The model's output muscle tension serves as the control signal for the pneumatic soft hand. Changes in the air pressure driven by the pneumatic soft hand are converted and calculated into changes in muscle length in the biomimetic model and sent to the muscle model. The muscle spindle model senses changes in muscle length in real time, while the Golgi tendon organ senses the output muscle tension in real time, adjusting the firing of motor neurons, and thus adjusting the muscle tension output calculated by the muscle model in real time. The entire loop constitutes a complete spinal reflex circuit, enabling the pneumatic soft hand control to possess human-like neuromuscular reflex characteristics.

[0071] Secondly, since motor nerve commands are a series of pulse sequences, these pulse sequences are input into the neuromimetic muscle model, and the pulse sequences are simulated as chemical signals inside the muscle fibers. For example, differential equations are used to simulate the functional relationship of calcium ion changes over time, and differential equations are used to simulate the functional relationship of muscle fiber activation degree changes over time. Then, based on the contraction dynamics and the simulated tension signal of the muscle generated by force, a more accurate and refined tension estimate is obtained.

[0072] Then, by combining the mechanism of muscle length change and muscle volume conservation in the Hill model, the pressure signal for controlling the pneumatic soft hand was accurately converted.

[0073] S300: The input pressure of the pneumatic soft hand is controlled proportionally according to the pressure signal to drive the corresponding action of the pneumatic soft hand.

[0074] Specifically, the pressure signal can be proportionally amplified into a current signal, and then the input pressure of the pneumatic soft hand can be controlled proportionally through a proportional valve, thereby driving the pneumatic soft hand to perform grasping actions more accurately.

[0075] Therefore, the application of the control method for the biomimetic pneumatic soft hand based on electromyography (EMG) signals provided in this application can accurately obtain the underlying instructions of the action intention based on EMG activation when extracting motion commands from surface EMG signals. When calculating the tension signal from the motion, it can simulate tetanic contraction, force-frequency relationship, muscle fatigue and other situations from a deeper physiological mechanism to obtain a more accurate and refined tension estimate. When converting the tension signal into a pressure signal, it can accurately convert the pressure signal for controlling the pneumatic soft hand by combining the mechanism of muscle length change and muscle volume conservation in the Hill model. This allows for real-time and proportional control of the pneumatic soft hand to complete adaptive grasping actions, which improves the accuracy of control to a certain extent, making it more human-like and refined, thereby enhancing the user experience.

[0076] It is worth noting that this control method for the biomyograph-based pneumatic soft hand breaks through the limitations of traditional pneumatic control. By simulating the human neural control mechanism, it establishes a complete biomyograph-based conversion pathway from force intention recognition to pneumatic output. This enables real-time control of the pneumatic soft hand through biomyograph-based control and conversion of real-time human electromyographic signals, resulting in human-like compliant operation performance. When grasping objects, it adjusts muscle force in real time based on tactile feedback, forming a closed loop of "electromyographic activation → fine force control → adaptive grasping," thereby meeting the application needs of high-end medical rehabilitation and precision operation.

[0077] Moreover, unlike rigid robotic hands that require the integration of motors, circuits, transmission systems, and sensors, pneumatic soft hands are typically driven by an external air source and air pump, making them lighter and more portable. Compared to motor-driven soft hands, pneumatic soft hands only require compressed air, avoiding the potential hazards to users caused by motor leakage or short circuits. Additionally, the compressibility of air in a collision can slow down and reduce the impact of rigid forces.

[0078] In some embodiments, the tension-pressure conversion model is obtained by combining the Gaussian simplified model of the relationship between muscle tension and length with the muscle volume conservation model, satisfying the following relationship:

[0079]

[0080] In the formula: T represents the instantaneous tension of the muscle, in N; T maxThe maximum tension of the muscle at length L0 is expressed in N, which can be 250 N; L0 is the length of the muscle when it is naturally relaxed, expressed in meters, which can be 0.156 m; L is the instantaneous length of the muscle, expressed in meters; k is a constant coefficient, usually taken as 0.2 to 0.3.

[0081] Based on the assumption of muscle volume conservation, treating muscle as an incompressible material whose volume does not change, the muscle volume conservation model satisfies the following relationship:

[0082]

[0083] In the formula: V is the muscle volume, in meters. 3 This can be 0.000025m 3 S represents the cross-sectional area of ​​the muscle, in meters. 2 ;

[0084] The tension-pressure conversion model (eliminating S and substituting L) satisfies the following relationship:

[0085]

[0086] In the formula: P in This is a pressure signal, measured in Pa.

[0087] In this way, by capturing the nonlinear changes in muscle tension (such as Gaussian distribution characteristics), the tension response of human muscles at different lengths can be simulated more realistically. This nonlinear conversion process makes the mapping from electromyographic signals to muscle tension and then to air pressure signals more consistent with the actual force application process of the human body, realizing a continuous and smooth adjustment from "gentle touch to strong grip", improving the smoothness and naturalness of operation.

[0088] Furthermore, in this embodiment, controlling the input pressure of the pneumatic soft hand proportionally according to the pressure signal includes:

[0089] S310: Convert the pressure signal into a voltage signal via digital-to-analog conversion;

[0090] S320: The input pressure of the pneumatic soft hand is controlled proportionally by an electronically controlled proportional valve according to the voltage signal.

[0091] This allows for precise control from low to high pressure signals, meeting the demand for greater driving force in pneumatic soft hands. It should be noted that the real-time acquisition and conversion of electromyographic signals, along with the proportional valve's signal reception for real-time air pressure control, eliminates inertial resistance, allowing for a very smooth simulation of human hand grasping movements.

[0092] Furthermore, in this embodiment, the pressure signal is converted from digital to analog signal to voltage signal, satisfying the following relationship:

[0093]

[0094] In the formula: V out The voltage signal output by the digital-to-analog converter, in volts (V). ref is the reference voltage of the digital-to-analog converter, in volts (V); n is the resolution of the digital-to-analog converter.

[0095] The input pressure of the pneumatic soft hand is controlled proportionally through an electronically controlled proportional valve based on the voltage signal, satisfying the following relationship:

[0096]

[0097] In the formula: P act P represents the input pressure of the pneumatic soft hand, measured in Pa. max P represents the internal gas pressure of the pneumatic soft hand in its fully bent state, expressed in Pa. min V represents the internal gas pressure of the pneumatic soft hand in its unbent state, expressed in Pa. max The maximum nominal voltage of the electronically controlled proportional valve, in volts (V). min This is the minimum nominal voltage of the electronically controlled proportional valve, in volts (V).

[0098] In some embodiments, motion commands are input into a neuromimetic muscle model to calculate tension signals, including:

[0099] Receive the pulse sequence in the motion command;

[0100] Generates simulated muscle tension signals based on movement commands;

[0101] Among them, the change in length converted from the fingertip force of the pneumatic soft hand is fed back in real time, and the firing frequency of the pulse sequence is dynamically adjusted according to the tension signal.

[0102] In one possible implementation, the surface electromyography signal is preprocessed, its features extracted, and then decoded, including:

[0103] The surface electromyography signal is amplified and filtered.

[0104] Extract motor neural features;

[0105] Generate a pulse sequence.

[0106] In summary, the control method for the biomimetic pneumatic soft hand based on electromyography (EMG) signals designed in this application achieves real-time conversion and calculation from EMG signals (motor intention) to pneumatic pressure signals (control commands) through real-time signal processing (such as EMG signal acquisition, amplification, and filtering), neuromorphic calculation, and efficient mathematical transformation (such as Gaussian models and volume conservation formulas). Combined with stepless smooth pressure control using a proportional valve, the system can adjust the air pressure without delay, ensuring smooth and unimpeded soft hand movements.

[0107] Secondly, such as Figure 5 As shown, this application provides a control device for a biomimetic pneumatic soft hand based on electromyographic signals, comprising:

[0108] The acquisition module acquires surface electromyography signals corresponding to the gestures of the pneumatic soft hand;

[0109] The processing module preprocesses, extracts, and decodes the surface electromyography signals to obtain estimated motion commands; it inputs the motion commands into the neuromimetic muscle model to calculate the tension signal; and it inputs the tension signal into the tension-pressure conversion model to calculate the pressure signal.

[0110] The drive module proportionally controls the input pressure of the pneumatic soft hand according to the pressure signal, so as to drive the corresponding action of the pneumatic soft hand;

[0111] The tension-pressure conversion model is obtained by combining the simplified Gaussian model of the relationship between muscle tension and length with the muscle volume conservation model.

[0112] Therefore, the control device for the biomimetic pneumatic soft hand based on electromyography (EMG) signals provided in this application can accurately obtain the underlying instructions of the action intention based on EMG activation when extracting motion commands from surface EMG signals. When calculating the tension signal from the motion, it can simulate tetanic contraction, force-frequency relationship, muscle fatigue and other situations from a deeper physiological mechanism to obtain a more accurate and refined tension estimate. When converting the tension signal into a pressure signal, it can accurately convert the pressure signal for controlling the pneumatic soft hand by combining the mechanism of muscle length change and muscle volume conservation in the Hill model. This allows for real-time and proportional control of the pneumatic soft hand to complete adaptive grasping actions, which improves the accuracy of control to a certain extent, making it more human-like and refined, and thus enhancing the user experience.

[0113] Thirdly, this application provides a prosthetic system, such as Figure 6 As shown, it includes:

[0114] Pneumatic soft hand body;

[0115] Electromyography (EMG) sensors used to fit the limbs corresponding to the pneumatic soft hand body;

[0116] Main controller connected to the electromyography (EMG) signal sensor:

[0117] The main controller is used to acquire surface electromyography (EMG) signals collected by the EMG signal sensor and execute the steps of the control method for the bionic pneumatic soft hand based on EMG signals provided in the first aspect to control the movement of the pneumatic soft hand body.

[0118] Specifically, the pneumatic soft hand body can be an anthropomorphic silicone inflatable structure with five fingers and multiple phalanges. Electromyography (EMG) sensors can be attached to the user's limb surface to collect surface EMG signals in real time. The main controller receives data from the EMG sensors in real time and executes the steps of the control method for the biomimetic pneumatic soft hand based on EMG signals to control the movement of the pneumatic soft hand body.

[0119] Therefore, the prosthetic system provided in this application, when extracting motion commands from surface electromyography (EMG) signals, accurately obtains the underlying commands of the movement intention based on EMG activation. When calculating tension signals from motion, it simulates tetanic contraction, force-frequency relationship, muscle fatigue, and other situations from a deeper physiological mechanism, resulting in a more accurate and refined tension estimate. When converting the tension signal into a pressure signal, it accurately converts the pressure signal for controlling the pneumatic soft hand by combining the mechanism of muscle length change and muscle volume conservation in the Hill model. This allows for real-time and proportional control of the pneumatic soft hand to complete adaptive grasping actions, improving the accuracy of control to a certain extent, making it more human-like and refined, and thus enhancing the user experience.

[0120] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the control method for the bionic pneumatic soft hand based on electromyographic signals provided in the first aspect.

[0121] Specifically, the aforementioned instructions can be executed by a processor to complete the above method. For example, a non-transitory computer-readable storage medium can be ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc. When the instructions in the storage medium are executed by the processor of the terminal device, the terminal device is able to perform the aforementioned electromyography signal processing method.

[0122] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the control method for a bionic pneumatic soft hand based on electromyographic signals provided in the first aspect.

[0123] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0124] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A control method for a biomimetic pneumatic soft hand based on electromyographic signals, characterized in that, include: Acquire surface electromyography signals corresponding to gestures of the pneumatic soft hand; The surface electromyography signal is preprocessed, features are extracted and decoded to obtain an estimated motion command; the motion command is input into a neuromimetic muscle model to calculate the tension signal; The tension signal is input into the tension-pressure conversion model to calculate the pressure signal; The input pressure of the pneumatic soft hand is controlled proportionally according to the pressure signal to drive the corresponding action of the pneumatic soft hand; The tension-pressure conversion model is obtained by combining the simplified Gaussian model of the relationship between muscle tension and length with the muscle volume conservation model.

2. The control method for the biomyopic pneumatic soft hand based on electromyographic signals according to claim 1, characterized in that, The tension-pressure conversion model is obtained by combining the Gaussian simplified model of the relationship between muscle tension and length with the muscle volume conservation model, and satisfies the following relationship: In the formula: T represents the instantaneous tension of the muscle, in N; T max is the maximum tension of the muscle at length L0, in N; L0 is the length of the muscle when it is naturally relaxed, in m; L is the instantaneous length of the muscle, in m; k is a constant coefficient. The muscle volume conservation model satisfies the following relationship: In the formula: V is the muscle volume, in meters. 3 S represents the cross-sectional area of ​​the muscle, in meters. 2 ; The tension-pressure conversion model obtained satisfies the following relationship: In the formula: P in The pressure signal is expressed in Pa.

3. The control method for the biomyopic pneumatic soft hand based on electromyographic signals according to claim 2, characterized in that, The step of proportionally controlling the input pressure of the pneumatic soft hand based on the pressure signal includes: The pressure signal is converted into a voltage signal via digital-to-analog conversion; The input pressure of the pneumatic soft hand is controlled proportionally by an electronically controlled proportional valve according to the voltage signal.

4. The control method for the biomyoscopic pneumatic soft hand based on electromyographic signals according to claim 3, characterized in that, The pressure signal is converted into a voltage signal via digital-to-analog conversion, satisfying the following relationship: In the formula: V out The voltage signal output by the digital-to-analog converter, in volts (V). ref is the reference voltage of the digital-to-analog converter, in volts (V); n is the resolution of the digital-to-analog converter. The input pressure of the pneumatic soft hand is controlled proportionally via an electronically controlled proportional valve according to the voltage signal, satisfying the following relationship: In the formula: P act P represents the input pressure of the pneumatic soft hand, in Pa. max P represents the intracavitary gas pressure when the pneumatic soft hand is in a fully bent state, in Pa. min The pressure of the gas inside the cavity of the pneumatic soft hand when it is not bent, in Pa; V max The maximum nominal voltage of the electro-hydraulic proportional valve, in V; V min The minimum nominal voltage of the electronically controlled proportional valve is expressed in volts (V).

5. The control method for the biomyographed pneumatic soft hand based on electromyography signals according to any one of claims 1 to 4, characterized in that, The motion command is input into the neuromimetic muscle model to calculate the tension signal, including: Receive the pulse sequence in the motion command; The tension signal of the simulated muscle is generated according to the motion command; Specifically, the system provides real-time feedback on the changes in length derived from the fingertip force of the pneumatic soft hand, and dynamically adjusts the firing frequency of the pulse sequence based on the tension signal.

6. The control method for the biomyoscopic pneumatic soft hand based on electromyographic signals according to claim 5, characterized in that, The preprocessing, feature extraction, and decoding of the surface electromyography signal include: The surface electromyography signal is amplified and filtered. Extract motor neural features; The pulse sequence is generated.

7. A control device for a biomimetic pneumatic soft hand based on electromyographic signals, characterized in that, include: The acquisition module acquires surface electromyography signals corresponding to the gestures of the pneumatic soft hand; The processing module preprocesses, extracts, and decodes the surface electromyography signals to obtain estimated motion commands; it then inputs the motion commands into a neuromimetic muscle model to calculate tension signals. The tension signal is input into the tension-pressure conversion model to calculate the pressure signal; The drive module controls the input pressure of the pneumatic soft hand proportionally according to the pressure signal, so as to drive the pneumatic soft hand to perform corresponding actions. The tension-pressure conversion model is obtained by combining the simplified Gaussian model of the relationship between muscle tension and length with the muscle volume conservation model.

8. A prosthetic system, characterized in that, include: Pneumatic soft hand body; Electromyography (EMG) sensors used to fit the limbs corresponding to the pneumatic soft hand body; Main controller connected to the electromyography signal sensor: The main controller is used to acquire the surface electromyography (EMG) signals collected by the EMG signal sensor and execute the steps of the control method for the bionic pneumatic soft hand based on EMG signals as described in any one of claims 1 to 6, so as to control the movement of the pneumatic soft hand body.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the control method for the biomyographed pneumatic soft hand based on electromyographic signals as described in any one of claims 1-6.

10. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the control method for an inspired pneumatic soft hand based on electromyographic signals as described in any one of claims 1-6.

Images