Wrist-exoskeleton system tremor suppression method, system, storage medium, device

Abstract

The wrist-exoskeleton system tremor suppression method based on equivalent input disturbance is provided, and relates to the technical field of robot control system, and comprises the following steps: a polyhedral model is established to describe the coefficient matrix of the wrist-exoskeleton system dynamics model; a linear band-pass filter is designed to extract the tremor signal from the measurement signal according to the frequency band difference between the active motion and the tremor motion; a state observer is constructed according to the coefficient matrix of the model, and the estimated value of the equivalent input disturbance of the wrist-exoskeleton system is obtained, and the estimated value of the tremor torque is obtained according to the estimated value of the disturbance and the output of the filter; the estimated value of the tremor torque is compensated in reverse, and the system tremor is eliminated. The controller design process is simplified, and efficient tremor suppression is realized without affecting the active motion of the patient.

Description

Technical Field

[0001] This invention relates to the field of robot control system technology, and in particular to methods, systems, storage media, and devices for tremor suppression in wrist-exoskeleton systems. Background Technology

[0002] Wrist tremor is the most prominent symptom of Parkinson's disease. It is an involuntary, irregular oscillating movement whose amplitude varies over time, while its frequency remains relatively stable. Current treatments for tremor mainly include medication and deep cranial stimulation, but these have drawbacks such as drug resistance and significant side effects. In recent years, the rapid development of robotics has led to the development of shock-damping exoskeletons, which offer a new solution for traditional medicine. The working mechanism of a shock-damping exoskeleton involves controlling the output of actuators mounted on the exoskeleton structure to adjust the damping and inertia of the interactive coupling system formed by the patient's upper limb and the exoskeleton's mechanical structure, thereby altering the system's frequency response to achieve shock suppression. Compared to traditional treatment methods, non-invasive shock-damping exoskeletons offer the advantages of lower risk and better wearability.

[0003] Tremor, as a form of involuntary movement, often accompanies voluntary movement. Sensors such as pressure sensors, gyroscopes, or accelerometers can be used to measure signals from the human body, which include both tremor and voluntary movement signals. Studies have shown that wrist tremors in Parkinson's patients occur in the 3.5-7.7 Hz frequency band, while voluntary movement occurs in the 0-2 Hz frequency band. Since these frequency bands are very close, achieving efficient tremor suppression without affecting voluntary movement is crucial for designing a shock-suppressing exoskeleton control system, and has significant theoretical and practical value. Summary of the Invention

[0004] To address the aforementioned problems, this invention proposes a control strategy design method for a tremor suppression wrist-exoskeleton system based on equivalent input disturbances. By coordinating the setting of the bandpass filter bandwidth and the state observer gain, the method resolves the trade-off between tremor suppression performance and the impact on active motion, enabling more effective tuning of control parameters. This invention includes the following steps:

[0005] S1. Establish a dynamic model of the wrist-exoskeleton system, considering the differences in wrist inertia among different patients, and establish a polyhedral model to describe the coefficient matrix of the dynamic model of the wrist-exoskeleton system.

[0006] S2. Based on the frequency band difference between active motion and tremor motion, design a linear bandpass filter to extract the tremor signal from the measurement signal;

[0007] S3. Based on the model and its coefficient matrix, construct a state observer for the wrist-exoskeleton system;

[0008] S4. Based on the state observer, obtain the estimated value of the equivalent input disturbance of the wrist-exoskeleton system, and based on the estimated value of the equivalent input disturbance and the output of the filter, obtain the estimated value of the tremor torque.

[0009] S5. Inversely compensate for the estimated value of the obtained flutter torque in the wrist-exoskeleton system to suppress wrist-exoskeleton system flutter.

[0010] Furthermore, step S1 specifically includes:

[0011] Establish a dynamic model of the wrist-exoskeleton system:

[0012]

[0013] Where the coefficient matrix is:

[0014]

[0015] Where x(t) is the state vector, Let J be the first derivative of x(t), y(t) be the system output, and J be the first derivative of x(t). w and J e These are the moments of inertia of the patient's wrist and exoskeleton in the flexion and extension directions, respectively. m It is the coefficient of friction, τ V (t) is the torque that enables the patient to move actively, τ T (t) is the torque that causes the patient to tremble, τ O (t) is the torque applied by the exoskeleton, τ in this system O (t) is considered as the system control input, taking J and B as examples. m Given the parameter uncertainty, J and B are given. m The range of values ​​for J1 is: J1≤J≤J2, B m1 ≤B m ≤B m2 The coefficient matrix is ​​described using a polyhedral model:

[0016]

[0017] in, λ i (i = 1, 2, 3, 4) to satisfy Uncertain parameters.

[0018] Furthermore, step S2 specifically includes:

[0019] S21, Define Ω respectively v =[0,ω vh ] and Ω t =[ω tl ,ω th[This refers to] the angular frequency band of active motion and motion including tremor.

[0020] S22, the fourth-order bandpass filter is expressed as:

[0021]

[0022] Where s represents the Laplace operator, K is the filter gain, and K is adjusted so that the amplitude of the fourth-order bandpass filter F(s) is within Ω. t =[ω tl ,ω th The range is close to 0dB.

[0023] Furthermore, the state observer is constructed as follows:

[0024]

[0025] in For the observed values ​​of the state vector x(t), for The first derivative, A0 is the coefficient matrix A(J,B) m The nominal value of ), where L is the observer gain.

[0026] Furthermore, the estimated value of the equivalent input disturbance is:

[0027]

[0028] in, It is an estimate of the equivalent input disturbance, and is τ. V (t)+τ T The estimated value of (t), where B0 is the nominal value of the coefficient matrix B(J). yes The output after filtering by F(s) is then filtered by a bandpass filter F(s) to remove... τ in V The estimated value of (t) is used to obtain the flutter torque τ. T The estimated value of (t)

[0029] This invention also proposes a wrist-exoskeleton system tremor suppression system based on equivalent input interference, comprising:

[0030] The first building unit is used to construct a dynamic model of the wrist-exoskeleton system. It takes into account the differences in wrist inertia among different patients and establishes a polyhedral model to describe the coefficient matrix of the dynamic model of the wrist-exoskeleton system.

[0031] The second building unit is used to construct a linear bandpass filter to extract the tremor signal from the measurement signal based on the frequency band difference between active motion and tremor motion.

[0032] The third construction unit is used to construct a state observer based on the model and the model's coefficient matrix;

[0033] The acquisition unit is used to obtain an estimate of the equivalent input disturbance of the wrist-exoskeleton system based on the state observer, and to obtain an estimate of the tremor torque based on the estimate of the equivalent input disturbance and the output of the filter.

[0034] The reverse compensation unit is used to reverse compensate for the estimated value of the obtained flutter torque in the wrist-exoskeleton system, thereby suppressing wrist-exoskeleton system flutter.

[0035] The present invention also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the wrist-exoskeleton system tremor suppression method described above.

[0036] The present invention also proposes an electronic device including a processor and a memory, wherein the processor and the memory are interconnected, wherein the memory is used to store a computer program, the computer program including computer-readable instructions, and the processor is configured to invoke the computer-readable instructions to execute the above-described wrist-exoskeleton system tremor suppression method.

[0037] The beneficial effects of the technical solution provided by this invention are:

[0038] This invention constructs a state observer using a dynamic model of the wrist-exoskeleton system. Based on the model, the model coefficient matrix, and the state observer, it obtains estimates of the equivalent input disturbances of the wrist-exoskeleton system. A bandpass filter is then designed to obtain estimates of the tremor torque for inverse compensation of the system. This achieves efficient tremor suppression without affecting the patient's active movement. This method effectively reduces the complexity of system design and problem analysis, simplifies the controller design process, and provides technical guidance for solving control and human-computer interaction problems in practical exoskeleton systems. It has significant theoretical and applied value. Attached Figure Description

[0039] Figure 1 This is a flowchart of a wrist-exoskeleton system tremor suppression method based on equivalent input interference according to an embodiment of the present invention;

[0040] Figure 2 This is a block diagram of a wrist-exoskeleton system tremor suppression method based on equivalent input interference according to an embodiment of the present invention;

[0041] Figure 3 This is a comparison graph of the transfer functions before and after using the method of the present invention under different model parameters in the embodiments of the present invention. Figure 3(a) shows the amplitude-frequency curves of the open-loop system and the closed-loop system using the method of this invention under different model parameters. Figure 3 (b) shows the phase frequency curves of the open-loop system and the closed-loop system using the method of this invention under different model parameters;

[0042] Figure 4 This is a block diagram of an electronic device according to an exemplary embodiment of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0044] A flowchart of a wrist-exoskeleton system tremor suppression method based on equivalent input interference according to an embodiment of the present invention is shown below. Figure 1 Specifically, it includes the following steps:

[0045] S1. Establish a dynamic model of the wrist-exoskeleton system, considering the differences in wrist inertia among different patients, and establish a polyhedral model to describe the coefficient matrix of the dynamic model of the wrist-exoskeleton system.

[0046] Specifically:

[0047] Establish a dynamic model of the wrist-exoskeleton system:

[0048]

[0049] Where the coefficient matrix is:

[0050]

[0051] Where x(t) is the state vector, Let J be the first derivative of x(t), y(t) be the system output, and J be the first derivative of x(t). w and J e These are the moments of inertia of the patient's wrist and exoskeleton in the flexion and extension directions, respectively. m It is the coefficient of friction, τ V (t) is the torque that enables the patient to move actively, τ T (t) is the torque that causes the patient to tremble, τ O (t) is the torque applied by the exoskeleton, τ in this system O (t) is considered as the system control input, taking J and B as examples. m Given the parameter uncertainty, J and B are given. m The range of values ​​for J1 is: J1≤J≤J2, B m1 ≤B m ≤B m2 The coefficient matrix is ​​described using a polyhedral model:

[0052]

[0053] in, λ i (i = 1, 2, 3, 4) to satisfy Uncertain parameters.

[0054] S2. Based on the frequency band difference between active motion and tremor motion, design a linear bandpass filter to extract the tremor signal from the measurement signal.

[0055] Specifically

[0056] S21, Define Ω respectively v =[0,ω vh ] and Ω t =[ω tl ,ω th [ ] is the angular frequency band of active motion and including tremor motion.

[0057] Most active motion has a frequency less than 0.5 Hz, which will affect Ω. v = [0,π] rad / s is defined as the angular frequency band of active motion. The frequency band for Parkinson's tremor is 3-6 Hz, and for essential tremor it is 4-8 Hz. Therefore, Ω is chosen. t = [6π, 16π] rad / s as the angular frequency band containing flutter motion. According to Ω v and Ω t The band difference between them is used to design a bandpass filter using loop shaping. Where ω vh =πrad / s, ω tl =6πrad / s, ω th = 16πrad / s. Based on design requirements, the amplitude of the bandpass filter in this study is within Ω. v The time should be very small, while in Ω t The value should be close to 0dB. Considering Ω... v and Ω t Since the frequency bands are similar, a fourth-order bandpass filter is selected to achieve the above design.

[0058] S22, the fourth-order bandpass filter is expressed as:

[0059]

[0060] Where s represents the Laplace operator, K is the filter gain, and K is adjusted so that the amplitude of the fourth-order bandpass filter F(s) is within Ω. t =[ω tl ,ω th The range is close to 0dB.

[0061] S3. Based on the model and the coefficient matrix of the model, construct a state observer for the wrist-exoskeleton system.

[0062] The state observer is constructed as follows:

[0063]

[0064] in For the observed values ​​of the state vector x(t), for The first derivative, A0 is the coefficient matrix A(J,B) m The nominal value of ), where L is the observer gain.

[0065] S4. Based on the state observer, obtain an estimate of the equivalent input disturbance of the wrist-exoskeleton system. Based on the estimate of the equivalent input disturbance and the output of the filter, obtain an estimate of the tremor torque.

[0066] Based on the state observer, the estimated value of the equivalent input disturbance is:

[0067]

[0068] in, It is an estimate of the equivalent input disturbance, and is τ. V (t)+τ T The estimated value of (t), where B0 is the nominal value of the coefficient matrix B(J). yes The output after filtering by F(s) is then filtered by a bandpass filter F(s) to remove... τ in V The estimated value of (t) is used to obtain the flutter torque τ. T The estimated value of (t)

[0069] S5. Perform inverse compensation on the estimated value of the flutter torque obtained in the wrist-exoskeleton system to suppress flutter in the wrist-exoskeleton system. Put the exoskeleton control system into torque control mode and give the control input. This helps to suppress tremors.

[0070] Figure 2 This is a block diagram of a wrist-exoskeleton system tremor suppression method based on equivalent input interference according to an embodiment of the present invention, including a wrist-exoskeleton polyhedral model, a state observer, and a tremor torque estimator.

[0071] This invention also proposes a wrist-exoskeleton system tremor suppression system based on equivalent input interference, comprising:

[0072] The first building unit is used to construct a dynamic model of the wrist-exoskeleton system. It takes into account the differences in wrist inertia among different patients and establishes a polyhedral model to describe the coefficient matrix of the dynamic model of the wrist-exoskeleton system.

[0073] The second building unit is used to construct a linear bandpass filter to extract the tremor signal from the measurement signal based on the frequency band difference between active motion and tremor motion.

[0074] The third construction unit is used to construct a state observer based on the model and the model's coefficient matrix;

[0075] The acquisition unit is used to obtain an estimate of the equivalent input disturbance of the wrist-exoskeleton system based on the state observer, and to obtain an estimate of the tremor torque based on the estimate of the equivalent input disturbance and the output of the filter.

[0076] The reverse compensation unit is used to reverse compensate for the estimated value of the obtained flutter torque in the wrist-exoskeleton system, thereby suppressing wrist-exoskeleton system flutter.

[0077] To demonstrate the robustness of the proposed tremor suppression strategy to unknown system parameters, four different sets of J and Bm parameters were defined. Please refer to [reference needed]. Figure 3 , Figure 3 This is a comparison graph of the transfer functions before and after using the method of the present invention under different model parameters in the embodiments of the present invention.

[0078]

[0079] The Bode plots for the open-loop system (dashed line) and the closed-loop system (solid line) after applying the strategy of this invention are given. Figure 3 As shown, the closed-loop responses of different open-loop models are almost identical, at Ω t The tremor was suppressed to a low level, indicating that our tremor suppression strategy is robust to changes in model parameters. Furthermore, in Ω... t When the closed-loop amplitude is lower than the open-loop amplitude, the tremor suppression ratio ranges from 10.27 dB (69.34% suppression) to 75.82 dB (99.99% suppression). Therefore, our tremor suppression strategy can suppress Ω t All tremors. Furthermore, due to tremor inhibition Ω t The angular frequency band amplitude is very low, and the phase lag is relative to Ω. t The impact on closed-loop performance is minimal. At low frequencies below 0.5 Hz, which are related to the autonomous motion frequency band, the amplitudes of the open-loop and closed-loop responses are almost identical, with a maximum phase difference of 28 degrees (Case 1).

[0080] In one exemplary embodiment, the method further includes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the wrist-exoskeleton system tremor suppression method described above.

[0081] In one exemplary embodiment, reference is made to Figure 4 It also includes an electronic device comprising at least one processor, at least one memory, and at least one communication bus.

[0082] The memory contains a computer program, which includes computer-readable instructions. The processor calls the computer-readable instructions stored in the memory via a communication bus to execute the aforementioned wrist-exoskeleton system tremor suppression method.

[0083] This invention addresses the widespread problem of designing exoskeleton control systems for wrist tremor suppression by proposing a tremor suppression strategy based on an equivalent input interference method. This strategy aims to improve the exoskeleton control system's ability to filter tremor signals and reduce hysteresis in response to active motion signals. The method effectively reduces the complexity of system design and problem analysis, simplifies the controller design process, and provides technical guidance for solving control and human-computer interaction problems in practical exoskeleton systems, possessing significant theoretical and practical value.

[0084] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for tremor suppression in the wrist-exoskeleton system based on equivalent input interference, characterized in that, Includes the following steps: S1. Establish a dynamic model of the wrist-exoskeleton system, considering the differences in wrist inertia among different patients, and establish a polyhedral model to describe the coefficient matrix of the dynamic model of the wrist-exoskeleton system. S2. Based on the frequency band difference between active motion and trembling motion, design a linear bandpass filter; S3. Based on the model and its coefficient matrix, construct a state observer for the wrist-exoskeleton system; S4. Based on the state observer, obtain the estimated value of the equivalent input disturbance of the wrist-exoskeleton system, and based on the estimated value of the equivalent input disturbance and the output of the filter, obtain the estimated value of the tremor torque. S5. In the wrist-exoskeleton system, the estimated value of the obtained flutter torque is reversed to suppress wrist-exoskeleton system flutter. Step S1 is as follows: Establish a dynamic model of the wrist-exoskeleton system: Where the coefficient matrix is: in, Let this be the state vector of the wrist-exoskeleton system. for The first derivative, Output to the wrist-exoskeleton system and These are the moments of inertia of the patient's wrist and exoskeleton in the flexion and extension directions, respectively. It is the coefficient of friction. To generate the torque needed for the patient to move actively, The torque required to induce tremors in the patient The exoskeleton applies torque. Consider J and as system control inputs. Given the parameter uncertainty, J and The range of values ​​for: The coefficient matrix is ​​described using a polyhedral model: in, , To meet Uncertain parameters.

2. The wrist-exoskeleton system tremor suppression method based on equivalent input interference according to claim 1, characterized in that, Step S2 is as follows: S21, Define respectively and It is the angular frequency band of active motion and including tremor motion; S22, the fourth-order bandpass filter is expressed as: Where s represents the Laplace operator, K is the filter gain, and K is adjusted so that the amplitude of the fourth-order bandpass filter F(s) is within the range of... The range is close to 0dB.

3. The wrist-exoskeleton system tremor suppression method based on equivalent input interference according to claim 2, characterized in that, The state observer is constructed as follows: in State vector The observed values, for The first derivative, Coefficient matrix The nominal value of L is the observer gain.

4. The wrist-exoskeleton system tremor suppression method based on equivalent input interference according to claim 3, characterized in that, The estimated value of equivalent input interference is: in, It is an estimate of the equivalent input disturbance. The estimated value, Coefficient matrix The nominal value, yes pass The filtered output is then filtered by a bandpass filter F(s). In The estimated value is used to obtain the flutter torque. The estimated value .

5. A wrist-exoskeleton system tremor suppression system based on equivalent input interference, used to implement the method of claim 1, characterized in that, include: The first building unit is used to construct a dynamic model of the wrist-exoskeleton system. It takes into account the differences in wrist inertia among different patients and establishes a polyhedral model to describe the coefficient matrix of the dynamic model of the wrist-exoskeleton system. The second building unit is used to construct a linear bandpass filter to extract the tremor signal from the measurement signal based on the frequency band difference between active motion and tremor motion. The third construction unit is used to construct a state observer based on the model and the model's coefficient matrix; The acquisition unit is used to obtain an estimate of the equivalent input disturbance of the wrist-exoskeleton system based on the state observer, and to obtain an estimate of the tremor torque based on the estimate of the equivalent input disturbance and the output of the filter. The reverse compensation unit is used to reverse compensate for the estimated value of the obtained flutter torque in the wrist-exoskeleton system, thereby suppressing wrist-exoskeleton system flutter.

6. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-4.

7. An electronic device, characterized in that, The device includes a processor and a memory interconnected thereto, wherein the memory is used to store a computer program, the computer program including computer-readable instructions, and the processor is configured to invoke the computer-readable instructions to perform the method as described in any one of claims 1-4.

Images