Musculoskeletal collaborative deformity correction method and robot

Abstract

The embodiment of the application provides a kind of musculoskeletal collaborative deformity correction method and robot, comprising: with the expected pose of bone correction, the change rate of expected pose, the actual displacement amount of pneumatic component, the rigidity parameter and damping parameter of pre-set, the torque signal of torque sensor acquisition as input, by bone correction controller, the first pressure signal for controlling pneumatic component inflation and deflation is output, so that pneumatic component is applied to bone hard tissue by inflation and deflation Correction force;With the collected electromyographic signal as input, the predicted muscle fatigue degree is output using the pre-constructed muscle fatigue prediction model;According to the predicted muscle fatigue degree, determine the air bag adjustment pressure;With air bag adjustment pressure and the current pressure of the air bag as the input of muscle correction controller, the second pressure signal for controlling air bag inflation and deflation is output by the muscle correction controller, so that air bag is applied to muscle soft tissue by inflation and deflation Correction force.The application can promote the collaborative rehabilitation of bone hard tissue and muscle soft tissue.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of rehabilitation assistive devices, in particular to a muscle-bone collaborative deformity correction method and robot. BACKGROUND

[0002] Traditional lower limb skeletal deformity correction can wear a rigid correction support for correcting the deformity of the skeleton after the operation. In the rehabilitation process, the support needs to be adjusted multiple times to apply different correction forces to the affected limb. The operation is complicated and the adjustment is inaccurate. Moreover, since the effect of muscle soft tissue on skeletal hard tissue correction is not considered, the synchronous and collaborative rehabilitation of skeletal hard tissue and muscle soft tissue cannot be realized, and the rehabilitation effect is not ideal. SUMMARY

[0003] Therefore, the purpose of the embodiment of the present application is to provide a muscle-bone collaborative deformity correction method and robot, which can adaptively control the correction force applied to the affected limb by the auxiliary correction robot, and promote the collaborative rehabilitation of skeletal hard tissue and muscle soft tissue.

[0004] To achieve the above purpose, the embodiment of the present application provides a muscle-bone collaborative deformity correction method for controlling an auxiliary correction robot, the auxiliary correction robot comprising a variable stiffness external fixator for correcting skeletal hard tissue and a flexible corrector for correcting muscle soft tissue, the variable stiffness external fixator being provided with a pneumatic component and a torque sensor, and the flexible corrector being provided with an air bag; the method comprising:

[0005] Taking the expected pose of skeletal correction, the change rate of the expected pose, the actual displacement amount of the pneumatic component, the preset stiffness parameter and damping parameter, and the torque signal collected by the torque sensor as inputs, a first pressure signal for controlling the inflation and deflation of the pneumatic component is output by a skeletal correction controller, so that the pneumatic component applies a correction force to the skeletal hard tissue through inflation and deflation;

[0006] Taking the collected electromyographic signal as input, a predicted muscle fatigue degree is output by using a pre-constructed muscle fatigue degree prediction model;

[0007] According to the relationship between the predicted muscle fatigue degree and the air pressure of the air bag, the air bag adjustment pressure is determined;

[0008] Taking the air bag adjustment pressure and the current pressure of the air bag as inputs of a muscle correction controller, a second pressure signal for controlling the inflation and deflation of the air bag is output by the muscle correction controller, so that the air bag applies a correction force to the muscle soft tissue through inflation and deflation.

[0009] Optionally, the bone correction controller comprises an adaptive impedance controller and a first fuzzy PID controller, the adaptive impedance controller is configured to determine control parameters of the first fuzzy PID controller according to the desired pose, a change rate of the desired pose, the stiffness parameter, a damping parameter, a torque signal and an actual displacement of the pneumatic component, and the first fuzzy PID controller is configured to output the first pressure signal according to the control parameters.

[0010] Optionally, a mathematical model of the adaptive impedance controller is:

[0011]

[0012] wherein, is a target velocity, q d is the desired pose, is a change rate of the desired pose, q is an actual pose of the pneumatic component, K is the stiffness parameter, B is the damping parameter, and τ is the torque signal; wherein the actual pose of the pneumatic component is determined based on a forward kinematics method according to the actual displacement of the pneumatic component.

[0013] Optionally, an input of the first fuzzy PID controller is a displacement error between a displacement of the pneumatic component determined based on the control parameters and the actual displacement of the pneumatic component, and an output of the first fuzzy PID controller is the first pressure signal.

[0014] Optionally, the actual pose of the pneumatic component is determined based on a forward kinematics method according to the actual displacement of the pneumatic component; the displacement of the pneumatic component determined based on the control parameters is a displacement of the pneumatic component calculated based on a target pose according to an inverse kinematics method, and the target pose is obtained by integrating the target velocity.

[0015] Optionally, the method further comprises:

[0016] determining a predicted damping parameter according to the predicted muscle fatigue degree, a preset maximum fatigue degree and a preset maximum damping parameter;

[0017] determining a target damping parameter according to a relationship between the predicted damping parameter and a preset minimum damping parameter and the maximum damping parameter;

[0018] adjusting the preset damping parameter according to the target damping parameter.

[0019] Optionally, the method further comprises:

[0020] processing the electromyographic signal to obtain an electromyographic feature;

[0021] inputting the myoelectric feature, and outputting a predicted muscle force by using a pre-constructed muscle force prediction model;

[0022] determining a predicted stiffness parameter according to the predicted muscle force, a preset maximum muscle force, and a preset maximum stiffness parameter;

[0023] determining a target stiffness parameter according to a relationship between the predicted stiffness parameter and a preset minimum stiffness parameter and the maximum stiffness parameter;

[0024] adjusting the preset stiffness parameter according to the target stiffness parameter.

[0025] Optionally, a gasbag adjustment pressure is determined according to a relationship between the predicted muscle fatigue degree and the gasbag pressure, and the method is as follows:

[0026]

[0027] wherein P is the gasbag adjustment pressure, is the muscle fatigue degree, and a proportional coefficient k is determined according to a maximum value of the gasbag pressure and a maximum value of the muscle fatigue degree.

[0028] Optionally, the muscle fatigue degree prediction model is constructed based on an autoregressive model, and the muscle force prediction model is constructed based on a GD-FNN model.

[0029] Embodiments of the present application also provide a muscle-bone collaborative deformity correction robot for controlling an auxiliary correction robot, the auxiliary correction robot including a variable stiffness external fixator for correcting a bone hard tissue and a flexible corrector for correcting a muscle soft tissue, the variable stiffness external fixator being provided with a pneumatic component and a torque sensor, and the flexible corrector being provided with a gasbag; and the robot including:

[0030] a bone auxiliary rehabilitation module for inputting a desired pose of a bone correction, a change rate of the desired pose, an actual displacement amount of the pneumatic component, a preset stiffness parameter and a damping parameter, and a torque signal collected by the torque sensor, and outputting, by a bone correction controller, a first pressure signal for controlling inflation and deflation of the pneumatic component to enable the pneumatic component to apply a correction force to the bone hard tissue through inflation and deflation;

[0031] a fatigue degree prediction module for inputting a collected myoelectric signal, and outputting a predicted muscle fatigue degree by using a pre-constructed muscle fatigue degree prediction model;

[0032] an adjustment pressure determination module for determining a gasbag adjustment pressure according to a relationship between the predicted muscle fatigue degree and the gasbag pressure;

[0033] The muscle auxiliary rehabilitation module is used for taking the air bag adjustment pressure and the current pressure of the air bag as inputs of a muscle correction controller, outputting a second pressure signal for controlling air charging and discharging of the air bag from the muscle correction controller, so that the air bag exerts a correction force on the muscle soft tissue through air charging and discharging.

[0034] From the above, it can be seen that the musculoskeletal collaborative deformity correction method and robot provided by the embodiments of the present application utilize the bone correction controller to perform feedback control on the pneumatic component according to the expected pose of the bone recovery, the interaction force between the variable stiffness external fixator and the bone, the interaction between the bone hard tissue and the muscle soft tissue, and the displacement amount of the pneumatic component of the variable stiffness external fixator, and adjust the correction force exerted on the bone hard tissue by controlling air charging and discharging of the pneumatic component; the muscle correction controller is utilized to perform feedback control on the air bag according to the air bag adjustment pressure and the current pressure, and adjust the correction force exerted on the muscle soft tissue by controlling air charging and discharging of the air bag; at the same time, in the rehabilitation process, the damping parameter and the stiffness parameter of the bone correction controller are dynamically adjusted, and the collaborative correction force exerted on the bone and muscle is dynamically adjusted, so as to promote the collaborative rehabilitation of the bone hard tissue and the muscle soft tissue. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only the embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0036] Figure 1 It is a structural schematic diagram of an auxiliary correction robot of the embodiments of the present application;

[0037] Figure 2 It is a structural schematic diagram of a flexible corrector of the embodiments of the present application;

[0038] Figure 3 It is a structural schematic diagram of a variable stiffness external fixator of the embodiments of the present application;

[0039] Figure 4 It is a structural schematic diagram of a pneumatic component of the embodiments of the present application;

[0040] Figure 5 It is a control method flowchart of the embodiments of the present application;

[0041] Figure 6 It is a control flowchart of the embodiments of the present application;

[0042] Figure 7 It is an expected pose diagram of bone hard tissue correction of the embodiments of the present application;

[0043] Figure 8 Control flowchart for bone hard tissue correction of embodiments of the present application;

[0044] Figure 9 Control flowchart for first fuzzy PID controller of embodiments of the present application;

[0045] Figure 10 Control flowchart for muscle soft tissue correction of embodiments of the present application;

[0046] Figure 11 Processing flowchart for predicting muscle force of embodiments of the present application;

[0047] Figure 12 Processing flowchart for predicting muscle fatigue of embodiments of the present application;

[0048] Figure 13 Structural block diagram of robot of embodiments of the present application;

[0049] Figure 14 Structural block diagram of electronic device of embodiments of the present application. DETAILED DESCRIPTION

[0050] In order to make the objects, technical solutions and advantages of the present disclosure clearer, the present disclosure will be further described in detail below with reference to specific embodiments and drawings.

[0051] It should be noted that, unless otherwise defined, technical terms or scientific terms used in embodiments of the present application should be understood as their common meanings to those skilled in the art to which the present disclosure belongs. The terms "first", "second" and similar terms used in embodiments of the present application do not represent any order, number or importance, but are only used to distinguish different components. The terms "include" or "contain" and similar terms mean that the elements or objects before the terms cover the elements or objects listed after the terms and their equivalents, and do not exclude other elements or objects. The terms "connect" or "connected" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "up", "down", "left", "right" and the like only represent relative positional relationships, and when the absolute positions of the described objects change, the relative positional relationships may also change accordingly.

[0052] In the following, the technical solutions of the present application will be further described in detail through specific embodiments.

[0053] As Figures 1-4As shown, in this embodiment, to promote the rehabilitation of the affected limb, an auxiliary correction robot is worn on the affected limb, the auxiliary correction robot includes a flexible corrector 10 for correcting muscle soft tissue and a variable stiffness external fixator 20 for correcting skeletal hard tissue, the flexible corrector 10 is provided with an air bag and an air pressure sensor, the variable stiffness external fixator 20 is provided with a pneumatic component and a torque sensor, and an air source 30 is used to provide gas for the pneumatic component and the air bag.

[0054] Specifically, the flexible corrector 10 includes a thigh medial muscle patch 11, a thigh lateral muscle patch 12, a first air bag 13, and a magnetic suction joint 14. The thigh medial muscle patch 11 and the thigh lateral muscle patch 12 are respectively attached to the thigh medial muscle and the thigh lateral muscle. The first air bag 13 is arranged on the side of the muscle patch away from the thigh muscle, one end of the first air bag 13 is connected to the muscle patch, the other end is connected to the magnetic suction joint 14, and the magnetic suction joint 14 is adsorbed on the variable stiffness external fixator 20. In some ways, the first air bag 13 is a double-cavity auxiliary traction air bag, which can assist in traction correction of soft tissues such as muscles by elongation or contraction during inflation and deflation.

[0055] The variable stiffness external fixator 20 includes an upper fixing ring 21, a lower fixing ring 24, at least one bone needle 23, and at least one pneumatic component 22. The two ends of the pneumatic component 22 are respectively detachably connected to the upper fixing ring 21 and the lower fixing ring 24, and one end of the bone needle 23 is inserted into the bone and the other end is detachably connected to the upper fixing ring 21 through a bone needle fixing part. The pneumatic component 22 includes an upper support rod 22-2, a lower support rod 22-5, a gas cylinder, and a second air bag 22-4 arranged in the gas cylinder. One end of the upper support rod 22-2 is hinged to the upper fixing ring 21, and the other end extends into the gas cylinder and abuts against the second air bag 22-4. One end of the lower support rod 22-5 is hinged to the lower fixing ring 24, and the other end extends into the gas cylinder and abuts against the second air bag 22-4. The second air bag 22-4 is connected in communication with the air source 30 through a quick insertion air port 22-3. The second air bag 22-4 is elongated or contracted in the gas cylinder during inflation and deflation, thereby elongating or contracting the upper support rod 22-2 and the lower support rod 22-5, adjusting the length of the pneumatic component, adjusting the pose of the variable stiffness external fixator, and thereby adjusting the traction force and traction angle of the variable stiffness external fixator on the affected limb bone. The pneumatic component 22 is also provided with a sensor mounting part for mounting sensors such as a torque sensor and a displacement sensor. The torque sensor is used to collect torque signals between the pneumatic component and the muscle and bone in real time, and the displacement sensor is used to collect displacement of the pneumatic component in real time.

[0056] As shown in Figures 5-8 , 10, the muscle-bone collaborative deformity correction method provided by the embodiments of the present application is used for adaptive control of the auxiliary correction robot, and the method comprises:

[0057] S501: taking the expected pose of the bone correction, the change rate of the expected pose, the actual displacement amount of the pneumatic component, the preset stiffness parameter and damping parameter, and the torque signal collected by the torque sensor as inputs, outputting, by the bone correction controller, a first pressure signal for controlling the inflation and deflation of the pneumatic component, so that the pneumatic component applies a correction force to the hard tissue of the bone through inflation and deflation;

[0058] In this embodiment, in order to correct the deformed bone to a normal bone, the expected pose of the bone in the correction process needs to be determined first. In some ways, the deformed parameters of the bone can be obtained through X-ray, including: valgus, varus angular deformity and lateral, medial displacement deformity on the frontal image, hyperextension, flexion angular deformity and anterior, posterior displacement deformity on the lateral image, shortening, overlength displacement deformity on the axial image, and external rotation, internal rotation rotation deformity on the axial image obtained through clinical surface measurement. According to the deformed parameters and the preset normal parameters, the A* algorithm is used to determine the initial trajectory of the bone correction, and the B-spline curve is used to optimize the initial trajectory to obtain the expected trajectory in the bone correction process, and the expected pose is determined according to the expected trajectory. The above algorithms for determining the initial trajectory and the expected trajectory are only exemplary, and the specific process of determining the initial trajectory and the expected trajectory is not described in detail in this embodiment.

[0059] After determining the expected pose, the bone correction controller is used to perform feedback control on the pneumatic component of the variable stiffness external fixator, and the correction force of the variable stiffness external fixator on the affected limb bone is adjusted through the control of the pneumatic component, so that the affected limb bone can gradually recover to the normal bone state according to the expected pose.

[0060] In some embodiments, the bone correction controller includes a first fuzzy PID controller and an adaptive impedance controller, the first fuzzy PID controller outputs a control signal based on the control parameters determined by the adaptive impedance controller, so that the variable stiffness external fixator applies a correction force to the affected limb bone to promote muscle-bone coordinated rehabilitation under the control of the control signal. Although the first fuzzy PID controller alone can control the variable stiffness external fixator to apply a correction force to the affected limb bone to correct it to a normal bone according to the expected pose, since the interaction between the affected limb bone and the variable stiffness external fixator during the recovery process and the interaction between the affected limb bone and the muscle soft tissue are not considered, the use of the first fuzzy PID controller alone cannot achieve the best effect of coordinated rehabilitation of the bone hard tissue and the muscle soft tissue.

[0061] Thus, in the embodiment, the control parameters of the first fuzzy PID controller are determined by using the adaptive impedance controller in combination with the interaction between the limb bone and the variable stiffness external fixator during the recovery process, and the interaction between the limb bone and the muscle soft tissue. The adaptive impedance controller is used to determine the control parameters of the first fuzzy PID controller according to the desired pose, the rate of change of the desired pose, the stiffness parameter, the damping parameter, the torque signal and the actual displacement amount of the pneumatic component, and the first fuzzy PID controller is used to output the first pressure signal according to the control parameters. The mathematical model of the adaptive impedance controller is:

[0062]

[0063] wherein, and q respectively represent the actual acceleration, velocity and pose of the variable stiffness external fixator during the correction process, and q d respectively represent the desired acceleration, desired velocity and desired pose, F represents the interaction force signal, M, B and K respectively represent the inertia parameter, damping parameter and stiffness parameter.

[0064] To ensure the safety during the correction of deformity, the speed of the variable stiffness external fixator is usually slow, and thus the acceleration term can be ignored, and the mathematical model can be:

[0065]

[0066] The speed controlled by the adaptive impedance controller can be obtained as:

[0067]

[0068] When applied to an auxiliary correction robot, the torque signal can be used to represent the interaction force signal, and the following can be obtained:

[0069]

[0070] wherein τ is the torque signal, is the rate of change of the desired pose, i.e. the desired velocity, which is obtained by derivation of the desired pose.

[0071] After determining the target speed that can be controlled by the adaptive impedance controller according to formula (4), the target speed is integrated to obtain the target pose q t The target pose is used as the control parameter determined by the adaptive impedance controller, and the first fuzzy PID controller is used to control the variable stiffness external fixator to perform actions based on the target pose.

[0072] In some embodiments, the input of the first fuzzy PID controller is a displacement error between a displacement amount of the pneumatic component of the variable stiffness external fixator determined based on the control parameter and a displacement amount of the pneumatic component collected by the displacement sensor, and the output of the first fuzzy PID controller is the first pressure signal. In some ways, the displacement amount of the pneumatic component can be determined based on a kinematics inverse solution method according to the target pose, and the kinematics inverse solution method and the kinematics positive solution method are not described in detail.

[0073] In the present embodiment, the first fuzzy PID controller outputs the first pressure signal for controlling the pneumatic component of the variable stiffness external fixator to charge and discharge, and the pneumatic component charges and discharges under the control of the first pressure signal, and exerts a correction force on the bone hard tissue during the charging and discharging process. The first fuzzy PID controller includes a first PID controller and a first fuzzy controller, the first fuzzy controller takes as input a displacement error between a displacement amount of the pneumatic component collected by the displacement sensor and a displacement amount of the pneumatic component determined based on the control parameter, and a change rate of the displacement error, and outputs the first control parameter K P , K I , K D of the first PID controller; and the first PID controller takes as input the displacement error and outputs the first pressure signal according to the first control parameter. The pneumatic component generates displacement during the charging and discharging process, and the pose q o of the variable stiffness external fixator is calculated based on the actual displacement amount generated according to the kinematics positive solution method, and the calculated pose q o is approximated as the actual pose q of the variable stiffness external fixator in formula (4) considering that the tracking error of the first fuzzy PID controller is small.

[0074] In some embodiments, the pneumatic component charges and discharges according to the pressure difference between the first pressure signal and the current pressure, that is, the pneumatic component charges and discharges under the control of the pressure difference between the first pressure signal output by the first PID controller and the current pressure, and adjusts the traction force and the traction angle of the variable stiffness external fixator on the limb bone during the charging and discharging process.

[0075] As Figure 9As shown, in some modes, the variable stiffness external fixator 20 is provided with 6 pneumatic components for correcting the hard tissue of the bone from different angles. Correspondingly, a first fuzzy PID controller is configured for controlling the 6 pneumatic components respectively, each first fuzzy PID controller takes the displacement error between the displacement amount of the corresponding pneumatic component determined based on the control parameters and the displacement amount of the corresponding pneumatic component collected by the displacement sensor as input, and outputs a first pressure signal for controlling the corresponding pneumatic component, each pneumatic component is controlled by the pressure difference between the corresponding first pressure signal and the current pressure to inflate and deflate, and in the process of inflation and deflation, the displacement amount of the corresponding pneumatic component is adjusted, the overall pose of the variable stiffness external fixator is adjusted, and appropriate correction force is applied to the limb bone.

[0076] S502: Taking the collected electromyographic signal as input, a pre-constructed muscle fatigue degree prediction model is used to output the predicted muscle fatigue degree;

[0077] S503: According to the relationship between the predicted muscle fatigue degree and the air pressure of the air bag, the air bag adjustment pressure is determined;

[0078] S504: Taking the air bag adjustment pressure and the current pressure of the air bag as the input of the muscle correction controller, the second pressure signal for controlling the inflation and deflation of the air bag is output by the muscle correction controller, so that the air bag applies a correction force to the muscle soft tissue through inflation and deflation.

[0079] In this embodiment, in order to promote the synchronous recovery of muscle soft tissue, a muscle fatigue degree prediction model is pre-constructed, the muscle fatigue degree is predicted according to the input electromyographic signal by using the muscle fatigue degree prediction model, the air bag adjustment pressure is determined according to the relationship between the predicted muscle fatigue degree and the air pressure of the air bag, and the second pressure signal for controlling the inflation and deflation of the air bag is output by the muscle correction controller according to the air bag adjustment pressure and the current pressure, so that the air bag applies a suitable correction force to the muscle soft tissue through inflation and deflation.

[0080] As shown in some embodiments, Figure 12 The muscle fatigue degree prediction model is established based on an autoregressive model. If {ε t} is white noise, the p-order difference equation is called a p-order autoregressive model, and the p-order autoregressive model can be regarded as the relationship between the electromyographic signal at the current time and the electromyographic signal at p times before the current time. The model is represented as:

[0081]

[0082] Where a i (i=0, 1, 2, …, p) is the autoregressive coefficient, X t is the observation value of the electromyographic signal at the current time t, and X t-i is the observation value sequence of the electromyographic signal before the current time t.

[0083] The parameter estimation of the autoregressive model is determined by using the least square method, denoted as:

[0084]

[0085] In the formula, N is 2p+1 to match the dimension of a; combining formula (5) and (6) can obtain:

[0086] Y=Xa+ε (7)

[0087] In order to minimize the square sum of the residual error ε, the objective function can be expressed as:

[0088] S(a)=(Y-Xa) T (Y-Xa) (8)

[0089] The derivative of the parameter a in formula (8) is zero, and the least square estimation of the parameter a can be obtained as:

[0090]

[0091] According to the change rate of the element a i in the parameter a, the muscle fatigue degree is determined, denoted as:

[0092] CR=(a iM -a i ) / a i ×100% (10)

[0093] In the formula, a i is the parameter value of the autoregressive model at the current time t, and a iM is the parameter value of the autoregressive model at the time t+M after M sampling periods.

[0094] Experiments prove that the parameter a1 in the autoregressive model can directly reflect the degree of change of the muscle state over time, and therefore, the muscle fatigue degree predicted by the muscle fatigue degree prediction model is:

[0095]

[0096] In the formula, a1 is the parameter value of the autoregressive model at the current time t, and a 1M is the parameter value of the autoregressive model at the time t+M.

[0097] In some modes, since the autoregressive model has sensitivity to the input signal, the collected electromyography signal (surface electromyography signal) needs to be filtered for enhancing the effective signal component in the electromyography signal and suppressing the influence of noise; the electromyography signal after the filtering processing is normalized for reducing the error caused by the difference in muscle activity of different subjects. The electromyography signal after the normalization processing is taken as the observation value of the electromyography signal of the muscle fatigue degree prediction model, and is used for predicting the muscle fatigue degree.

[0098] After the muscle fatigue degree is predicted by using the muscle fatigue degree prediction model according to the collected electromyography signal, the airbag adjustment pressure is determined according to the relationship between the predicted muscle fatigue degree and the air pressure of the airbag, and is expressed as:

[0099]

[0100]

[0101] wherein P is the airbag adjustment pressure, the muscle fatigue degree is inversely proportional to the airbag adjustment pressure P, that is, when the muscle fatigue degree is large, the pressure applied by the airbag is reduced to relax the local muscle, and when the muscle fatigue degree is small, the pressure applied by the airbag to the muscle is increased. k is a proportional coefficient, P max is the maximum value of the airbag pressure, is the maximum value of the muscle fatigue degree.

[0102] After the airbag adjustment pressure is determined, the airbag adjustment pressure and the current pressure of the airbag collected by the air pressure sensor are taken as the input of the muscle correction controller, the second pressure signal for controlling the airbag of the flexible corrector to inflate and deflate is output by using the muscle correction controller, and the airbag inflates and deflates under the control of the second pressure signal, so that appropriate correction force is applied to the muscle soft tissue during the inflation and deflation process, and the muscle soft tissue is promoted to recover.

[0103] In some modes, the muscle correction controller is implemented based on a second fuzzy PID controller, and the airbag of the flexible corrector is feedback controlled by using the second fuzzy PID controller. The second fuzzy PID controller includes a second PID controller and a second fuzzy controller, the pressure error between the airbag adjustment pressure and the current pressure collected by the air pressure sensor and the change rate of the pressure error are taken as the input of the second fuzzy controller, and the second control parameters K I , K D , and K P are output; the second PID controller takes the pressure error as the input, and outputs the second pressure signal according to the second control parameters K I , K D , and K P .

[0104] In some embodiments, during the recovery of the affected limb bone, the bone hard tissue and muscle soft tissue are constantly changing. In order to promote the effective coordinated recovery of the muscle and bone, the control parameters of the adaptive impedance controller, i.e. the damping parameter B and the stiffness parameter K in formula (4), need to be dynamically adjusted at different stages of muscle and bone recovery, so that the auxiliary correction robot can apply the most suitable coordinated correction force to the muscle and bone at different stages.

[0105] Therefore, in order to adjust the appropriate damping parameter, the muscle and bone coordinated correction method further comprises:

[0106] determining the predicted damping parameter according to the predicted muscle fatigue degree, the preset maximum fatigue degree and the preset maximum damping parameter;

[0107] determining the target damping parameter according to the relationship between the predicted damping parameter and the preset minimum damping parameter and the preset maximum damping parameter;

[0108] adjusting the preset damping parameter according to the target damping parameter.

[0109] In this embodiment, for the damping parameter, the muscle fatigue degree predicted by the muscle fatigue degree prediction model the preset maximum fatigue degree and the preset maximum damping parameter B hb determining the predicted damping parameter B1, and then determining the target damping parameter B according to the size relationship between the predicted damping parameter B1 and the preset minimum damping parameter B lb and the preset maximum damping parameter B hb determining the control parameters of the adaptive impedance controller according to the target damping parameter, and determining the correction force suitable for the coordinated recovery of the muscle and bone at the current stage by dynamically adjusting the damping parameter.

[0110] In some ways, the method for determining the target damping parameter is:

[0111]

[0112]

[0113] The damping parameter B is adjusted according to the muscle fatigue degree. When the muscle fatigue degree deepens, the damping parameter is increased, the elasticity of the system is increased, and the muscle fatigue recovery is helped.

[0114] On the other hand, in order to adjust the appropriate stiffness parameter, the muscle and bone coordinated correction method further comprises:

[0115] processing the electromyographic signal to obtain an electromyographic feature;

[0116] using the electromyographic feature as input and using a pre-constructed muscle force prediction model to output a predicted muscle force;

[0117] determining a predicted stiffness parameter according to the predicted muscle force, a preset maximum muscle force and a preset maximum stiffness parameter;

[0118] determining a target stiffness parameter according to a relationship between the predicted stiffness parameter and a preset minimum stiffness parameter and a preset maximum stiffness parameter;

[0119] adjusting the preset stiffness parameter according to the target stiffness parameter.

[0120] In the embodiment, a muscle force prediction model is constructed, and the muscle force prediction model is used to predict muscle force according to muscle electrical characteristics of muscle soft tissue. First, an electromyography (EMG) signal acquisition instrument is used to collect EMG signals (surface EMG signals) of muscle soft tissue, and then the EMG signals are processed to obtain EMG characteristics (surface EMG signal characteristics) suitable for processing of the muscle force prediction model. The EMG characteristics are input into the muscle force prediction model, and the muscle force prediction model outputs predicted muscle force. Then, the predicted muscle force is used to determine a stiffness parameter that needs to be adjusted.

[0121] In some ways, the EMG characteristics obtained by processing the EMG signals include a root mean square (RMS) value characteristic, a zero-crossing point (ZC) number characteristic and an integrated EMG (IEMG) value characteristic, which can reflect the degree of muscle activity, wherein:

[0122] The RMS value characteristic RMS is used to represent the discharge amount of a predetermined periodic motor neuron, and is expressed as:

[0123]

[0124] wherein EMG(t) is a time-domain amplitude variation curve of the EMG signal, T is a selected period, and t is a starting time of the period.

[0125] The IEMG value characteristic IEMG is used to represent the strength of muscle activity in a predetermined period, and is expressed as:

[0126]

[0127] The ZC number characteristic ZC is used to represent the speed of muscle contraction, and is expressed as:

[0128]

[0129]

[0130] wherein x i and x i+1 are the i-th EMG signal and the i+1-th EMG signal, i.e., two consecutive EMG signals, and N is the number of EMG signal sampling points in a predetermined period.

[0131] For example, Figure 11As shown, in some embodiments, the muscle force prediction model is implemented based on a GD-FNN model. The model uses a Gaussian function as the membership function, and the output parameter range is 0-1 for optimal use. To improve the generalization ability of the model, after processing the collected electromyographic signals to obtain electromyographic features, the electromyographic features need to be normalized, and then the normalized electromyographic features are input into the muscle force prediction model. After the model outputs the predicted muscle force, the output muscle force is denormalized to obtain the final prediction result.

[0132] The method for normalizing the electromyographic features is:

[0133]

[0134] In the formula, X(t) is any one of the root mean square value feature, the zero-crossing point feature, and the integral electromyographic value feature, and x(t) is the corresponding feature after normalization.

[0135] The method for denormalizing the predicted muscle force is:

[0136] Y(t) = y(t) [maxX(t) - minX(t)] + minX(t) (21)

[0137] y(t) is the muscle force output by the model, and Y(t) is the muscle force after denormalization.

[0138] In some ways, the GD-FNN model is trained using the constructed training data set, and the muscle force prediction model is obtained after training. The training data set includes two parts. One part is the input data of the model, which is obtained by processing the collected electromyographic signals to obtain electromyographic features X`(t), normalizing the electromyographic features, and taking the normalized electromyographic features x`(t) as the input data. The other part is the output data, which is obtained by obtaining the muscle force Y`(t) through the musculoskeletal model based on physical modeling, normalizing the obtained muscle force, and taking the normalized muscle force y`(t) as the output data.

[0139] In some embodiments, the muscle force prediction model includes an input layer, a membership function layer, a planning layer, and an output layer. The input of the input layer is the normalized root mean square value feature, the zero-crossing point feature, and the integral electromyographic value feature.

[0140] Each input variable of the membership function layer has u membership functions u ij The expression of the membership function is:

[0141]

[0142] In the formula, μ ij (xi ) is the jth membership function of input variable x i , r is the number of input variables, c ij and σ ij represent the center and width of the Gaussian function respectively;

[0143] The rule layer corrects the existing rules and the center and width of the membership function through network training, and the output of the jth membership function is:

[0144]

[0145] Each node of the output layer represents an output variable of the weighted sum of input signals:

[0146]

[0147] where y is the predicted muscle force, ω j is the connection weight of the jth membership function, and y is the output of the jth membership function.

[0148] In this embodiment, after the muscle force is predicted by the muscle force prediction model, the predicted muscle force MF, the preset maximum muscle force MF max , and the preset damping parameter maximum value K hb are used to determine the predicted stiffness parameter K1, and then the target stiffness parameter K is determined according to the size relationship between the predicted stiffness parameter K1 and the preset minimum stiffness parameter K lb , the maximum stiffness parameter K hb . The control parameters of the adaptive impedance controller are determined according to the target stiffness parameter, and the correction force suitable for the current stage of muscle-skeleton collaborative recovery is determined by dynamically adjusting the stiffness parameter.

[0149] In some ways, the method for determining the target stiffness parameter is:

[0150]

[0151]

[0152] The stiffness parameter K is adjusted according to the size of the muscle force. When the muscle force increases, the stiffness parameter increases, and the role of the muscle force is played to reduce the pressure on the skeleton.

[0153] The method for correcting musculoskeletal deformity provided by the embodiment of the present application is used for adaptive control of an auxiliary correction robot worn on a diseased limb. On one hand, a skeletal correction controller is used to perform feedback control on a pneumatic component according to a desired pose of skeletal recovery, an interaction force between a variable stiffness external fixator and a musculoskeletal system, an interaction between a skeletal hard tissue and a muscle soft tissue, and a displacement amount of the pneumatic component of the variable stiffness external fixator, so as to control a correction force applied to the skeletal hard tissue by adjusting inflation and deflation of the pneumatic component. On the other hand, a muscle correction controller is used to perform feedback control on an air bag according to an air bag adjustment pressure and a current pressure, so as to control a correction force applied to the muscle soft tissue by adjusting inflation and deflation of the air bag. On the other hand, in a rehabilitation process, a damping parameter and a stiffness parameter of the skeletal correction controller are dynamically adjusted, and a collaborative correction force applied to the musculoskeletal system is dynamically adjusted, so as to promote collaborative rehabilitation of the skeletal hard tissue and the muscle soft tissue.

[0154] It should be noted that the method of the embodiment of the present application can be executed by a single device, such as a computer or a server. The method of the embodiment can also be applied to a distributed scenario, and completed by multiple devices in cooperation. In the case of such a distributed scenario, one of the multiple devices can only execute one or more steps in the method of the embodiment of the present application, and the multiple devices can interact with each other to complete the method.

[0155] It should be noted that the above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than the order in which they are recited and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

[0156] As shown in Figure 13 The embodiment of the present application also provides a musculoskeletal collaborative deformity correction robot, which comprises:

[0157] The skeletal auxiliary rehabilitation module is used for taking a desired pose of skeletal correction, a change rate of the desired pose, an actual displacement amount of the pneumatic component, a preset stiffness parameter and a damping parameter, and a torque signal collected by a torque sensor as inputs, and outputting, by a skeletal correction controller, a first pressure signal for controlling inflation and deflation of the pneumatic component, so that the pneumatic component applies a correction force to the skeletal hard tissue through inflation and deflation;

[0158] The fatigue degree prediction module is used for taking the collected electromyographic signal as input, and outputting a predicted muscle fatigue degree by using a pre-constructed muscle fatigue degree prediction model;

[0159] The pressure adjustment determining module is configured to determine the air bag adjustment pressure according to the predicted muscle fatigue degree.

[0160] The muscle auxiliary rehabilitation module is configured to take the air bag adjustment pressure and the current pressure of the air bag as inputs of the muscle correction controller, and output a second pressure signal for controlling air charging and discharging of the air bag from the muscle correction controller, so that the air bag applies a correction force to the muscle soft tissue through air charging and discharging.

[0161] For the convenience of description, the above apparatus is described in various modules in terms of functions. Of course, the functions of the modules can be implemented in one or more software and / or hardware when implementing the embodiments of the present application.

[0162] The apparatus of the above embodiments is used to implement the corresponding method in the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be described here.

[0163] Figure 14 A more specific electronic device hardware structure schematic diagram provided by the embodiment is shown, which can include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040 and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030 and the communication interface 1040 are connected to each other through the bus 1050 for communication within the device.

[0164] The processor 1010 can be implemented in the form of a general CPU (Central Processing Unit, central processor), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits, etc., for executing related programs to implement the technical solutions provided by the embodiments of the present specification.

[0165] The memory 1020 can be implemented in the form of a ROM (Read Only Memory, read-only memory), a RAM (Random Access Memory, random access memory), a static storage device, a dynamic storage device, etc. The memory 1020 can store an operating system and other application programs, and when the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the related program codes are saved in the memory 1020 and called and executed by the processor 1010.

[0166] The input / output interface 1030 is configured to connect an input / output module to realize information input and output. The input / output module can be configured as a component in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. The input device can include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device can include a display, a speaker, a vibrator, an indicator light, etc.

[0167] The communication interface 1040 is configured to connect a communication module (not shown in the figure) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired manner (such as a USB, a network cable, etc.) or a wireless manner (such as a mobile network, WIFI, Bluetooth, etc.).

[0168] The bus 1050 includes a channel to transmit information between various components (such as the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040) of the device.

[0169] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in the specific implementation process, the device can also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device can also only contain components necessary for implementing the embodiments of the present specification, and does not necessarily contain all the components shown in the figure.

[0170] The electronic device of the above embodiment is used to realize the corresponding method in the above embodiment, and has the beneficial effects of the corresponding method embodiment, which will not be described here.

[0171] The computer readable medium of the present embodiment includes permanent and non-permanent, removable and non-removable media, which can be realized by any method or technology to store information. The information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0172] Those skilled in the art should understand that the above discussion of any embodiment is only intended to be exemplary in nature and is not intended to suggest any limitation as to the scope of the disclosure including the claims. Alternate embodiments of both known and / or novel configurations can implement the same or similar techniques in a manner consistent with the disclosure. In other words, any of the embodiments of the present disclosure (including all features and / or embodiments thereof) can be altered or modified and can function within a system in a manner consistent with the disclosure. Furthermore, no one feature or component of an embodiment should be a disclaimer of either the whole of any part of the disclosure; the embodiment should be read to function in conjunction with other embodiments or in place of there embodiments. Accordingly, the disclosure is not intended to be limited to the examples described herein but rather is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0173] In addition, to simplify the description and discussion, and so as not to make the embodiments of the present application difficult to understand, the well-known power / ground connections of integrated circuit (IC) chips and other components can or can not be shown in the provided drawings. Furthermore, the apparatuses can be shown in block diagram form in order to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details regarding the implementation of these block diagram apparatuses are highly dependent on the platform to be implemented (i.e., these details should be well within the understanding of those skilled in the art). Where specific details (e.g., circuitry) are set forth in order to describe an illustrative embodiment of the disclosure, it should be apparent to those skilled in the art that the embodiments of the present application can be practiced without or with variation of these specific details. Thus, the description should not be viewed as limiting and should be understood to be illustrative only. The description is presented for the purpose of simplifying the present application and making it easier to understand.

[0174] Although the present disclosure has been described in conjunction with the specific embodiments thereof, many alternatives, modifications and variations will be apparent to those skilled in the art in light of the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) can use the embodiments discussed.

[0175] The embodiments of the present application are intended to cover all such alternatives, modifications and variations as falling within the scope of the broadest possible interpretation of the appended claims. Accordingly, any and all such alternatives, modifications and variations should be included within the scope of the present disclosure.

Claims

1. A musculoskeletal coordinated deformity correction robot, used to control an auxiliary correction robot, characterized in that, The assistive correction robot includes a variable stiffness external fixator for correcting hard bone tissue and a flexible corrector for correcting soft muscle tissue. The variable stiffness external fixator is equipped with pneumatic components and a torque sensor, and the flexible corrector is equipped with an airbag. The robot includes: The skeletal-assisted rehabilitation module is used to take the desired posture of skeletal correction, the rate of change of the desired posture, the actual displacement of the pneumatic component, the preset stiffness parameters and damping parameters, and the torque signal collected by the torque sensor as inputs, and output a first pressure signal from the skeletal correction controller to control the inflation and deflation of the pneumatic component, so that the pneumatic component applies a corrective force to the hard tissue of the bone through inflation and deflation. The fatigue prediction module is used to take the collected electromyographic signals as input and output the predicted muscle fatigue using a pre-built muscle fatigue prediction model. The regulating pressure determination module is used to determine the airbag regulating pressure based on the relationship between the predicted muscle fatigue level and the airbag pressure. The muscle-assisted rehabilitation module is used to take the airbag adjustment pressure and the current pressure of the airbag as inputs to the muscle correction controller, and the muscle correction controller outputs a second pressure signal to control the inflation and deflation of the airbag, so that the airbag applies a corrective force to the muscle soft tissue through inflation and deflation. The skeletal correction controller includes an adaptive impedance controller and a first fuzzy PID controller. The adaptive impedance controller is used to determine the control parameters of the first fuzzy PID controller based on the desired pose, the rate of change of the desired pose, the stiffness parameter, the damping parameter, the torque signal, and the actual displacement of the pneumatic component. The first fuzzy PID controller is used to output the first pressure signal based on the control parameters. The mathematical model of the adaptive impedance controller is as follows: （4） in, For the target velocity, q d For the desired pose, Let be the rate of change of the desired pose, q be the actual pose of the pneumatic component, K be the stiffness parameter, and B be the damping parameter. The torque signal is defined as follows: the actual pose of the pneumatic component is determined based on the actual displacement of the pneumatic component using the forward kinematics method.

2. The robot according to claim 1, characterized in that, The input to the first fuzzy PID controller is the displacement error between the displacement of the pneumatic component determined based on the control parameters and the actual displacement of the pneumatic component, and the output of the first fuzzy PID controller is the first pressure signal.

3. The robot according to claim 2, characterized in that, The actual pose of the pneumatic component is determined based on the actual displacement of the pneumatic component using forward kinematics; the displacement of the pneumatic component is determined based on the control parameters by calculating the displacement of the pneumatic component based on the target pose using inverse kinematics, where the target pose is obtained by integrating the target velocity.

4. The robot according to claim 1, characterized in that, Also includes: The damping parameter adjustment module is used to determine the predicted damping parameter based on the predicted muscle fatigue, the preset maximum fatigue value, and the preset maximum damping parameter value. The target damping parameter is determined based on the relationship between the predicted damping parameter and the preset minimum and maximum damping parameter values. Adjust the preset damping parameters according to the target damping parameters.

5. The robot according to claim 1, characterized in that, Also includes: The stiffness parameter adjustment module is used to process the electromyographic signals to obtain electromyographic characteristics; Using the electromyographic features as input, the predicted muscle force is output using a pre-built muscle force prediction model; Based on the predicted muscle force, the preset maximum muscle force, and the preset maximum stiffness parameter, a predicted stiffness parameter is determined; based on the relationship between the predicted stiffness parameter, the preset minimum stiffness parameter, and the preset maximum stiffness parameter, a target stiffness parameter is determined; and the preset stiffness parameter is adjusted according to the target stiffness parameter.

6. The robot according to claim 1, characterized in that, Based on the relationship between the predicted muscle fatigue level and the air pressure of the airbag, the airbag adjustment pressure is determined by the following method: （12） Wherein, P is the airbag regulating pressure. The proportionality coefficient k is determined based on the maximum value of the airbag pressure and the maximum value of the muscle fatigue.

7. The robot according to claim 5, characterized in that, The muscle fatigue prediction model is based on an autoregressive model; the muscle strength prediction model is based on a GD-FNN model.

Images