CN116578024A - Multi-mode control method and system for rehabilitation robot based on mixed mode signals - Google Patents
Multi-mode control method and system for rehabilitation robot based on mixed mode signals Download PDFInfo
- Publication number
- CN116578024A CN116578024A CN202310533530.1A CN202310533530A CN116578024A CN 116578024 A CN116578024 A CN 116578024A CN 202310533530 A CN202310533530 A CN 202310533530A CN 116578024 A CN116578024 A CN 116578024A
- Authority
- CN
- China
- Prior art keywords
- rehabilitation
- controller
- mode
- modes
- mixed mode
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000012549 training Methods 0.000 claims abstract description 45
- 230000009471 action Effects 0.000 claims abstract description 38
- 230000008569 process Effects 0.000 claims abstract description 23
- 210000004556 brain Anatomy 0.000 claims abstract description 6
- 230000005611 electricity Effects 0.000 claims abstract description 4
- 238000013528 artificial neural network Methods 0.000 claims description 36
- 238000005070 sampling Methods 0.000 claims description 9
- 230000000737 periodic effect Effects 0.000 claims description 8
- 230000003044 adaptive effect Effects 0.000 claims description 5
- 230000003993 interaction Effects 0.000 claims description 4
- 210000003205 muscle Anatomy 0.000 claims description 4
- 230000005284 excitation Effects 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 210000001364 upper extremity Anatomy 0.000 description 14
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000003909 pattern recognition Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000003542 behavioural effect Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 208000006011 Stroke Diseases 0.000 description 2
- 238000011217 control strategy Methods 0.000 description 2
- 238000002567 electromyography Methods 0.000 description 2
- 210000003414 extremity Anatomy 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 206010008190 Cerebrovascular accident Diseases 0.000 description 1
- 208000012661 Dyskinesia Diseases 0.000 description 1
- 241000282412 Homo Species 0.000 description 1
- 206010061225 Limb injury Diseases 0.000 description 1
- 208000033952 Paralysis flaccid Diseases 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000003169 central nervous system Anatomy 0.000 description 1
- 230000002490 cerebral effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004064 dysfunction Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000010304 firing Methods 0.000 description 1
- 208000028331 flaccid paralysis Diseases 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003183 myoelectrical effect Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000005312 nonlinear dynamic Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000001148 spastic effect Effects 0.000 description 1
- 230000002459 sustained effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B24/00—Electric or electronic controls for exercising apparatus of preceding groups; Controlling or monitoring of exercises, sportive games, training or athletic performances
- A63B24/0087—Electric or electronic controls for exercising apparatus of groups A63B21/00 - A63B23/00, e.g. controlling load
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61H—PHYSICAL THERAPY APPARATUS, e.g. DEVICES FOR LOCATING OR STIMULATING REFLEX POINTS IN THE BODY; ARTIFICIAL RESPIRATION; MASSAGE; BATHING DEVICES FOR SPECIAL THERAPEUTIC OR HYGIENIC PURPOSES OR SPECIFIC PARTS OF THE BODY
- A61H1/00—Apparatus for passive exercising; Vibrating apparatus; Chiropractic devices, e.g. body impacting devices, external devices for briefly extending or aligning unbroken bones
- A61H1/02—Stretching or bending or torsioning apparatus for exercising
- A61H1/0274—Stretching or bending or torsioning apparatus for exercising for the upper limbs
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
- G05B19/0423—Input/output
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61H—PHYSICAL THERAPY APPARATUS, e.g. DEVICES FOR LOCATING OR STIMULATING REFLEX POINTS IN THE BODY; ARTIFICIAL RESPIRATION; MASSAGE; BATHING DEVICES FOR SPECIAL THERAPEUTIC OR HYGIENIC PURPOSES OR SPECIFIC PARTS OF THE BODY
- A61H2201/00—Characteristics of apparatus not provided for in the preceding codes
- A61H2201/01—Constructive details
- A61H2201/0173—Means for preventing injuries
- A61H2201/0176—By stopping operation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61H—PHYSICAL THERAPY APPARATUS, e.g. DEVICES FOR LOCATING OR STIMULATING REFLEX POINTS IN THE BODY; ARTIFICIAL RESPIRATION; MASSAGE; BATHING DEVICES FOR SPECIAL THERAPEUTIC OR HYGIENIC PURPOSES OR SPECIFIC PARTS OF THE BODY
- A61H2201/00—Characteristics of apparatus not provided for in the preceding codes
- A61H2201/50—Control means thereof
- A61H2201/5007—Control means thereof computer controlled
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/25—Pc structure of the system
- G05B2219/25257—Microcontroller
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Physical Education & Sports Medicine (AREA)
- Epidemiology (AREA)
- Pain & Pain Management (AREA)
- Rehabilitation Therapy (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Rehabilitation Tools (AREA)
Abstract
The invention provides a multi-mode control method and a multi-mode control system for a rehabilitation robot based on mixed mode signals, wherein a discrete determination learning theory is adopted to establish an experience-based control library and an identifier library under different rehabilitation modes and rehabilitation actions; and then based on a control theory of the mode, identifying a rehabilitation mode suitable for the rehabilitation grade of the current patient and the rehabilitation action wanted by the patient by identifying various physiological signals such as brain electricity, myoelectricity, human body posture, contact force and the like of the current patient and kinematic parameters, and further selecting a corresponding controller to drive the patient to train. In addition, mixed mode signals of a patient can be analyzed in real time in the training process, and once the intention of the patient is changed, the experience-based control library and the identifier library can work quickly to quickly identify and switch the controllers, so that the performance of the rehabilitation training system is improved.
Description
Technical Field
The invention belongs to the technical field of rehabilitation robot control, and relates to a multi-mode control method and system for a rehabilitation robot based on mixed mode signals.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Two-thirds of patients after stroke have different degrees of dyskinesias, which severely affect their daily lives. The rehabilitation training can help the patient to fully repeat the movement and restore the movement function. Aiming at the defects of large manpower consumption, poor repeatability and the like of the traditional rehabilitation, the rehabilitation robot can provide concentrated and repeated training for patients. And clinical results prove that: the robot auxiliary rehabilitation training has positive effect on restoring the movement function.
Rehabilitation of cerebral apoplexy patients is a long process, and can be roughly divided into a flaccid paralysis period, a spastic period and a recovery period, and the rehabilitation training modes required by each period are different. Rehabilitation robots are expected to possess humanoid intelligent control capabilities, i.e. to learn and recognize different patient conditions and to take accurate control strategies based on experience like humans. Inspired by the human learning and control ideas, a combined idea of pattern learning, pattern recognition and pattern control, i.e. pattern-based control, is proposed. Learning in a dynamic environment is regarded as the most difficult problem in the fields of adaptive and learning control, and conventional control methods cannot achieve learning at all or have limited learning effects. Moreover, existing pattern recognition based on statistical decisions is based on some known pattern characteristics and is not a control performance, and thus cannot be directly applied in pattern recognition and control.
Currently, rehabilitation robots for upper limbs have developed various rehabilitation modes, such as a passive mode, a resistance mode, an auxiliary mode, and the like, and the rehabilitation training actions involved are also various. Accurate identification of patient movement intention is key to mode-based rehabilitation training implementation and ensuring safety and accuracy of human-computer interaction. However, physiological information (such as brain electricity, myoelectricity, etc.) or behavioral information (such as body posture, contact force, etc.) of a single modality cannot fully reflect the movement intention, and particularly in a patient group with central nervous system or peripheral limb injury, due to dysfunction, the physiological information or the behavioral information of a single source cannot accurately provide the basis for distinguishing the movement intention. Therefore, development of a rehabilitation robot system driven by various physiological information or behavioral information is urgently needed, and comprehensive and accurate judgment of the motion intention of a patient by the system is improved.
Disclosure of Invention
In order to solve the problems, the invention provides a multi-mode control method and a multi-mode control system for a rehabilitation robot based on mixed mode signals, wherein a discrete determination learning theory is adopted to establish an experience-based control library and an identifier library under different rehabilitation modes and rehabilitation actions; and then based on a control theory of the mode, identifying a rehabilitation mode suitable for the rehabilitation grade of the current patient and the rehabilitation action wanted by the patient by identifying various physiological signals such as brain electricity, myoelectricity, human body posture, contact force and the like of the current patient and kinematic parameters, and further selecting a corresponding controller to drive the patient to train. In addition, mixed mode signals of a patient can be analyzed in real time in the training process, and once the intention of the patient is changed, the experience-based control library and the identifier library can work quickly to quickly identify and switch the controllers, so that the performance of the rehabilitation training system is improved.
According to some embodiments, the present invention employs the following technical solutions:
a multi-mode control method of a rehabilitation robot based on mixed mode signals comprises the following steps:
adopting a discrete determination learning theory to establish an experience-based controller library under different rehabilitation modes and rehabilitation actions;
establishing an experience-based identifier library under all rehabilitation modes and rehabilitation actions by adopting a discrete determination learning theory;
according to the current mixed mode signal, the kinematic parameters are combined with an experience-based identifier library to be compared, and a rehabilitation mode suitable for the rehabilitation grade of the current patient and a rehabilitation action wanted by the patient are identified;
and selecting a corresponding controller from the controller library according to the identified rehabilitation mode and rehabilitation action, and controlling the rehabilitation robot to execute corresponding rehabilitation training.
As an alternative embodiment, the method further comprises the steps of:
in the training process, the method is characterized in that the method is used for identifying according to the mixed mode signals and the kinematic parameters, judging whether the current training intention is changed or not, and if so, identifying the rehabilitation mode and the rehabilitation action which are suitable for the rehabilitation grade of the current patient, and switching from a controller library to a matched controller.
As an alternative embodiment, using discrete deterministic learning theory, the specific process of building an experience-based controller library for different rehabilitation patterns and rehabilitation actions includes:
converting the discrete rehabilitation robot dynamic model into a dynamic model in a standard system form with output feedback;
designing a discrete self-adaptive neural network controller by utilizing a discrete determination theory;
determining a weight updating law of the discrete self-adaptive neural network controller according to a Lyapunov stability theory and a sampling determination learning theory;
based on the discrete self-adaptive neural network controller and the weight updating law, learning the unknown dynamics of the system along the periodic track, and constructing an experience-based controller library under different training modes by using the learned knowledge.
Further, based on the discrete adaptive neural network controller and the weight update law, learning the unknown dynamics of the system along the periodic track, and constructing an experience-based controller library under different training modes by using the learned knowledge, wherein when neurons of the discrete adaptive neural network controller along the periodic track meet the continuous excitation condition, the state error and the neural network weight are bounded and exponentially converged, storing the neural network weight after stable convergence in the form of a constant neural network, and then constructing an experienced controller for different training modes by using the learned knowledge;
and optimizing the experienced controller according to the characteristics of different modes to form a controller library.
Further, according to the characteristics of different modes, the specific process of optimizing the experienced controller comprises the following steps:
a passive mode employing the experienced controller;
an active mode, wherein admittance control is added on the experienced controller, and the parameters are changed to adjust the flexibility in the human-computer interaction process;
and a stopping mode, wherein a threshold control is added to the experienced controller, and when the muscle strength is detected to be greater than a set threshold, the control movement of the controller is forcibly stopped.
As an alternative implementation mode, the discrete determination learning theory is adopted, and the specific process for establishing the experience-based identifier library under all rehabilitation modes and rehabilitation actions comprises the following steps:
based on a discrete rehabilitation robot dynamics model, building a dynamic neural network identifier for modeling the intrinsic dynamics information of the subsystem under different training modes;
updating the weight of the identifier and proving the system stability by using the Lyapunov theory;
the index stability principle in the learning theory is determined through sampling, so that the neural network can be enabled to converge to a true value in an index, and the accurate modeling of the intrinsic dynamics information of the subsystem is ensured;
and constructing an experience-based identifier library under all rehabilitation modes and rehabilitation actions by utilizing the converged constant neural network.
As an alternative embodiment, when the controller library based on experience under different rehabilitation modes and rehabilitation actions is established, mixed mode signals in the process of each rehabilitation action under each rehabilitation mode are acquired, the mixed mode signals are identified, and corresponding relations between the different mixed mode signals and the different rehabilitation modes and between the different rehabilitation mode signals and the different rehabilitation actions are established.
As an alternative embodiment, the mixed mode signals include brain electrical, myoelectrical, body posture, contact force signals, and the like.
As an alternative implementation manner, according to the current mixed mode signal, the kinematic parameters are combined with the experience-based identifier library to be compared, the corresponding rehabilitation mode is determined, and then the corresponding controller is selected to drive the patient to move along with the rehabilitation robot.
A multi-mode control system for a rehabilitation robot based on mixed mode signals, comprising:
the construction module is configured to adopt a discrete determination learning theory to establish an experience-based controller library under different rehabilitation modes and rehabilitation actions and an experience-based identifier library under all rehabilitation modes and rehabilitation actions;
the identification module is configured to identify according to the current mixed mode signal and the combination of the kinematic parameters, and identify the rehabilitation mode suitable for the rehabilitation grade of the current patient;
and the control module is configured to select a corresponding controller from the controller library according to the identified rehabilitation mode and control the rehabilitation robot to execute the training of the corresponding rehabilitation action.
As an alternative embodiment, the control module is further configured to identify according to the mixed mode signal in combination with the kinematic parameters, determine whether the current training intention is changed, and if so, identify a rehabilitation mode and a rehabilitation action suitable for the current patient rehabilitation level, and switch from the controller library to the matched controller.
A terminal device comprising a processor and a computer readable storage medium, the processor configured to implement instructions; the computer readable storage medium is for storing a plurality of instructions adapted to be loaded by a processor and to perform the steps in the method.
Compared with the prior art, the invention has the beneficial effects that:
the invention establishes two libraries based on sampling to determine the learning theory, which are respectively experience-based controller libraries for controlling robots in different actions and modes to move according to task requirements; and comparing the newly generated multi-mode signal characteristics with information in the library to obtain actions or modes corresponding to the signal characteristics with highest similarity based on the experience identifier library, so as to switch. The intelligent control strategy of a person is simulated, an experience-based controller library and an experience-based identifier library in different modes are established, and then the mode is rapidly identified and the controller is rapidly switched based on the rehabilitation level of the patient and the rehabilitation willingness of the patient, so that the system performance is greatly improved.
The control system provided by the invention utilizes the mixed mode physiological signals of the patient to carry out the intention recognition and rehabilitation grade evaluation of the patient, and utilizes the definite learning theory to carry out the experience-based rapid mode recognition, thereby greatly improving the recognition precision and speed.
The control method fully considers the discrete time characteristic of the actual rehabilitation robot, constructs an experience-based discrete learning controller for the passive mode to realize accurate rehabilitation in the passive training of the patient, and adds admittance control in the experience-based discrete learning controller to improve the flexibility of the active mode process so as to excite the active movement intention of the patient.
According to the invention, the controller based on experience and the identifier based on experience can be used for locally and accurately modeling unknown dynamic and unpredictable interference in a nonlinear system, the learned experience knowledge is stored in a constant neural network form, and then the same or similar control tasks can directly call the stored knowledge for control or identification, so that the controller parameters are not required to be calculated on line, the energy and control time of a rehabilitation training system are saved, and the dynamic control performance is further improved.
In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
Fig. 1 is a block diagram of a multimode switching mechanism of an upper limb rehabilitation robot in the present embodiment.
Fig. 2 is a schematic diagram of the operation of the controller in the passive mode and the active mode of the upper limb rehabilitation robot in the present embodiment.
Fig. 3 is a flowchart of a specific implementation of the multi-mode control method of the upper limb rehabilitation robot based on the mixed-mode physiological signals in the present embodiment.
Detailed Description
The invention will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the present invention. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
The technical scheme of the invention is described by taking an upper limb rehabilitation robot as an application object/control object as an example.
But is not representative of the fact that the present invention can only be used with upper limb rehabilitation robots.
As shown in fig. 3, the method comprises the following steps:
acquiring mixed mode signals in different modes in advance, constructing a mode-mixed mode signal relation, and storing the mode-mixed mode signal relation;
and (3) establishing an experience-based controller library under different rehabilitation modes and rehabilitation actions by adopting a discrete determination learning theory.
And (3) establishing an experience-based identifier library under all rehabilitation modes and rehabilitation actions by adopting a discrete determination learning theory.
After the intervention of the patient, the corresponding modal signals and the corresponding kinematic signal detection equipment are provided for the patient, then the detected multi-modal signals are compared with the information in the identifier library, the rehabilitation mode which is most in line with the rehabilitation level of the current patient and the rehabilitation action which is most in line with the rehabilitation willingness of the current patient are selected through the principle of the minimum residual error through identification and pattern recognition, and the corresponding experience-based controller is selected to drive the patient to carry out rehabilitation training.
The multi-mode signal acquisition of the invention can adopt the prior art, such as the acquisition of brain electrical signals by wearing an electroencephalogram cap, the acquisition of electromyography signals by wearing an electromyography instrument, the acquisition of human body gestures by using an image acquisition device, the acquisition of contact force signals by using a force sensor and the like.
In some embodiments, to ensure constant adaptation of the rehabilitation training, the mixed mode signal of the patient is still detected in real time during the rehabilitation training, and once the movement intention of the patient changes or the rehabilitation level of the patient is improved along with the progress of the rehabilitation training, the dynamic mode recognition mechanism based on the determination learning can be rapidly recognized and switched to the corresponding controller in a very short time. The multi-mode control method based on the mixed mode signals fully considers the rehabilitation level of the patient and the rehabilitation willingness of the patient, and can greatly improve the rehabilitation efficiency of the patient.
The following describes the specific implementation technical scheme in detail.
As shown in fig. 1, the process of building and training and learning the controller specifically includes:
step (1): designing a discrete self-adaptive neural network according to a discrete upper limb rehabilitation robot dynamic model and a discrete determination theory;
according to the following discrete time dynamic model of the upper limb rehabilitation robot:
the dynamic model form converted into the canonical system form with output feedback is as follows:
wherein k is a sampling time point, and T is a sampling time interval; q (k), v (k) represent joint position and joint velocity, respectively,m (q (k)) and->Is an inertial matrix; f (q (k), v (k)) is the Coriolis-centrifugal force and gravity moment matrix; τ (k) is the control input torque;
x(k)＝[q(k),v(k)] T ，τ d (k) Is an external force moment;
the discrete time self-adaptive neural network controller is properly designed based on a discrete upper limb rehabilitation robot dynamics model and sampling determination learning, so as to accurately identify (learn) the unknown dynamics of the upper limb rehabilitation robot in the tracking control process, specifically:
wherein c 2 The gain constant of the controller is designed to meet 0-1 Tc 2 ≤1，e 2 As a tracking error of the velocity of the joint,output for radial basis function neural network (Radial basis function neural network, RBFNN), ++>Is the weight of RBFNN, S is the regression vector of RBFNN, < ->For RBFNN input, alpha 1 (k) The virtual controller is specifically:
in the formula e 1 C is the tracking error of the joint position 1 The gain constant of the controller is designed to meet 0-1 Tc 1 ≤1，x d Is a reference track (a track corrected by admittance control in the active mode).
According to Lyapunov stability theory and sampling, determining a learning theory to design a neural network weight updating law, wherein the method comprises the following steps:
wherein Γ=Γ T >0 is a positive diagonal matrix, σ>0 is a constant.
Step (2): learning the unknown dynamics of the system along the periodic track, and constructing an upper limb rehabilitation robot controller library under different training modes by using the learned knowledge;
when the neurons of the RBFNN along the periodic trajectory meet the sustained firing (persistent excitation, PE) condition, the state error is bounded by the neural network weights and converges exponentially. The neural network weight after stable convergence is obtainedWith constant neural network->Is stored and then uses the learned knowledge to construct an experienced controller for different training patterns, specifically:
wherein, the liquid crystal display device comprises a liquid crystal display device,the method comprises the following steps:
in the formula, [ k ] a ,k b ]The time interval after the system is stably converged is set.
Step (3): accurately identifying subsystem dynamics in different training modes, and constructing an identifier library by using learned knowledge;
aiming at a discrete nonlinear dynamics system v (k+1) =v (k) +T [ f (x) +g (x) tau (k) ] of the upper limb rehabilitation robot shown in a formula (2), modeling the intrinsic dynamics information of subsystems under different training modes by building a dynamic neural network identifier, wherein the method specifically comprises the following steps:
in the formula, s represents a training mode,for the state of the ith dynamic neural network identifier, β i For identifier gain, ++>Weight to be estimated for RBFNN in identifier, ++>Is the regression vector of RBFNN in identifier, < >>To approximate the subsystem unknown dynamics f (x) +g (x) τ (k).
Neural network weight update law given based on Lyapunov design:
wherein gamma represents the neural network weightLearning gain, z of value update law i Representing tracking errors.
By adopting the dynamic neural network identifier and the weight updating law, the intrinsic dynamics information of the subsystem under different training modes can be accurately identified locally by the constant RBFNN:
in the method, in the process of the invention,for the state of the ith discrete dynamic identifier in the s-th training mode, +.>Is a constant neural network.
Step (4): rapidly identifying a change in the control state using the estimator;
the data of the test patterns is compared with discrete dynamic identifiers in an estimation library, resulting in synchronization errors (or recognition errors) corresponding to the different training patterns.
To increase the effectiveness of the recognition process, the average L1 norm is used to make decisions:
wherein T is e Is a predetermined range of calculated average L1 norms.
Step (5): and switching the corresponding controller according to the principle of the minimum average L1 norm.
Another important part is the design and choice of controller for different rehabilitation modes.
As shown in fig. 2, the rehabilitation mode can be broadly divided into a passive mode, an active mode, and a stop mode. In the passive mode, the accurate track tracking training is mainly performed by using the discrete learning controller based on experience and shown in the formula (6), the device is suitable for early rehabilitation, the patient has no muscle strength, and the patient mainly depends on an upper limb rehabilitation robot to drive the affected limb to move. The active mode is divided into an auxiliary mode and a resistance mode, is suitable for middle and late stages of rehabilitation, and has certain active movement capability for the affected limbs of a patient. The assist mode provides assistance when the patient is not in motion, and the resistance mode provides resistance when the patient is in motion.
In the embodiment, the controller in the active mode adds admittance control based on the formula (6), so that the switching from the passive mode to the active mode is realized, and the flexibility in the man-machine interaction process is adjusted by changing the admittance parameter of the system.
The stopping mode is mainly used for guaranteeing the safety of a patient, and when abnormal behavior of the patient is detected, namely, the muscle strength is greater than the force range required by training, the movement of the upper limb rehabilitation robot is forcefully stopped.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
While the foregoing description of the embodiments of the present invention has been presented in conjunction with the drawings, it should be understood that it is not intended to limit the scope of the invention, but rather, it is intended to cover all modifications or variations within the scope of the invention as defined by the claims of the present invention.
Claims (10)
1. The multi-mode control method for the rehabilitation robot based on the mixed mode signals is characterized by comprising the following steps of:
adopting a discrete determination learning theory to establish an experience-based controller library under different rehabilitation modes and rehabilitation actions;
establishing an experience-based identifier library under all rehabilitation modes and rehabilitation actions by adopting a discrete determination learning theory;
identifying a rehabilitation mode suitable for the rehabilitation grade of the current patient according to the current mixed mode signal and by combining with the kinematic parameters;
and selecting a corresponding controller from the controller library according to the identified rehabilitation mode, and controlling the rehabilitation robot to execute the training of the corresponding rehabilitation action.
2. The multi-mode control method for a rehabilitation robot based on mixed mode signals according to claim 1, further comprising the steps of:
in the training process, the mixed mode signals are combined with the kinematic parameters to identify, whether the current training intention is changed or not is judged, if so, the rehabilitation mode suitable for the rehabilitation grade of the current patient is identified, and the controller is switched to a matched controller from a controller library.
3. The multi-mode control method of a rehabilitation robot based on mixed mode signals according to claim 1 or 2, wherein the specific process of establishing the experience-based controller library under different rehabilitation modes and rehabilitation actions by adopting a discrete determination learning theory comprises the following steps:
converting the discrete rehabilitation robot dynamic model into a dynamic model in a standard system form with output feedback;
designing a discrete self-adaptive neural network controller by utilizing a discrete determination theory;
determining a weight updating law of the discrete self-adaptive neural network controller according to a Lyapunov stability theory and a sampling determination learning theory;
based on the discrete self-adaptive neural network controller and the weight updating law, the unknown dynamics of the system is learned along the periodic track, and the learned knowledge is used for constructing a controller library under different training modes.
4. The multi-mode control method of a rehabilitation robot based on mixed mode signals according to claim 3, wherein the specific process of constructing a controller library in different training modes by using learned knowledge based on the discrete adaptive neural network controller and a weight update law along a periodic trajectory learning system unknown dynamic includes that when neurons of the discrete adaptive neural network controller along the periodic trajectory meet a continuous excitation condition, state errors and neural network weights are bounded and exponentially converged, the neural network weights after stable convergence are stored in the form of a constant neural network, and then an experienced controller is constructed for different training modes by using the learned knowledge;
and optimizing the experienced controller according to the characteristics of different modes to form a controller library.
5. The multi-mode control method of a rehabilitation robot based on mixed mode signals according to claim 4, wherein the specific process of optimizing the experienced controller according to the characteristics of different modes comprises:
a passive mode employing the experienced controller;
an active mode, wherein admittance control is added on the experienced controller, and the parameters are changed to adjust the flexibility in the human-computer interaction process;
and a stopping mode, wherein a threshold control is added to the experienced controller, and when the muscle strength is detected to be greater than a set threshold, the control movement of the controller is forcibly stopped.
6. The multi-mode control method of the rehabilitation robot based on the mixed mode signals, according to claim 1, is characterized in that when an experience-based controller library under different rehabilitation modes and rehabilitation actions is established, mixed mode signals in the process of each rehabilitation action under each rehabilitation mode are acquired, the mixed mode signals are identified, and the corresponding relation between the mixed mode signals and the different rehabilitation modes and the rehabilitation actions is established;
or, the mixed mode signals comprise brain electricity, myoelectricity, human body posture and contact force signals.
7. The multi-mode control method of a rehabilitation robot based on mixed mode signals according to claim 1, wherein the specific process of establishing the experience-based identifier library under all rehabilitation modes and rehabilitation actions by adopting a discrete determination learning theory comprises the following steps:
based on a discrete rehabilitation robot dynamics model, building a dynamic neural network identifier for modeling the intrinsic dynamics information of the subsystem under different training modes;
updating the weight of the identifier and proving the system stability by using the Lyapunov theory;
the index stability principle in the learning theory is determined through sampling, so that the neural network can be enabled to converge to a true value in an index, and the accurate modeling of the intrinsic dynamics information of the subsystem is ensured;
and constructing an experience-based identifier library under all rehabilitation modes and rehabilitation actions by utilizing the converged constant neural network.
8. The multi-mode control method of the rehabilitation robot based on the mixed mode signals, according to the current mixed mode signals, the combination of the kinematic parameters and the experience-based identifier library is compared, the corresponding rehabilitation mode is determined, and the corresponding controller is selected to drive the patient to follow the rehabilitation robot to move.
9. A rehabilitation robot multi-mode control system based on mixed mode signals is characterized by comprising:
the construction module is configured to adopt a discrete determination learning theory to establish an experience-based controller library under different rehabilitation modes and rehabilitation actions and an experience-based identifier library under all rehabilitation modes and rehabilitation actions;
the identification module is configured to identify according to the current mixed mode signal and the combination of the kinematic parameters, and identify the rehabilitation mode suitable for the rehabilitation grade of the current patient;
and the control module is configured to select a corresponding controller from the controller library according to the identified rehabilitation mode and control the rehabilitation robot to execute the training of the corresponding rehabilitation action.
10. The multi-mode control system of a rehabilitation robot based on mixed mode signals according to claim 9, wherein the control module is further configured to identify based on the mixed mode signals in combination with kinematic parameters, determine whether the current training intent is changed, and if so, identify a rehabilitation mode and rehabilitation actions appropriate for the current patient rehabilitation level, and switch from the controller library to the matched controller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310533530.1A CN116578024A (en) | 2023-05-09 | 2023-05-09 | Multi-mode control method and system for rehabilitation robot based on mixed mode signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310533530.1A CN116578024A (en) | 2023-05-09 | 2023-05-09 | Multi-mode control method and system for rehabilitation robot based on mixed mode signals |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116578024A true CN116578024A (en) | 2023-08-11 |
Family
ID=87543752
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310533530.1A Pending CN116578024A (en) | 2023-05-09 | 2023-05-09 | Multi-mode control method and system for rehabilitation robot based on mixed mode signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116578024A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117539153A (en) * | 2023-11-21 | 2024-02-09 | 山东大学 | Upper limb rehabilitation robot self-adaptive control method and system based on definite learning |
-
2023
- 2023-05-09 CN CN202310533530.1A patent/CN116578024A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117539153A (en) * | 2023-11-21 | 2024-02-09 | 山东大学 | Upper limb rehabilitation robot self-adaptive control method and system based on definite learning |
| CN117539153B (en) * | 2023-11-21 | 2024-05-28 | 山东大学 | Upper limb rehabilitation robot self-adaptive control method and system based on definite learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Deng et al. | A learning-based hierarchical control scheme for an exoskeleton robot in human–robot cooperative manipulation | |
| Wang et al. | Controlling object hand-over in human–robot collaboration via natural wearable sensing | |
| US11062617B2 (en) | Training system for autonomous driving control policy | |
| CN111956452A (en) | Control method and device for upper limb rehabilitation robot | |
| CN116578024A (en) | Multi-mode control method and system for rehabilitation robot based on mixed mode signals | |
| Zeng et al. | Encoding multiple sensor data for robotic learning skills from multimodal demonstration | |
| CN108304767A (en) | Human action's intention assessment training method based on more brain area cooperated computings | |
| JP2022546644A (en) | Systems and methods for automatic anomaly detection in mixed human-robot manufacturing processes | |
| Nomm et al. | Monitoring of the human motor functions rehabilitation by neural networks based system with kinect sensor | |
| Anvaripour et al. | Collision detection for human-robot interaction in an industrial setting using force myography and a deep learning approach | |
| Zhang et al. | A fusion-based spiking neural network approach for predicting collaboration request in human-robot collaboration | |
| Liu et al. | A human-simulated fuzzy membrane approach for the joint controller of walking biped robots | |
| CN113478462B (en) | Method and system for controlling intention assimilation of upper limb exoskeleton robot based on surface electromyogram signal | |
| Crowley | Integration and control of reactive visual processes | |
| KR20210069410A (en) | Learning system for robotic object manipulation | |
| JPH04156610A (en) | Automatic recognition device for man action using neural network | |
| CN116604532A (en) | Intelligent control method for upper limb rehabilitation robot | |
| CN114952791A (en) | Control method and device for musculoskeletal robot | |
| Koropouli et al. | Generalization of Force Control Policies from Demonstrations for Constrained Robotic Motion Tasks: A Regression-Based Approach | |
| CN113760099A (en) | Movement intention prediction method and system | |
| KR101606543B1 (en) | HRI Robot System | |
| Li et al. | Guest editorial for special issue on human-centered intelligent robots: issues and challenges | |
| Hirkoawa et al. | Coaching robots: online behavior learning from human subjective feedback | |
| CN117539153B (en) | Upper limb rehabilitation robot self-adaptive control method and system based on definite learning | |
| KR20200083338A (en) | Hand gesture recognition apparatus and method using an artificial neural network learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |