CN115054412A - Intelligent artificial limb foot plate system with touchdown gait perception function based on machine learning - Google Patents
Intelligent artificial limb foot plate system with touchdown gait perception function based on machine learning Download PDFInfo
- Publication number
- CN115054412A CN115054412A CN202210603682.XA CN202210603682A CN115054412A CN 115054412 A CN115054412 A CN 115054412A CN 202210603682 A CN202210603682 A CN 202210603682A CN 115054412 A CN115054412 A CN 115054412A
- Authority
- CN
- China
- Prior art keywords
- foot plate
- artificial limb
- gait
- machine learning
- strain
- 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
- 230000005021 gait Effects 0.000 title claims abstract description 34
- 238000010801 machine learning Methods 0.000 title claims description 20
- 230000006870 function Effects 0.000 title claims description 18
- 230000008447 perception Effects 0.000 title claims description 7
- 238000005259 measurement Methods 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 24
- 238000003062 neural network model Methods 0.000 claims abstract description 8
- 239000000853 adhesive Substances 0.000 claims abstract description 4
- 230000001070 adhesive effect Effects 0.000 claims abstract description 4
- 210000002683 foot Anatomy 0.000 claims description 106
- 210000003414 extremity Anatomy 0.000 claims description 13
- 229920000049 Carbon (fiber) Polymers 0.000 claims description 11
- 239000004917 carbon fiber Substances 0.000 claims description 11
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 claims description 11
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 claims description 4
- 210000000544 articulatio talocruralis Anatomy 0.000 claims description 4
- 229910052744 lithium Inorganic materials 0.000 claims description 4
- 229910000831 Steel Inorganic materials 0.000 claims description 3
- 238000002266 amputation Methods 0.000 claims description 3
- 239000002131 composite material Substances 0.000 claims description 3
- 239000010959 steel Substances 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 2
- 239000000463 material Substances 0.000 claims 1
- 230000001960 triggered effect Effects 0.000 abstract description 3
- 230000003993 interaction Effects 0.000 abstract description 2
- 238000000034 method Methods 0.000 description 12
- 210000003141 lower extremity Anatomy 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000004146 energy storage Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 210000000629 knee joint Anatomy 0.000 description 2
- 238000012805 post-processing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 229920004943 Delrin® Polymers 0.000 description 1
- 206010049565 Muscle fatigue Diseases 0.000 description 1
- 206010033372 Pain and discomfort Diseases 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 210000003423 ankle Anatomy 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000004519 grease Substances 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 230000002232 neuromuscular Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 210000004243 sweat Anatomy 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/60—Artificial legs or feet or parts thereof
- A61F2/66—Feet; Ankle joints
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/68—Operating or control means
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/60—Artificial legs or feet or parts thereof
- A61F2/66—Feet; Ankle joints
- A61F2002/6614—Feet
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/68—Operating or control means
- A61F2002/689—Alarm means, e.g. acoustic
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/68—Operating or control means
- A61F2/70—Operating or control means electrical
- A61F2002/704—Operating or control means electrical computer-controlled, e.g. robotic control
Landscapes
- Health & Medical Sciences (AREA)
- Transplantation (AREA)
- Biomedical Technology (AREA)
- Cardiology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Engineering & Computer Science (AREA)
- Heart & Thoracic Surgery (AREA)
- Vascular Medicine (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Orthopedic Medicine & Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
An artificial limb connecting piece of the intelligent artificial limb foot plate system is arranged above a quadrangular frustum pyramid, the quadrangular frustum pyramid is arranged above the artificial limb foot plate, the contact surface between an upper foot plate and a lower foot plate of the artificial limb foot plate is bonded together by adopting an adhesive and is fixed by 2 bolts, the rear part of the lower foot plate and the upper foot plate form an included angle of 35 degrees, the artificial limb foot plate is provided with a toe separating gap, and the upper foot plate and the lower foot plate are both arc-shaped structures imitating normal foot plates; the strain measurement module is arranged in a cuboid groove on the surface of a prosthetic foot, the main processing module and the charging device are arranged on a prosthetic foot plate, the STM32 single chip microcomputer is arranged inside the main processing module, the strain measurement module is electrically connected with the main processing module, and the charging device supplies power to the STM32 single chip microcomputer; the interaction between the artificial limb foot plate and the ground is sensed through the strain measurement module, so that the gait characteristics are recognized, the ground contact gait stage of a user is recognized in real time through the multilayer neural network model, and when an output value deviates from a predicted value, an alarm device is triggered in time to remind the user of paying attention to safety, and the feedback function similar to a human body sensing function is played.
Description
Technical Field
The invention belongs to the technical field of lower limb artificial limbs and orthotics, and particularly relates to an intelligent artificial limb foot plate system with a touchdown gait perception function based on machine learning.
Background
It is known that amputated patients lose relevant limb functions, the movement mode of the amputated patients is greatly changed, the amputated patients suffer great pain and discomfort, and the psychological states of the amputated patients are also greatly overturned. The lower limb artificial limb refers to a wearable technical device specially developed for compensating the loss of the motion function of the lower limb amputee. Therefore, wearing the lower limb prosthesis, rebuilding the lower limb movement ability of the amputee, recovering the standing and walking ability of the amputee, and is an effective means for improving the life quality of the amputee and returning to the social work activities. The prosthetic foot is an important component of a lower limb prosthesis, and the performance of the prosthetic foot is significant to the movement gait of a lower limb amputee. Literature research has shown that current prosthetic foot plates are broadly divided into: one is the early conventional non-energy storing prosthetic foot plate that does not release energy when the prosthesis is stepped off, including a solid ankle with a fixed heel (SACH), a single axis foot and a multi-axis foot. The non-energy-storing artificial limb foot plate has heavy weight and consumes a large amount of energy in the walking process of a wearer. Secondly, the energy storage foot plate which can release the energy stored in the standing stage when the foot is kicked off, and various energy storage artificial limb foot plate products are provided by the current famous artificial limb companies at home and abroad, for example: the flex foot of the orsoi company consists of a carbon fiber body and a heel spring, allowing compression of the carbon fiber body and spring, with the aim of increasing the energy released during the kick-off period. The energy storage foot plates of Chinese Delrin company are generally designed in a split toe mode, and have the advantages of light weight, small load of a wearer and low energy consumption. The energy storage foot plate is light in weight and high in strength, and saves energy in the walking process of a wearer. However, the artificial foot plates are passive, and cannot provide touchdown information feedback, so that the artificial foot plates are difficult to play a sensory feedback role similar to that of the feet of the human body. Further literature research shows that most of the existing researches mainly focus on intelligent control of knee joints and ankle joints of lower limb prostheses, and the development on artificial limb baseplates with intelligent sensing functional characteristics is lacked, so that good ground contact walking state feedback cannot be provided for lower limb amputees.
Therefore, research on human body movement intention identification is an important direction for the development of current intelligent artificial limbs, and the movement intention prediction of artificial limbs so far is mainly divided into the following two categories according to different collected signal sources: firstly, the intention identification based on mechanical signals, such as an insole with a pressure sensor, can identify the gait by sensing the pressure change in the walking process, but the service life of the insole needs to be improved; the Inertial Measurement Unit (IMU) is placed on the knee joint, consists of an accelerometer, a gyroscope and magnetometers of each axis, is mainly used for detecting linear acceleration and rotation speed, has certain advantages in cost, service life and application range, but needs to obtain stable gait recognition, the sensor is placed in a central position, usually, a plurality of model prediction methods are used for searching the relation between gait intention and original signals in signal post-processing, and the processing process is complicated. Secondly, intention recognition based on biological electric signals, such as electromyographs which can measure electromyographic signals generated by neuromuscular activity, is a measuring method for directly detecting and predicting movement intention recognition, but has higher cost, and needs further improvement and optimization in the aspects of accurate sensor pasting position and complex signal post-processing process.
In summary, no matter the inertial sensor and the pressure sensor are adopted based on mechanical signals, or the intention identification of collecting electromyographic signals based on bioelectric signals is based on bioelectric signals, friction loss can be generated due to the fact that the electromyographic signals are placed on soles and the like, the service life is short, the operation process is complex, meanwhile, the electromyographic signals are weak and unstable signals, in order to collect accurate data, metal electrodes need to be tightly attached to the surface of the skin to be detected in the test, and grease or sweat on the surface of the skin can enable the signals to be easily polluted by motion fake signals and muscle fatigue signals. In addition, the surface electromyogram signal is very sensitive to the position change of the pasted electrode, needs to be calibrated again when a test is prepared each time, is complex in operation process and unstable in signal, and is very inconvenient for an amputee to wear, so that the development of the intelligent artificial limb foot plate system with the touchdown gait perception function is challenging, and a solution is needed urgently.
Disclosure of Invention
In order to solve the technical problems, the invention provides an intelligent artificial limb foot plate system with a touchdown gait sensing function based on machine learning, which is a method for detecting gait characteristics during touchdown by using a strain gauge type sensor.
An intelligent artificial limb foot plate system with a ground contact gait sensing function based on machine learning comprises a foot plate connecting piece, a quadrangular frustum pyramid, an artificial limb foot plate, a cuboid groove, a strain measuring module, a bolt, a general processing module, an STM32 single chip microcomputer, a charging device and an alarm device; the artificial limb connecting piece is arranged above the quadrangular frustum pyramid and is used for connecting an ankle joint; the four-edged platform is arranged above the artificial limb foot plate, the four-edged platform is made of 45# steel, when an amputation patient wears the artificial limb foot plate to walk, the carbon fiber foot plate is caused to deform, the artificial limb foot plate comprises a bolt, an upper foot plate and a lower foot plate, a contact surface between the upper foot plate and the lower foot plate is bonded together by adopting an adhesive and is fixed by 2 bolts, an included angle of 35 degrees is formed between the rear part of the lower foot plate and the upper foot plate, the artificial limb foot plate is provided with a toe separating gap, the upper foot plate and the lower foot plate are made of carbon fiber composite materials, and the upper foot plate and the lower foot plate are both arc-shaped structures imitating normal foot plates;
the surface of the artificial limb foot plate is provided with a cuboid groove, the strain measurement module is arranged in the cuboid groove, the total processing module and the charging device are arranged on the artificial limb foot plate, the STM32 singlechip is arranged in the total processing module, the strain measurement module is electrically connected with the total processing module, the alarm device is arranged on the rear surface of the lower foot plate, and the charging device supplies power to the STM32 singlechip;
the strain measurement module measures in real time to obtain a strain value of the artificial limb foot plate, and the strain value measured by the strain measurement module is input to the total processing module for filtering and calculating two characteristic values of the slope and the mean value of a strain curve; then inputting the gait cycle into an STM32 single chip microcomputer, calculating a corresponding gait cycle by the STM32 single chip microcomputer according to the embedded multilayer neural network model, comparing an output value with a predicted value, and triggering an alarm device to remind a user of safety when the output value is different from the predicted value of the model; the charging device is used for driving the STM32 single chip microcomputer to complete machine learning.
The length-width-height ratio of the cuboid groove is 11: 9: 3, the proportion meets the strength requirement generated by the pressure when a 70kg human body is worn, and meets the requirement that the measured strain value is more sensitive.
The charging device is a lithium battery.
The alarm device is a sound-light alarm.
The STM32 singlechip trains three types of characteristic value data including a strain value acquired by a strain measurement module and a slope and a mean value calculated by a total processing module through a multilayer neural network model in machine learning, and the safety of a user is guaranteed.
The artificial limb foot plate is provided with the toe separating gap, so that the artificial limb foot plate can adapt to the rugged road condition, and the stability and the safety of the artificial limb foot plate are enhanced; the most strain sensitive area of the prosthetic foot plate in the walking process is obtained through finite element analysis software: at the separation part of the upper and lower layers of the foot plates, the distance between the toe separating area and the toe separating area is 5mm, and a length-width-height ratio of 11:9 is formed: 3, the strain is maximum on the premise of meeting the strength requirement of wearing an artificial limb foot plate by a 70kg amputee patient.
The strain measurement module strain gauge type sensor is pasted at an angle of 30 degrees with the vertical direction, and the deformation of the carbon fiber foot plate when the gait of the amputee patient changes can be sensed by only using a single strain gauge type sensor.
The general processing module mainly filters the acquired strain data and extracts three types of characteristic values including the slope and the mean value of a strain curve;
the STM32 single chip microcomputer is a machine learning module embedded into a multilayer neural network, and a strain value measured by the strain module is input, so that the gait cycle of a human body in the walking process can be output.
The specific multilayer neural network algorithm is as follows:
#MLP
from sklearn.neural_network import MLPClassifier
mlp＝MLPClassifier(alpha＝1e5,hidden_layer_sizes＝(30,50),random_state＝20, learning_rate＝"adaptive",learning_rate_init＝0.01)
mlp.fit(X,Y)
score3＝cross_val_score(mlp,X,Y,cv＝10).mean
y _ pred3 cross _ val _ predict (mlp, X, Y, cv 10) # obtains the predicted value
print ('MLP cross validation accuracy of', score3) # classification accuracy.
The alarm device is an audible and visual alarm, if the output value of the STM32 single chip microcomputer deviates from the predicted value, the alarm device is triggered to remind a user that the gait is unnatural at the moment and the user pays attention to safety.
Charging device places beside the STM32 singlechip, and charging device includes an external lithium cell, conveniently gives the power supply of STM32 singlechip, the operation of guarantee gait discernment.
The invention has the beneficial effects that:
1. according to the invention, a single strain gauge type sensor is combined into a cuboid groove with a definite length-width-height ratio, the strain gauge type sensor is adhered to the upper foot plate at an angle of 30 degrees with the vertical direction, and the interaction between the carbon fiber artificial limb foot plate of a user and the ground is sensed through the strain measurement module, so that the gait characteristics are identified.
2. According to the invention, the touchdown gait stage of the user is identified in real time through the multilayer neural network model, and when the output value deviates from the predicted value, the alarm device is triggered in time to remind the user of safety, so that a feedback function similar to a human body perception function is exerted.
Drawings
FIG. 1 is an isometric view of the present invention;
FIG. 2 is a top view of the present invention;
FIG. 3 is a cross-sectional view A-A of FIG. 2;
FIG. 4 is a bottom view of the present invention;
FIG. 5 is a top view of the upper foot plate of the present invention;
FIG. 6 is a cross-sectional view taken along line B-B of FIG. 5;
FIG. 7 is a top view of the lower foot plate of the present invention;
FIG. 8 is a lower footplate side view of the present invention;
FIG. 9 is a system flow diagram of the present invention.
In the drawings, 1-prosthetic foot attachment; 2-a quadrangular frustum pyramid; 3-prosthetic foot plate; 4-a cuboid groove; 5-a strain measurement module; 6-bolt; 7-a total processing module; 8-STM32 single chip microcomputer; 9-a charging module; 10-alarm device, 11-upper foot plate; 12-lower foot plate.
Detailed Description
As shown in fig. 1 to 9, an intelligent artificial limb footboard system with a touchdown gait perception function based on machine learning comprises a footboard connecting piece 1, a quadrangular frustum pyramid 2, an artificial limb footboard 3, a cuboid groove 4, a strain measurement module 5, a bolt 6, a total processing module 7, an STM32 single chip microcomputer 8, a charging device 9 and an alarm device 10; the artificial limb connecting piece 1 is arranged above the quadrangular frustum pyramid 2 and is used for connecting an ankle joint; the four-edged platform 2 is arranged above the artificial limb foot plate 3, the four-edged platform 2 is made of 45# steel, when an amputation patient wears the artificial limb foot plate 3 to walk, the carbon fiber foot plate is caused to deform, the artificial limb foot plate 3 comprises a bolt 6, an upper foot plate 11 and a lower foot plate 12, a contact surface between the upper foot plate 11 and the lower foot plate 12 is bonded together by an adhesive and is fixed by the 2 bolts 6, an included angle of 35 degrees is formed between the rear part of the lower foot plate 12 and an upper foot plate 10, the artificial limb foot plate 3 is provided with a toe separating gap 13, the upper foot plate 11 and the lower foot plate 12 are made of carbon fiber composite materials, and the upper foot plate 11 and the lower foot plate 12 are both arc-shaped structures imitating normal foot plates;
the surface of an artificial limb foot plate 3 is provided with a cuboid groove 4, a strain measurement module 5 is arranged in the cuboid groove 4, a main processing module 7 and a charging device 9 are arranged on the artificial limb foot plate 3, an STM32 single chip microcomputer 8 is arranged in the main processing module 7, the strain measurement module 5 is electrically connected with the main processing module 7, an alarm device 10 is arranged on the rear surface of a lower foot plate 12, and the charging device 9 supplies power to the STM32 single chip microcomputer 8; the strain measurement module 5 measures in real time to obtain a strain value of the artificial limb foot plate 3, the strain value measured by the strain measurement module 5 is input to the total processing module 7, filtering is carried out, and two characteristic values of the slope and the mean value of a strain curve are calculated; then inputting the gait cycle into an STM32 single chip microcomputer 8, calculating a corresponding gait cycle by the STM32 single chip microcomputer 8 according to the embedded multilayer neural network model, comparing an output value with a predicted value, and triggering an alarm device 10 to remind a user of safety when the output value is different from the predicted value of the model; and the charging device 9 is used for driving the STM32 singlechip 8 to complete machine learning.
The length-width-height ratio of the cuboid groove 4 is 11: 9: 3, the proportion meets the strength requirement generated by the pressure when a 70kg human body is worn, and meets the requirement that the measured strain value is more sensitive.
The charging device 9 is a lithium battery.
The alarm device 10 is a sound-light alarm.
The strain measurement module 5 is a strain gauge type sensor which is pasted at an angle of 30 degrees with the vertical direction, and the deformation of the carbon fiber foot plate when the gait of the amputee patient changes can be sensed by only using a single strain gauge type sensor.
The STM32 singlechip 8 trains three types of characteristic value data including a strain value acquired by the strain measurement module 5 and a slope and a mean value calculated by the total processing module 7 through a multilayer neural network model in machine learning, and the safety of a user is guaranteed.
Claims (7)
1. Intelligent artificial limb footboard system that has a gait perception function that contacts to earth based on machine learning, its characterized in that: the artificial limb foot plate comprises a foot plate connecting piece (1), a quadrangular frustum pyramid (2), an artificial limb foot plate (3), a cuboid groove (4), a strain measuring module (5), a bolt (6), a general processing module (7), an STM32 single chip microcomputer (8), a charging device (9) and an alarm device (10); the artificial limb connecting piece (1) is arranged above the quadrangular frustum pyramid (2) and is used for connecting an ankle joint; the quadrangular frustum pyramid (2) is arranged above the artificial limb foot plate (3), when an amputation patient wears the artificial limb foot plate (3) to walk, the carbon fiber foot plate is caused to deform, the artificial limb foot plate (3) comprises a bolt (6), an upper foot plate (11) and a lower foot plate (12), a contact surface between the upper foot plate (11) and the lower foot plate (12) is bonded together by adopting an adhesive and is fixed by 2 bolts (6), an included angle of 35 degrees is formed between the rear part of the lower foot plate (12) and the upper foot plate (11), the artificial limb foot plate (3) is provided with a toe separating gap (13), the upper foot plate (11) and the lower foot plate (12) are made of carbon fiber composite materials, and the upper foot plate (11) and the lower foot plate (12) are both arc-shaped structures imitating normal foot plates;
a cuboid groove (4) is formed in the surface of the artificial limb foot plate (3), a strain measurement module (5) is arranged in the cuboid groove (4), a general processing module (7) and a charging device (9) are arranged on the artificial limb foot plate (3), an STM32 single chip microcomputer (8) is arranged inside the general processing module (7), the strain measurement module (5) is electrically connected with the general processing module (7), an alarm device (10) is arranged on the rear surface of a lower foot plate (12), and the charging device (9) supplies power to the STM32 single chip microcomputer (8);
the strain measurement module (5) measures in real time to obtain a strain value of the artificial limb foot plate (3), the strain value measured by the strain measurement module (5) is input into the total processing module (7), and two characteristic values of the slope and the average value of a strain curve are filtered and calculated; then inputting the gait cycle into an STM32 singlechip (8), calculating a corresponding gait cycle according to the embedded multilayer neural network model, comparing an output value with a predicted value, and triggering an alarm device (10) to remind a user of safety when the output value is different from the predicted value of the model; and the charging device (9) is used for driving the STM32 singlechip (8) to complete machine learning.
2. The machine learning based intelligent prosthetic foot plate system with touchdown gait awareness function according to claim 1, wherein: the material of the quadrangular frustum pyramid (2) is 45# steel.
3. The machine learning based intelligent prosthetic foot plate system with touchdown gait awareness function according to claim 1, wherein: the length, width and height ratio of the cuboid groove (4) is 11: 9: 3.
4. the machine learning based intelligent prosthetic foot plate system with touchdown gait awareness function according to claim 1, wherein: the charging device (9) is a lithium battery.
5. The machine learning based intelligent prosthetic foot plate system with touchdown gait awareness function according to claim 1, wherein: the alarm device (10) is an audible and visual alarm.
6. An intelligent prosthetic footboard system with touchdown gait awareness function based on machine learning according to claim 1, characterized in that: the strain gauge type sensor of the strain measurement module (5) is pasted at an angle of 30 degrees with the vertical direction.
7. An intelligent prosthetic footboard system with touchdown gait awareness function based on machine learning according to claim 1, characterized in that: the STM32 single chip microcomputer (8) trains three types of characteristic value data including a strain value acquired by the strain measurement module (5) and a slope and a mean value calculated by the total processing module (7) through a multilayer neural network model in machine learning, and the safety of a user is guaranteed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210603682.XA CN115054412A (en) | 2022-05-31 | 2022-05-31 | Intelligent artificial limb foot plate system with touchdown gait perception function based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210603682.XA CN115054412A (en) | 2022-05-31 | 2022-05-31 | Intelligent artificial limb foot plate system with touchdown gait perception function based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115054412A true CN115054412A (en) | 2022-09-16 |
Family
ID=83198089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210603682.XA Pending CN115054412A (en) | 2022-05-31 | 2022-05-31 | Intelligent artificial limb foot plate system with touchdown gait perception function based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115054412A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115501014A (en) * | 2022-10-16 | 2022-12-23 | 吉林大学 | Integrated ankle system artificial limb |
| CN117012362A (en) * | 2023-10-07 | 2023-11-07 | 中国康复科学所(中国残联残疾预防与控制研究中心) | Adaptive data identification method, system, equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1513647A (en) * | 2003-08-21 | 2004-07-21 | 中国科学院合肥智能机械研究所 | Humanoid robot foot and foot force information detection method |
| US20060195197A1 (en) * | 2005-02-02 | 2006-08-31 | Clausen Arinbjorn V | Sensing systems and methods for monitoring gait dynamics |
| US20150134081A1 (en) * | 2013-03-15 | 2015-05-14 | Otto Bock Healthcare Gmbh | Hydraulic valve |
| US20160331557A1 (en) * | 2015-05-11 | 2016-11-17 | The Hong Kong Polytechnic University | Exoskeleton Ankle Robot |
| CN110711055A (en) * | 2019-11-07 | 2020-01-21 | 江苏科技大学 | Image sensor intelligence artificial limb leg system based on degree of depth learning |
| CN211485109U (en) * | 2019-09-13 | 2020-09-15 | 湖南嘉米机电科技有限公司 | Ankle foot device with sensor |
| CN113780223A (en) * | 2021-09-09 | 2021-12-10 | 北京信息科技大学 | Gait recognition method and device for artificial limb and storage medium |
-
2022
- 2022-05-31 CN CN202210603682.XA patent/CN115054412A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1513647A (en) * | 2003-08-21 | 2004-07-21 | 中国科学院合肥智能机械研究所 | Humanoid robot foot and foot force information detection method |
| US20060195197A1 (en) * | 2005-02-02 | 2006-08-31 | Clausen Arinbjorn V | Sensing systems and methods for monitoring gait dynamics |
| US20150134081A1 (en) * | 2013-03-15 | 2015-05-14 | Otto Bock Healthcare Gmbh | Hydraulic valve |
| US20160331557A1 (en) * | 2015-05-11 | 2016-11-17 | The Hong Kong Polytechnic University | Exoskeleton Ankle Robot |
| CN211485109U (en) * | 2019-09-13 | 2020-09-15 | 湖南嘉米机电科技有限公司 | Ankle foot device with sensor |
| CN110711055A (en) * | 2019-11-07 | 2020-01-21 | 江苏科技大学 | Image sensor intelligence artificial limb leg system based on degree of depth learning |
| CN113780223A (en) * | 2021-09-09 | 2021-12-10 | 北京信息科技大学 | Gait recognition method and device for artificial limb and storage medium |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115501014A (en) * | 2022-10-16 | 2022-12-23 | 吉林大学 | Integrated ankle system artificial limb |
| CN115501014B (en) * | 2022-10-16 | 2024-04-26 | 吉林大学 | Integrated ankle system artificial limb |
| CN117012362A (en) * | 2023-10-07 | 2023-11-07 | 中国康复科学所(中国残联残疾预防与控制研究中心) | Adaptive data identification method, system, equipment and storage medium |
| CN117012362B (en) * | 2023-10-07 | 2024-01-12 | 中国康复科学所(中国残联残疾预防与控制研究中心) | Adaptive data identification method, system, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115054412A (en) | Intelligent artificial limb foot plate system with touchdown gait perception function based on machine learning | |
| Howell et al. | Kinetic gait analysis using a low-cost insole | |
| Miller et al. | Myoelectric walking mode classification for transtibial amputees | |
| Zheng et al. | A noncontact capacitive sensing system for recognizing locomotion modes of transtibial amputees | |
| US8951211B2 (en) | Prosthetic limb monitoring system | |
| US8623098B2 (en) | Systems and method for volitional control of jointed mechanical devices based on surface electromyography | |
| CN105795571B (en) | A kind of data collecting system and method for ectoskeleton pressure footwear | |
| Alaqtash et al. | Application of wearable sensors for human gait analysis using fuzzy computational algorithm | |
| CN113273999B (en) | Wearable multidimensional gait analysis system and method | |
| CN106419929B (en) | Quantitative detection device for posture balance stability of Parkinson disease patient | |
| Crowe et al. | The influence of walking speed on parameters of gait symmetry determined from ground reaction forces | |
| Liu et al. | Local dynamic stability assessment of motion impaired elderly using electronic textile pants | |
| Tiwari et al. | An infrared sensor-based instrumented shoe for gait events detection on different terrains and transitions | |
| Negi et al. | FSR and IMU sensors-based human gait phase detection and its correlation with EMG signal for different terrain walk | |
| Kim et al. | The effect of powered ankle prostheses on muscle activity during walking | |
| CN108186025A (en) | A kind of amount of exercise sole detection device | |
| Haque et al. | Real time level ground walking vs stair-climbing locomotion mode detection | |
| Jamshidi et al. | A case study on the value of in-socket force measurements in gait monitoring of lower-limb prosthesis users | |
| Lou et al. | IMU-Based Gait Phase Recognition for Stroke Survivors: Preliminary Results | |
| Patterson et al. | Using a shoe mounted tri-axial accelerometer to detect kinematic changes during stiff ankle walking | |
| Jasni | Configuration and analysis of piezoelectric-based in socket sensory system for transfemoral prosthetic gait detection | |
| Aminian | Monitoring human movement with body-fixed sensors and its clinical applications | |
| WO2022126608A1 (en) | Waterproof intelligent insole capable of replacing battery | |
| CN214128549U (en) | Portable gait analysis device | |
| CN211187240U (en) | Device for gait cycle recognition |
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 |