CN114678097A - Artificial intelligence and digital twinning system and method for intelligent clothes - Google Patents
Artificial intelligence and digital twinning system and method for intelligent clothes Download PDFInfo
- Publication number
- CN114678097A CN114678097A CN202210573379.XA CN202210573379A CN114678097A CN 114678097 A CN114678097 A CN 114678097A CN 202210573379 A CN202210573379 A CN 202210573379A CN 114678097 A CN114678097 A CN 114678097A
- Authority
- CN
- China
- Prior art keywords
- module
- user
- data
- sensor
- state
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 24
- 238000004891 communication Methods 0.000 claims abstract description 29
- 238000013135 deep learning Methods 0.000 claims abstract description 28
- 238000005516 engineering process Methods 0.000 claims abstract description 19
- 230000003993 interaction Effects 0.000 claims abstract description 13
- 239000011159 matrix material Substances 0.000 claims description 33
- 238000013528 artificial neural network Methods 0.000 claims description 28
- 239000013598 vector Substances 0.000 claims description 27
- 238000007781 pre-processing Methods 0.000 claims description 18
- 238000013507 mapping Methods 0.000 claims description 15
- 230000000717 retained effect Effects 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 11
- 238000012886 linear function Methods 0.000 claims description 11
- 230000000694 effects Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 230000001133 acceleration Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 6
- 238000005452 bending Methods 0.000 claims description 4
- 230000036760 body temperature Effects 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 210000002414 leg Anatomy 0.000 claims description 4
- 230000035790 physiological processes and functions Effects 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 4
- 210000001015 abdomen Anatomy 0.000 claims description 3
- 210000003423 ankle Anatomy 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000013523 data management Methods 0.000 claims description 3
- 230000009191 jumping Effects 0.000 claims description 3
- 210000003127 knee Anatomy 0.000 claims description 3
- 230000037081 physical activity Effects 0.000 claims description 3
- 210000000707 wrist Anatomy 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 description 5
- 230000036541 health Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 3
- 238000013475 authorization Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 241001456553 Chanodichthys dabryi Species 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000001217 buttock Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 210000000038 chest Anatomy 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 210000001513 elbow Anatomy 0.000 description 1
- 230000005802 health problem Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000002832 shoulder Anatomy 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
- A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
- A61B5/6804—Garments; Clothes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1116—Determining posture transitions
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Theoretical Computer Science (AREA)
- Veterinary Medicine (AREA)
- Physiology (AREA)
- Cardiology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Heart & Thoracic Surgery (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Fuzzy Systems (AREA)
- Pulmonology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses an artificial intelligence and digital twinning system and method for intelligent clothes, wherein the system comprises an intelligent clothes module and a digital twinning module; the intelligent clothing module comprises clothing, a sensor module and a wireless communication module, and is used for acquiring user data and performing data interaction with the digital twin module; the digital twin module comprises a deep learning algorithm module, a digital twin body module and a cloud server module, the cloud server module judges and stores the current state of the user by using the deep learning algorithm, the user entity is mapped to the digital twin body by using a digital twin technology to be displayed and interacted in real time, and when the user is detected to be in danger, the cloud server module automatically warns. The invention innovatively combines an intelligent wearing technology, an artificial intelligence technology and a digital twin technology, obtains more comprehensive user data, accurately judges the user state, can record and analyze the current state of the user in real time and take corresponding measures when the user encounters danger, and effectively ensures the life safety of the user.
Description
Technical Field
The invention belongs to the field of intelligent clothes, and particularly relates to an artificial intelligence and digital twinning system and method for intelligent clothes.
Background
With the innovative development of the clothing industry, intelligent clothing has been widely applied to a plurality of fields such as life health, medical health, safety protection, military equipment and the like, and is developed towards the direction that materials are more intelligent, functions are more comprehensive and services are more diversified. The intelligent wearable equipment can improve the difference of uneven distribution of medical resources, so that the laggard areas can fairly enjoy medical services; meanwhile, aiming at the safety problem of children and the aging population health problem, the physiological state and the daily activity state of the children can be monitored in real time through the intelligent clothes, and sufficient protection is provided.
The existing intelligent clothes only have a simple detection function on human bodies, the acquired data volume is small, the user state judgment is not accurate enough, the display is not visual enough, the product function is not comprehensive enough, the practicability is poor, and the actual requirements of users cannot be met. For example, chinese patent publication No. CN 107432739 a discloses "an intelligent clothing system for health monitoring", according to the scheme, information such as heart rate, body temperature, and acceleration of a human body can be collected, and a remote intelligent terminal is provided. However, the scheme has less detection data, the current posture of the user is difficult to be effectively judged only by the three-axis acceleration sensor at the waist of the human body, the detection algorithm is simple and crude, user feedback is lacked, no user interaction exists, historical data cannot be checked, and the protection on the user is poor.
Disclosure of Invention
Aiming at the defects or the improvement requirements of the prior art, the invention provides an artificial intelligence and digital twin system and method for intelligent clothing, which aim to monitor, record and display the self state of a user in real time, remind the user when the posture of the user is identified to be incorrect, automatically send out a warning when the danger of the user is detected, inform relatives and guarantee the health of the user.
To achieve the above object, according to one aspect of the present invention, there is provided an artificial intelligence and digital twinning system for intelligent clothing, comprising an intelligent clothing module and a digital twinning module;
the intelligent clothing module comprises clothing, a sensor module and a wireless communication module, and is used for acquiring user data and performing data interaction with the digital twin module; the digital twin module comprises a deep learning algorithm module, a digital twin body module and a cloud server module, wherein the deep learning algorithm module is embedded in the cloud server module, the cloud server module judges and stores the current state of a user by using a deep learning algorithm in the deep learning algorithm module, and a digital twin technology in the digital twin body module is used for mapping a user entity to a digital twin body for real-time display and interaction;
the deep learning algorithm module comprises a data preprocessing module and a neural network module;
the data preprocessing module is used for preprocessing the sensor data to obtain input data of the neural network module;
the neural network module performs supervised training on the labeled data by using a neural network combined with attention so as to accurately judge the physiological state of a user, and comprises two control units which are connected at a time and used for controlling information flow in the network, an intermediate state unit, an attention unit and an output unit.
Further, the specific processing procedure of the neural network module is as follows:
taking the data processed by the data preprocessing module as input;
using two control units for controlling in a networkInformation flow, respectively reset unitsRsAnd an update unitUpThe reset unit controls the number of the retained past states, the update unit controls how many copies of the old state are retained in the new state,
wherein the content of the first and second substances,is a parameter to be optimized for the network;is a bias parameter of the network;M t is a certain momenttA two-dimensional parameter matrix generated by a data preprocessing module;s t-1is the output immediately before the network;δ(X) Is a non-linear function of the signal,a matrix of R x C is represented,for each row in XAnd each columnIs traversed, compressing the value between 0 and 1, wherein Z represents an integer set;
intermediate state obtained by intermediate state unit in networkHSIt is shown that,wherein, in the step (A),is a parameter to be optimized for the intermediate state;B hs is a bias parameter for the intermediate state; the symbol is a binary operation with the input of two matrices of the same shape and the output of a matrix of the same shape with elements of each position equal to the product of the elements of the same position of the two input matrices;
the attention that the attention unit in the network takes is formed by a weighted sum of intermediate states, in particular,WhereinW a A weight matrix representing the attention of the user,SW t refers to an intermediate state of attention that is,SA t the attention value obtained by final calculation is referred to;
output unit in networktThe output of the time iss t ,In the formula (I), wherein,λfor a hyper-parameter, the last cell state is passedδNon-linear function to obtain user state output。
Further, a sensor module in the intelligent clothing module comprises an attitude sensor, a heart rate sensor, an infrared temperature sensor and a GPS chip; the infrared temperature sensor captures the current temperature of the human body, and the heart rate sensor captures the current heart rate of the human body; the attitude sensor is used for acquiring Euler angles, accelerations and coordinates of the attitude sensors of all parts; the working cycles of different sensors are different, the posture sensor takes 0.2s as one cycle, the heart rate sensor takes 30s as one cycle, and the infrared temperature sensor takes 300s as one cycle; the 15 posture sensors are distributed on key nodes of all parts of a human body, specifically, 1 posture sensor is distributed on each of the cervical vertebra, the chest and the abdomen, and 1 posture sensor is distributed on each of the left and right sides of the shoulder, the elbow, the wrist, the hip, the knee and the ankle.
Furthermore, the wireless communication module in the intelligent clothing module further comprises a linear motor, the wireless communication module is connected with the sensor on the intelligent clothing through a Bluetooth technology, data transmission is carried out through the wireless communication technology and the server, real-time acquisition and data management of data of the sensor are achieved, and when a specific instruction of the server is received, the linear motor is controlled to vibrate to prompt a user.
Furthermore, the data preprocessing module combines the acquired sensor data into a one-dimensional vector
Wherein, in the step (A),each representing the data of 15 sensors,hrrepresenting the data of the heartbeat of the human body,btrepresenting human body temperature data and generating human body activity labels in a mode of artificial markingLIncluding a plurality of human body states: walking, running, jumping, sitting still, lying down, lying on one side, sleeping, lifting legs, bending back and humpback, falling down, abnormal heart rate;
5 one-dimensional vectors acquired every 1s at inputvScreening maximum and minimum values to generate 2 one-dimensional vectorsv max Andv min the 2 one-dimensional vectors and the first 5 one-dimensional vectors are spliced to form a two-dimensional matrixPerforming an operation ofMAndLand sending the data to a neural network module for training.
Further, the digital twin module forms a five-tuple representation between the entity and the twin according to a mapping function, D = (RE, VT, AD, TD, MR), where RE in the formula represents a real entity, and refers to a part of the entity existing in reality, that is, a cloud server module; the VT represents a virtual twin, namely a digital twin module, the digital modeling technology is used for performing characteristic display on the entity in two physical and physiological dimensions, and the characteristic display is presented to clients such as Web, a mobile phone App and the like; AD represents activity data of an entity, in particular to data obtained by multi-source acquisition of the entity by a plurality of sensors; TD represents twin data, namely user state data obtained by a neural network module; MR represents the mapping relationship between the entity and the twin.
The invention also discloses an artificial intelligence and digital twinning method for the intelligent garment, which comprises the following steps:
s1: acquiring the current temperature and heart rate of a human body and Euler angles, accelerations and coordinates of all parts of the human body through a sensor;
s2: the method comprises the steps that sensor data are obtained through a wireless communication module, and data interaction is conducted between the sensor data and a cloud server; wherein the wireless communication module is internally provided with a linear motor;
s3: the cloud server judges and stores the current state of the user by using a deep learning algorithm, wherein input data of the deep learning algorithm is obtained by preprocessing the sensor data in the step S1;
s4: mapping a user entity to a digital twin body by utilizing a digital twin technology according to the current state of a user for showing and interacting;
s5: when the server judges that the user is in a bad state for a long time, a specific instruction is sent, the linear motor in the wireless communication module vibrates to prompt the user, and when the server judges that the user is in a danger, the current state and the coordinates of the user can be automatically reported.
Further, the deep learning algorithm in step S3 includes the following steps:
s31: combining the sensor data acquired at the same time into a one-dimensional vector, screening the maximum value and the minimum value of a plurality of acquired one-dimensional vectors to generate two new most-valued vectors, generating a two-dimensional matrix by splicing all the vectors, classifying the two-dimensional matrix data by using an artificial labeling mode to generate a human activity labelL;
S32: judging the current state of the user by using a neural network combining attention, wherein the specific processing process is as follows;
the data after the processing of step S31 is taken as input;
controlling information flow in a network using two control units, respectively reset unitsRsAnd an update unitUpThe reset unit controls the number of the retained past states, the update unit controls how many copies of the old state are retained in the new state,
wherein the content of the first and second substances,is a parameter to be optimized for the network;is a bias parameter of the network;M t is a certain momenttGenerating a two-dimensional parameter matrix;s t-1is the output immediately before the network;δ(X) Is a non-linear function of the signal,a matrix of R x C is represented,for each row in XAnd each columnIs traversed, compressing the value between 0 and 1, wherein Z represents an integer set;
intermediate states in a network are composed ofHSIt is shown that,wherein, in the step (A),is a parameter to be optimized for the intermediate state;B hs is the bias parameter for the intermediate state; the symbol is a binary operation with the input of two matrices of the same shape and the output of a matrix of the same shape with elements of each position equal to the product of the elements of the same position of the two input matrices;
attention in the network is formed by a weighted sum of intermediate states, in particular,WhereinW a A weight matrix representing the attention of the user,SW t refers to an intermediate state of attention that is,SA t the attention value obtained by final calculation is referred to;
network attThe output of the time iss t ,In the formula (I), wherein,λfor a hyper-parameter, the last cell state is passedδNon-linear function to obtain user state output。
Further, the digital twinning technique described in step S4 above forms a quintuple representation between the entity and the twin according to the mapping function; the digital twin body can show the current state of a user, and the user can send a specific instruction to the wireless communication module by operating the digital twin body, so that the wireless communication module reports the current sensor information or generates vibration.
In general, compared with the prior art, the above technical solution contemplated by the present invention can achieve the following beneficial effects:
(1) the artificial intelligence and digital twin system and method for intelligent clothing uses a neural network combining attention, and utilizes a nonlinear function compared with other neural networksThe calculation of the neural network is simplified, the utilization rate of the data is improved through attention, so that important data obtain larger weight, and the accuracy of the network is improved;
(2) according to the artificial intelligence and digital twin system and method for the intelligent garment, provided by the invention, the behavior category and the health state of a user can be accurately judged by identifying the physical condition of the user in real time in the system operation process; the digital twin body corresponding to the real entity is calculated and deduced, the state change of the user is reflected visually, and when the user operates the digital twin body to send a specific instruction or the system identifies an abnormal state, a warning can be sent out and corresponding measures can be taken, so that the life safety of the user is effectively guaranteed, and the risk of danger is reduced.
Drawings
Fig. 1 is a schematic flow chart of an artificial intelligence and digital twin system for intelligent clothing according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of an attitude sensor distribution provided by an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Fig. 1 is a schematic flow chart of an artificial intelligence and digital twin system for intelligent clothing according to an embodiment; the artificial intelligence and digital twin system for the intelligent garment comprises an intelligent garment module and a digital twin module.
The intelligent clothing module comprises clothing, a sensor module and a wireless communication module, and is used for acquiring user data and performing data interaction with the digital twin module;
the digital twin module comprises a deep learning algorithm module, a digital twin body module and a cloud server module, wherein the deep learning algorithm module is embedded in the cloud server module, the cloud server module judges and stores the current state of a user by using a deep learning algorithm in the deep learning algorithm module, a digital twin technology in the digital twin body module is used for mapping a user entity to the digital twin body for real-time display and interaction, and automatic warning is given when the user is detected to run into danger.
The sensor module comprises an attitude sensor, a heart rate sensor, an infrared temperature sensor and a GPS chip;
the infrared temperature sensor captures the current temperature of a human body, and the heart rate sensor captures the current heart rate of the human body; the attitude sensor is used for acquiring Euler angles, accelerations and coordinates of the attitude sensors of all parts; the working cycles of different sensors are different, the posture sensor takes 0.2s as one cycle, the heart rate sensor takes 30s as one cycle, and the infrared temperature sensor takes 300s as one cycle; the GPS chip is used for positioning;
wherein, 15 attitude sensors distribute on human every key node, specifically, 1 each of cervical vertebra, chest, belly, shoulder, elbow, wrist, buttock, knee, ankle respectively 1 about, as shown in fig. 2.
The wireless communication module is used for data interaction and further comprises a linear motor, the wireless communication module is connected with a sensor on the intelligent garment through a Bluetooth technology, data transmission is carried out through the wireless communication technology and the server, real-time acquisition and data management of sensor data are achieved, and when a specific instruction of the server is received, the linear motor can be controlled to vibrate to prompt a user.
The digital twin module comprises a deep learning algorithm module, a digital twin body module and a cloud server;
the deep learning algorithm module comprises a data preprocessing module and a neural network module;
the data preprocessing module combines the acquired sensor data into a one-dimensional vectorWherein, in the step (A),each representing the data of 15 sensors,hrrepresenting the data of the heartbeat of the human body,btrepresenting the human body temperature data and generating the human body activity label by using an artificial labeling modeLIncluding a plurality of human body states: walking, running, jumping, sitting still, lying down, lying on one side, sleeping, climbing the legs, bending over, falling down, and abnormal heart rate, inputting 5 one-dimensional vectors obtained every 1svScreening maximum and minimum values to generate 2 one-dimensional vectorsv max Andv min the 2 one-dimensional vectors and the first 5 one-dimensional vectors are spliced to form a two-dimensional matrixPerforming an operation ofMAndLand sending the data to a neural network module for training.
The neural network module performs supervised training on the labeled data by using a neural network combined with attention so as to accurately judge the physiological state of a user, and comprises two control units which are connected at a time and used for controlling information flow in the network, an intermediate state unit, an attention unit and an output unit. The specific processing process of the neural network module is as follows:
controlling information flow in a network using two control units, respectively reset unitsRsAnd an update unitUpThe reset unit controls the number of the retained past states, and the update unit controls the new statesHow many copies of the old state are retained in the state,
wherein, the first and the second end of the pipe are connected with each other,is a parameter to be optimized for the network;is a bias parameter of the network;M t is a certain momenttA two-dimensional parameter matrix generated by a data preprocessing module;s t-1is the output immediately before the network;δ(X) Is a non-linear function of the signal,a matrix of R x C is represented,for each row in XAnd each columnIs traversed, compressing the value between 0 and 1, wherein Z represents an integer set;
intermediate state obtained by intermediate state unit in networkHSIt is shown that,wherein, in the step (A),is in an intermediate stateChanging parameters;B hs is a bias parameter for the intermediate state; the symbol is a binary operation with the input of two matrices of the same shape and the output of a matrix of the same shape with elements of each position equal to the product of the elements of the same position of the two input matrices;
the attention acquired by the attention unit in the network is formed by a weighted sum of intermediate states, in particular,In whichW a A weight matrix representing the attention of the user,SW t refers to an intermediate state of attention that is,SA t the attention value obtained by final calculation is referred to;
output unit in networktThe output of the time iss t ,In the formula (I), the reaction is carried out,λfor a hyperparameter (default 0.3), the last cell state is passedδNonlinear function to obtain user state output。
Data obtained by sensorsvJudging the current state of the user and giving a human activity labelLAnd taking a plurality of data as a training set to train the neural network for a plurality of times, and storing the network weight with the best performance for prediction. And during prediction, loading the trained network weight, acquiring the current sensor data of the user, sending the data into a network, and outputting the current state of the user after network operation.
The digital twin body module forms a representation of five-tuple between an entity and a twin body according to a mapping function, D = (RE, VT, AD, TD, MR), wherein RE in the formula represents a real entity, and refers to an entity part existing in reality, namely a cloud server module; the VT represents a virtual twin, namely a digital twin module, the digital modeling technology is used for performing characteristic display on the entity in two physical and physiological dimensions, and the characteristic display is presented to clients such as Web, a mobile phone App and the like; AD represents activity data of an entity, in particular to data obtained by multi-source acquisition of the entity by a plurality of sensors; TD represents twin data, namely user state data obtained by a neural network module; MR represents the mapping relationship between the entity and the twin.
The cloud server module is used as a computing center of the whole system and is responsible for specific display of the virtual twin body, and the user data is stored in a data warehouse mode by adopting a distributed computing method. Specifically, modeling is carried out through Unity and other software, a digital twin portrait of a user is constructed, sensor data uploaded by an intelligent clothing module is calculated through a deep learning algorithm, accurate judgment is given, the current state of the user is reflected in a three-dimensional mode, and corresponding operation is adopted when the situation of the user is identified; the stored data is stored in a data warehouse, and a report can be generated and exported by taking the user information as a theme, and is sent to a doctor for analysis;
the user condition can be browsed by the user, and also can be given to a guardian, a child or a doctor of the user to check after authorization, so that the system not only can better look after the child or the old to judge the current activity condition of the child or the old, but also can be given to medical care personnel to analyze the physical condition; based on the digital twin module, the patient can provide long-term activity data when carrying out online consultation with experts in various places, and the patient can help doctors to judge physical conditions and adjust daily life habits.
After authorization, a guardian or a child of the user can operate a digital twin body or a cloud server to send a corresponding instruction to a wireless communication module after judging that the user is in an adverse state for a long time (30 seconds for lifting the legs, 3 minutes for bending over and humping back and 45 minutes for sitting), and the clothes are vibrated by a linear motor to prompt the user; when the user is judged to be in danger (fall, abnormal heart rate and the like), the relative is automatically contacted according to the reserved emergency contact way, the current state and the coordinates of the user are reported, and the life safety of the user is protected.
The embodiment of the invention provides an artificial intelligence and digital twinning method for intelligent clothing, which comprises the following steps:
s1: acquiring the current temperature and heart rate of a human body and Euler angles, accelerations and coordinates of all parts of the human body through a sensor;
s2: the method comprises the steps that sensor data are obtained through a wireless communication module, and data interaction is conducted between the sensor data and a cloud server; wherein the wireless communication module is internally provided with a linear motor;
s3: the cloud server judges and stores the current state of the user by using a deep learning algorithm, wherein input data of the deep learning algorithm is obtained by preprocessing the sensor data in the step S1;
s4: mapping a user entity to a digital twin body by utilizing a digital twin technology according to the current state of a user for showing and interacting;
s5: when the server judges that the user is in a bad state for a long time, a specific instruction is sent, the linear motor in the wireless communication module vibrates to prompt the user, and when the server judges that the user is in a danger, the current state and the coordinates of the user can be automatically reported.
Further, the deep learning algorithm in step S3 includes the following steps:
s31: combining the sensor data acquired at the same time into a one-dimensional vector, screening the maximum value and the minimum value of a plurality of acquired one-dimensional vectors to generate two new most-valued vectors, generating a two-dimensional matrix by splicing all the vectors, classifying the two-dimensional matrix data by using an artificial labeling mode to generate a human activity labelL;
S32: judging the current state of the user by using a neural network combined with attention, wherein the specific processing process is as follows;
the data after the processing of step S31 is taken as input;
controlling information flow in a network using two control units, respectively reset unitsRsAnd an update unitUpThe reset unit controls the number of the retained past states, the update unit controls how many copies of the old state are retained in the new state,
wherein, the first and the second end of the pipe are connected with each other,is a parameter to be optimized for the network;is a bias parameter of the network;M t is a certain momenttGenerating a two-dimensional parameter matrix;s t-1is the output immediately before the network;δ(X) Is a non-linear function of the signal,a matrix of R x C is represented,for each row in XAnd each columnIs traversed, compressing the value between 0 and 1, wherein Z represents an integer set;
intermediate states in a network are composed ofHSIt is shown that,wherein, in the step (A),is a parameter to be optimized for the intermediate state;B hs is a bias parameter for the intermediate state; the symbol is an Aldamard product, a binary operation with two inputsThe output is a matrix which has the same shape and the element of each position is equal to the product of the elements of the same position of the two input matrixes;
attention in the network is formed by a weighted sum of intermediate states, in particular,WhereinW a A weight matrix representing the attention of the user,SW t refers to an intermediate state of attention that is,SA t the attention value obtained by final calculation is referred to;
network attThe output of the time iss t ,In the formula (I), wherein,λfor a hyper-parameter, the last cell state is passedδNon-linear function to obtain user state output。
Further, the digital twinning technique described in step S4 forms a quintuple representation between the entity and the twin according to the mapping function; the digital twin body can show the current state of a user, and the user can send a specific instruction to the wireless communication module by operating the digital twin body, so that the wireless communication module reports the current sensor information or generates vibration.
The artificial intelligence and digital twin system and method for the intelligent garment, provided by the invention, can accurately acquire physiological data of a user, simplifies the complexity of the network by using a neural network combined with attention, accurately judges the current state of the user in real time, performs display and interaction by using a digital twin body, and gives a warning in time and takes corresponding measures when the user sends a specific instruction or the system identifies an abnormal state, so that the life safety of the user is effectively guaranteed, the risk of danger is reduced, and the system and method have certain application values.
It will be understood by those skilled in the art that the foregoing is only an exemplary embodiment of the present invention, and is not intended to limit the invention to the particular forms disclosed, since various modifications, substitutions and improvements within the spirit and scope of the invention are possible and within the scope of the appended claims.
Claims (9)
1. The utility model provides an artificial intelligence and digital twinning system towards wisdom clothing which characterized in that: the intelligent garment comprises an intelligent garment module and a digital twinning module, wherein the intelligent garment module comprises a garment, a sensor module and a wireless communication module and is used for acquiring user data and performing data interaction with the digital twinning module; the digital twin module comprises a deep learning algorithm module, a digital twin body module and a cloud server module, wherein the deep learning algorithm module is embedded in the cloud server module, the cloud server module judges and stores the current state of a user by using a deep learning algorithm in the deep learning algorithm module, and a digital twin technology in the digital twin body module is used for mapping a user entity to a digital twin body for real-time display and interaction;
the deep learning algorithm module comprises a data preprocessing module and a neural network module;
the data preprocessing module is used for preprocessing the sensor data to obtain input data of the neural network module;
the neural network module performs supervised training on the labeled data by using a neural network combined with attention so as to accurately judge the physiological state of a user, and comprises two control units which are connected at a time and used for controlling information flow in the network, an intermediate state unit, an attention unit and an output unit.
2. The intelligent garment-oriented artificial intelligence and digital twin system of claim 1, wherein: the specific processing process of the neural network module is as follows:
taking the data processed by the data preprocessing module as input;
controlling information flow in a network using two control units, respectively reset unitsRsAnd an update unitUpThe reset unit controls the number of the retained past states, the update unit controls how many copies of the old state are retained in the new state,
wherein the content of the first and second substances,is a parameter to be optimized for the network;is a bias parameter of the network;M t is a certain momenttA two-dimensional parameter matrix generated by a data preprocessing module;s t-1is the output immediately before the network;δ(X) Is a non-linear function of the signal,a matrix of R x C is represented,for each row in XAnd each columnIs traversed, compressing the value between 0 and 1, wherein Z represents an integer set;
in-network intermediate state unit acquisitionIs in an intermediate state ofHSIt is shown that,wherein, in the step (A),is a parameter to be optimized for the intermediate state;B hs is a bias parameter for the intermediate state; the symbol is a binary operation with the input of two matrices of the same shape and the output of a matrix of the same shape with elements of each position equal to the product of the elements of the same position of the two input matrices;
the attention acquired by the attention unit in the network is formed by a weighted sum of intermediate states, in particular,WhereinW a A weight matrix representing the attention of the user,SW t refers to an intermediate state of attention that is,SA t the attention value obtained by final calculation is referred to;
3. The intelligent garment-oriented artificial intelligence and digital twin system of claim 1, wherein: the sensor module in the intelligent clothing module comprises an attitude sensor, a heart rate sensor, an infrared temperature sensor and a GPS chip; the infrared temperature sensor captures the current temperature of the human body, and the heart rate sensor captures the current heart rate of the human body; the attitude sensor is used for acquiring Euler angles, accelerations and coordinates of the attitude sensors of all parts; the working cycles of different sensors are different, the posture sensor takes 0.2s as one cycle, the heart rate sensor takes 30s as one cycle, and the infrared temperature sensor takes 300s as one cycle; the 15 posture sensors are distributed on key nodes of all parts of a human body, specifically, 1 posture sensor is distributed on each of the cervical vertebra, the chest and the abdomen, and 1 posture sensor is distributed on each of the left and right sides of the shoulder, the elbow, the wrist, the hip, the knee and the ankle.
4. The intelligent garment-oriented artificial intelligence and digital twin system of claim 1, wherein: the wireless communication module in the intelligent clothing module further comprises a linear motor, the wireless communication module is connected with a sensor on the intelligent clothing through a Bluetooth technology, data transmission is carried out through the wireless communication technology and the server, real-time acquisition and data management of data of the sensor are achieved, and when a specific instruction of the server is received, the linear motor is controlled to vibrate to prompt a user.
5. The intelligent garment-oriented artificial intelligence and digital twin system of claim 3, wherein:
the data preprocessing module combines the acquired sensor data into a one-dimensional vectorWherein, in the step (A),each representing the data of 15 sensors,hrrepresenting the data of the heartbeat of the human body,btrepresenting human body temperature data and generating human body activity labels in a mode of artificial markingLIncluding a plurality of human body states: walking, running, jumping, sitting still, lying down, lying on side, sleeping, lifting legs, bending back, falling down, and having different heart ratesFrequently;
5 one-dimensional vectors acquired every 1s at inputvScreening maximum and minimum values to generate 2 one-dimensional vectorsv max Andv min the 2 one-dimensional vectors and the first 5 one-dimensional vectors are spliced to form a two-dimensional matrixPerforming an operation ofMAndLand sending the data to a neural network module for training.
6. The intelligent garment-oriented artificial intelligence and digital twin system of claim 1, wherein: the digital twin module forms a quintuple representation between an entity and the twin according to a mapping function, D = (RE, VT, AD, TD, MR), wherein RE in the formula represents a real entity, and refers to an entity part existing in reality, namely a cloud server module; the VT represents a virtual twin, namely a digital twin module, the digital modeling technology is used for performing characteristic display on the entity in two physical and physiological dimensions, and the characteristic display is presented to clients such as Web, a mobile phone App and the like; AD represents activity data of an entity, in particular to data obtained by multi-source acquisition of the entity by a plurality of sensors; TD represents twin data, namely user state data obtained by a neural network module; MR represents the mapping relationship between the entity and the twin.
7. An artificial intelligence and digital twinning method for intelligent clothes is characterized by comprising the following steps:
s1: acquiring the current temperature and heart rate of a human body and Euler angles, accelerations and coordinates of all parts of the human body through a sensor;
s2: the method comprises the steps that sensor data are obtained through a wireless communication module, and data interaction is conducted between the sensor data and a cloud server; wherein the wireless communication module is internally provided with a linear motor;
s3: the cloud server judges and stores the current state of the user by using a deep learning algorithm, wherein input data of the deep learning algorithm is obtained by preprocessing the sensor data in the step S1;
s4: mapping a user entity to a digital twin body by utilizing a digital twin technology according to the current state of a user for showing and interacting;
s5: when the server judges that the user is in a bad state for a long time, a specific instruction is sent, the linear motor in the wireless communication module vibrates to prompt the user, and when the server judges that the user is in a danger, the current state and the coordinates of the user can be automatically reported.
8. The intelligent garment-oriented artificial intelligence and digital twinning method of claim 7, wherein: the deep learning algorithm described in step S3 includes the following steps:
s31: combining the sensor data acquired at the same time into a one-dimensional vector, screening the maximum value and the minimum value of a plurality of acquired one-dimensional vectors to generate two new most-valued vectors, generating a two-dimensional matrix by splicing all the vectors, classifying the two-dimensional matrix data by using an artificial labeling mode to generate a human activity labelL;
S32: judging the current state of the user by using a neural network combining attention, wherein the specific processing process is as follows;
the data after the processing of step S31 is taken as input;
controlling information flow in a network using two control units, respectively reset unitsRsAnd an update unitUpThe reset unit controls the number of the retained past states, the update unit controls how many copies of the old state are retained in the new state,
wherein the content of the first and second substances,is a parameter to be optimized for the network;is a bias parameter of the network;M t is a certain momenttGenerating a two-dimensional parameter matrix;s t-1is the output immediately before the network;δ(X) Is a non-linear function of the signal,a matrix of R x C is represented,for each row in XAnd each columnIs traversed, compressing the value between 0 and 1, wherein Z represents an integer set;
intermediate states in a network are composed ofHSIt is shown that,wherein, in the step (A),is a parameter to be optimized for the intermediate state;B hs is a bias parameter for the intermediate state; the symbol is a binary operation with the input of two matrices of the same shape and the output of a matrix of the same shape with elements of each position equal to the product of the elements of the same position of the two input matrices;
attention in the network is formed by a weighted sum of intermediate states, in particular,In whichW a A weight matrix representing the attention of the user,SW t refers to an intermediate state of attention that is,SA t the attention value obtained by final calculation is referred to;
9. The intelligent garment-oriented artificial intelligence and digital twinning method of claim 7, wherein: the digital twinning technique described in step S4 forms a quintuple representation between the entity and the twin according to the mapping function; the digital twin body can show the current state of a user, and the user can send a specific instruction to the wireless communication module by operating the digital twin body, so that the wireless communication module reports the current sensor information or generates vibration.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210573379.XA CN114678097B (en) | 2022-05-25 | 2022-05-25 | Artificial intelligence and digital twinning system and method for intelligent clothes |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210573379.XA CN114678097B (en) | 2022-05-25 | 2022-05-25 | Artificial intelligence and digital twinning system and method for intelligent clothes |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114678097A true CN114678097A (en) | 2022-06-28 |
CN114678097B CN114678097B (en) | 2022-08-30 |
Family
ID=82080654
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210573379.XA Active CN114678097B (en) | 2022-05-25 | 2022-05-25 | Artificial intelligence and digital twinning system and method for intelligent clothes |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114678097B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116051810A (en) * | 2023-03-30 | 2023-05-02 | 武汉纺织大学 | Intelligent clothing positioning method based on deep learning |
CN116687409A (en) * | 2023-07-31 | 2023-09-05 | 武汉纺织大学 | Emotion recognition method and system based on digital twin and deep learning |
CN117426758A (en) * | 2023-12-20 | 2024-01-23 | 武汉纺织大学 | Intelligent clothing system and method based on multi-sensing information fusion |
CN117441980A (en) * | 2023-12-20 | 2024-01-26 | 武汉纺织大学 | Intelligent helmet system and method based on intelligent computation of multi-sensor information |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112818446A (en) * | 2021-01-26 | 2021-05-18 | 西安交通大学 | Construction method of intelligent workshop digital twin system |
EP3905225A1 (en) * | 2020-04-28 | 2021-11-03 | Université de Strasbourg | System and method for evaluating simulation-based medical training |
-
2022
- 2022-05-25 CN CN202210573379.XA patent/CN114678097B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3905225A1 (en) * | 2020-04-28 | 2021-11-03 | Université de Strasbourg | System and method for evaluating simulation-based medical training |
CN112818446A (en) * | 2021-01-26 | 2021-05-18 | 西安交通大学 | Construction method of intelligent workshop digital twin system |
Non-Patent Citations (5)
Title |
---|
HONGQIANG LI ET AL.: "《Wearable sensors in intelligent clothing for measursing human body temperature based on optical fiber bragg grating》", 《NATIONAL LIBRARY OF MEDICINE》 * |
孙金平: "《智能老年医护服装的开发》", 《服装学报》 * |
张捷等: "数字孪生技术在社区老年人安全健康监测领域的应用探究", 《中国医疗器械杂志》 * |
裴嘉晖等: "针对颈椎疾病的坐姿监测及理疗内衣设计", 《上海纺织科技》 * |
黄自鑫等: "可穿戴无线体域网中异质传感器数据融合算法研究", 《现代制造技术与装备》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116051810A (en) * | 2023-03-30 | 2023-05-02 | 武汉纺织大学 | Intelligent clothing positioning method based on deep learning |
CN116687409A (en) * | 2023-07-31 | 2023-09-05 | 武汉纺织大学 | Emotion recognition method and system based on digital twin and deep learning |
CN116687409B (en) * | 2023-07-31 | 2023-12-12 | 武汉纺织大学 | Emotion recognition method and system based on digital twin and deep learning |
CN117426758A (en) * | 2023-12-20 | 2024-01-23 | 武汉纺织大学 | Intelligent clothing system and method based on multi-sensing information fusion |
CN117441980A (en) * | 2023-12-20 | 2024-01-26 | 武汉纺织大学 | Intelligent helmet system and method based on intelligent computation of multi-sensor information |
CN117441980B (en) * | 2023-12-20 | 2024-03-22 | 武汉纺织大学 | Intelligent helmet system and method based on intelligent computation of multi-sensor information |
CN117426758B (en) * | 2023-12-20 | 2024-04-05 | 武汉纺织大学 | Intelligent clothing system and method based on multi-sensing information fusion |
Also Published As
Publication number | Publication date |
---|---|
CN114678097B (en) | 2022-08-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114678097B (en) | Artificial intelligence and digital twinning system and method for intelligent clothes | |
Forbes et al. | Fall prediction using behavioural modelling from sensor data in smart homes | |
Gao et al. | Wearable human motion posture capture and medical health monitoring based on wireless sensor networks | |
Fortino et al. | Enabling effective programming and flexible management of efficient body sensor network applications | |
Quwaider et al. | Body posture identification using hidden Markov model with a wearable sensor network | |
Tseng et al. | The elderly care system | |
Zhang et al. | A context-aware mhealth system for online physiological monitoring in remote healthcare | |
US20150088004A1 (en) | Method for monitoring an accurate heart rate | |
Chawla et al. | Edge AI with Wearable IoT: A Review on Leveraging Edge Intelligence in Wearables for Smart Healthcare | |
BC et al. | IOT Based Remote Health Monitoring System | |
CN117322852A (en) | Intelligent clothing health monitoring system and method based on deep learning | |
Koren et al. | Comparison of data-driven models for cleaning eHealth sensor data: Use case on ECG signal | |
Huang et al. | Implementation of a wireless sensor network for heart rate monitoring in a senior center | |
CN115841863A (en) | Rehabilitation training plan generation and rehabilitation training system | |
Santamaria et al. | Smart wearable device for health monitoring in the internet of things (IoT) domain | |
Barshan et al. | A Novel Heuristic Fall-Detection Algorithm Based on Double Thresholding, Fuzzy Logic, and Wearable Motion Sensor Data | |
Li et al. | Internet of things-based smart wearable system to monitor sports person health | |
Biswas et al. | Remote monitoring of soldier safety through body posture identification using wearable sensor networks | |
Rashid et al. | Emergency wireless health monitoring system using wearable technology for refugee camp and disaster affected people | |
CN112329812A (en) | Slow obstructive pulmonary acute exacerbation automatic early warning method and platform and readable storage medium | |
Kanimozhi et al. | Development of Health Monitoring Vest using Velostat | |
Banerjee et al. | IOT-based fluid and heartbeat monitoring for advanced healthcare | |
Sun et al. | An overview of aiomt applications | |
Jyotsna et al. | A Personalized Healthcare Platform for Monitoring Mental Health of a Person During COVID | |
Sabarmathi et al. | Application of Cognitive Computing in Healthcare |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20240320 Address after: No. 180 Qianyang West Road, Zhouji Office, Qianjiang City, Hubei Province, 433100 Patentee after: Hubei Bifan Garment Co.,Ltd. Country or region after: Zhong Guo Address before: 430073 No. 1 Textile Road, Hongshan District, Hubei, Wuhan Patentee before: Wuhan Textile University Country or region before: Zhong Guo |