CN111693024A - Wearable human body sensing monitoring equipment based on nine-axis inertia measurement unit - Google Patents
Wearable human body sensing monitoring equipment based on nine-axis inertia measurement unit Download PDFInfo
- Publication number
- CN111693024A CN111693024A CN202010530351.9A CN202010530351A CN111693024A CN 111693024 A CN111693024 A CN 111693024A CN 202010530351 A CN202010530351 A CN 202010530351A CN 111693024 A CN111693024 A CN 111693024A
- Authority
- CN
- China
- Prior art keywords
- axis
- falling
- fall
- module
- measurement unit
- 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
- 238000005259 measurement Methods 0.000 title claims abstract description 60
- 238000012544 monitoring process Methods 0.000 title claims abstract description 12
- 238000000034 method Methods 0.000 claims abstract description 22
- 210000003414 extremity Anatomy 0.000 claims abstract description 19
- 230000033001 locomotion Effects 0.000 claims abstract description 19
- 238000004891 communication Methods 0.000 claims abstract description 12
- 210000000707 wrist Anatomy 0.000 claims abstract description 12
- 230000001133 acceleration Effects 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims abstract description 6
- 239000013598 vector Substances 0.000 claims description 41
- 238000001514 detection method Methods 0.000 claims description 24
- 230000008569 process Effects 0.000 claims description 16
- 239000011159 matrix material Substances 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 13
- 230000036760 body temperature Effects 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 12
- 230000000295 complement effect Effects 0.000 claims description 12
- 238000001914 filtration Methods 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 10
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 9
- 239000008280 blood Substances 0.000 claims description 9
- 210000004369 blood Anatomy 0.000 claims description 9
- 229910052760 oxygen Inorganic materials 0.000 claims description 9
- 239000001301 oxygen Substances 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000010586 diagram Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 6
- 230000005484 gravity Effects 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 4
- 238000009825 accumulation Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 238000012821 model calculation Methods 0.000 claims description 3
- 238000012806 monitoring device Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000007665 sagging Methods 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 230000004069 differentiation Effects 0.000 claims description 2
- 230000010354 integration Effects 0.000 claims description 2
- 230000036544 posture Effects 0.000 description 23
- 230000036541 health Effects 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 3
- 230000002093 peripheral effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000474 nursing effect Effects 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C9/00—Measuring inclination, e.g. by clinometers, by levels
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C19/00—Gyroscopes; Turn-sensitive devices using vibrating masses; Turn-sensitive devices without moving masses; Measuring angular rate using gyroscopic effects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P15/00—Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration
- G01P15/18—Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration in two or more dimensions
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R33/00—Arrangements or instruments for measuring magnetic variables
- G01R33/02—Measuring direction or magnitude of magnetic fields or magnetic flux
- G01R33/0206—Three-component magnetometers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/285—Selection of pattern recognition techniques, e.g. of classifiers in a multi-classifier system
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
A wearable human sensing monitoring facilities based on nine inertial measurement units includes: the device comprises a power module, a sensor module, a nine-axis inertia measurement unit, a microprocessor module, a Bluetooth communication module and a mobile phone terminal; the nine-axis inertial measurement unit is integrated with a three-axis accelerometer, a three-axis gyroscope and a nine-axis inertial measurement unit of a three-axis magnetometer, is worn on the wrist position of a user, is used for detecting the posture conditions of the user in three directions of a preset three-dimensional coordinate system X, Y and Z, and judges the normal state of the human body and the accidental falling event by resolving multi-source data and combining a mode identification method; the nine-axis inertia measurement module is mainly used for collecting relevant parameters of human hand motion and processing and restoring the posture of the moving limb in space through the motion parameters; the motion parameters describing the moving limbs comprise acceleration, rotation angular velocity of corresponding joints and magnetic field information of the limbs in corresponding spaces.
Description
Technical Field
The invention relates to the field of sensor measurement, in particular to wearable human body sensing monitoring equipment based on a nine-axis inertia measurement unit.
Background
With the development of smart wearable technology, it is becoming easier and more popular to provide continuous healthcare for the elderly. An autonomous system that can provide rapid and continuous health status monitoring for people is clearly very attractive, which can help the elderly to improve their health and reduce the health cost expenditure of society for the elderly. Retrieving the technical solutions that have been developed, it can be seen that such systems are dedicated to monitoring important parameters such as Electrocardiogram (ECG), pulse and blood oxygen saturation, blood pressure, etc. in the home as well as in the outdoor environment, and have the ability to alert the hospital or doctor directly when a problem arises in the health condition. Some important parameter sensing platforms such as smart phones, smart watches and the like provide great convenience for developing mobile nursing devices for continuous health monitoring. However, power consumption, cost and human-machine comfort of home medical wearable devices are always issues to be perfected.
For the elderly who are in the single location, an accidental fall can have serious consequences, and therefore real-time detection of a fall event is very essential. In the traditional method, a three-axis acceleration sensor is adopted to detect a human body in real time, sometimes the human body cannot be accurately monitored, and especially the human body can miss judgment due to falling with a small amplitude or slow falling.
Disclosure of Invention
In order to solve the technical problems, the invention provides a wearable human body sensing monitoring device based on a nine-axis inertia measurement unit, which integrates a nine-axis inertia measurement unit comprising a three-axis accelerometer, a three-axis gyroscope and a three-axis magnetometer, is more accurate in measurement, can measure the body temperature, the heart rate and the blood oxygen level of the old in real time, and sends physiological data to a smart phone of a caregiver. In addition, the device can detect fall incidents encountered by elderly people and alert the caretaker.
The technical scheme adopted by the invention is as follows: a wearable human sensing monitoring facilities based on nine inertial measurement units includes: the device comprises a power module, a sensor module, a nine-axis inertia measurement unit, a microprocessor module, a Bluetooth communication module and a mobile phone terminal;
the power supply module provides a direct current power supply for the sensor module, the nine-axis inertia measurement unit, the microcontroller module and the Bluetooth communication module;
the nine-axis inertial measurement unit is integrated with a three-axis accelerometer, a three-axis gyroscope and a nine-axis inertial measurement unit of a three-axis magnetometer, is worn on the wrist position of a user, is used for detecting the posture conditions of the user in three directions of a preset three-dimensional coordinate system X, Y and Z, and judges the normal state of the human body and the accidental falling event by resolving multi-source data and combining a mode identification method;
the nine-axis inertia measurement module is mainly used for collecting relevant parameters of human hand motion and processing and restoring the posture of the moving limb in space through the motion parameters; the motion parameters describing the moving limbs comprise acceleration, rotation angular velocity of corresponding joints and magnetic field information of the limbs in corresponding spaces; firstly, a geographical coordinate system n system and a human body posture coordinate system b system are established, wherein the geographical coordinate system n system is OxnynznSelecting a northeast coordinate system, OxnThe axis pointing to the north, OynThe axis pointing east, OznThe axis points to the sky along the reverse direction of the plumb line, and the original point is the wearing position of the equipment;
the nine-axis inertia measurement module presets the wearing position of the equipment as a human body posture coordinate system b system, namely OxbybzbOrigin of coordinates of, OxbThe axis is directed forward, OybAxis pointing to the left, OzbThe axis points upward; as defined below:
r is the cosine of the true attitude direction,cosine of attitude direction, R, calculated for complementary filtering0For the attitude matrix, mu, observed by the accelerometer and the three-axis magnetometerHIs R0Is observed as noise R0=R+μH,RCAttitude matrix, mu, calculated for gyroscope measurement dataLAccumulating error R for a gyroscopec=R+μLS represents an intermediate operator, 1/s represents integration and s represents differentiation in the circuit diagram;
computingHigh-frequency components of the accelerometer and the magnetometer are filtered out firstly, and low-frequency components of the gyroscope are filtered out again, wherein:
GL(s) and GH(s) is the transfer function of the complementary filter, GL(s) having a first-order low-pass filtering characteristic, GH(s) having a first order high pass filtering characteristic, capable of eliminating the accumulation of high frequency noise and low frequency error by complementary filtering;
GL(s) and GH(s) complementary to give:
further, the fall detection algorithm based on the nine-axis inertial measurement unit specifically comprises the following steps:
designing and manufacturing an arm posture data set of the old people in a falling state, wherein the data set is divided into a training set and a testing set; the arm posture data set comprises relevant feature vectors for fall detection and is provided with fall/non-fall labels, the fall labels can be subdivided into six categories of forward fall, backward fall, left side fall, right side fall, supine fall and prone fall, and the data set is used for training the fall detection model.
Step two, constructing a falling detection model based on a deep neural network, wherein the input of the model is a feature vector related to falling, and the output is the probability of falling types and the probability of occurrence of various non-falling situations; training a falling detection model by using the training set in the arm posture data set to obtain model parameters, testing the performance of the model by using the test set in the arm posture data set and adjusting the model parameters.
Acquiring output data of the nine-axis inertia measurement unit in a falling state, and calculating to obtain a feature vector; the output data of the nine-axis inertia measurement unit is as follows: accelerationThe measured value of the meter is ax,ay,az(ii) a The gyroscope measured value is omegax,ωy,ωz(ii) a The three-axis magnetometer has m measured valuesx,my,mz(ii) a The treatment process comprises the following steps:
(1) initializing quaternions
Wherein,theta and gamma are initial attitude angles of the moving limbs, the arms of the user are in a natural sagging state in a static state, and the three initial attitude angles are assumed to be zero;
(2) converting the measured value of the accelerometer and the measured value of the triaxial magnetometer into a unit vector;
(3) obtaining a gravity vector (v) using a quaternionx,vy,vz)TAnd magnetic field vector (w)x,wy,wz)T
the three-axis magnetometer measures the magnitude and direction of the earth's magnetic field at an angle to each of the coordinate axes of the n-system, denoted (b)x,by,bz)TThe x-axis, which has been previously set to n, points to the north, so by0, i.e. (b)x,0,bz)T;
Calculating the magnetic field vector needs to be pushed from b to n systems, such thatWherein (m)x,my,mz)TIs the output of the magnetometer in the b-system. Since the geomagnetism meters have the same vector size on the XOY plane of the n system and the b system, the geomagnetism meters can be used for measuring the geomagnetism of the geomagnetismFor bzWithout change, bz=hzThereby obtaining (b)x,0,bz)T;
Obtaining the following result after converting the n system to the b system through the conversion matrix of the b system:
wherein,a transformation matrix between a geographic coordinate system and a carrier coordinate system;
(4) calculate the error e, i.e. (e)x,ey,ez)T
(ex,ey,ez)T=(vx,vy,vz)Tⅹ(ax,ay,az)T+(mx,my,mz)Tⅹ(wx,wy,wz)T
(5) Correction of gyroscope data using errors
=Kp*e+Ki*∫e,ω=ωg+, where KpDenotes the proportional gain, KiRepresenting integral gain, for controlling the rate of convergence, omega, of the gyro deviationg=(ωx,ωy,ωz)T。
(6) Updating quaternion and converting quaternion to euler angles
Solving this differential equation yields:
through the calculation, the attitude angle data of the arm at any moment can be obtained, and the two change values of the attitude angle and the corresponding time variation thereof when the falling event occurs are taken as the characteristic vector. That is, when a suspected fall event occurs, the timer is turned on, the size of the attitude angle reaches the peak/trough for the first time to serve as a first section, and the variation delta phi of the attitude angle of the first section is recorded1、Δθ1、Δγ1And corresponding time variation Δ t1(ii) a The change quantity delta phi of the attitude angle of the section is recorded from the first time of reaching the peak/trough to the second time of reaching the trough/peak as a second section2、Δθ2、Δγ2And corresponding time variation Δ t2. The recorded data constitute an 8-dimensional feature vector.
The suspected fall event is the condition that the combined acceleration is zero and is detected by the accelerometer.
Step four, a pattern matching process; and after normalization preprocessing is carried out on the obtained characteristic vectors, the characteristic vectors are input into the falling detection model, the probability of different falling types and the probability of non-falling situations are obtained through model calculation, and whether falling occurs and the types of falling are judged according to the probability.
The sensor module is worn on the wrist of a user and is in contact with the skin of the user, comprises a high-sensitivity pulse oximeter, a heart rate sensor and a body temperature sensor and is used for measuring the heart rate, the blood oxygen level and the body temperature of the user in real time;
the microprocessor module comprises a relevant necessary control circuit, the input end of the microprocessor module is connected with the output ends of the sensor module and the inertia measurement module, and the microprocessor module is used for establishing a detection model of a human body falling event according to the acquired sensor data, researching the falling process of the human body, extracting characteristic parameters of hand motion changes in several falling postures, and processing the parameters through the falling detection model so as to accurately distinguish the normal state of the human body from an accidental falling event; the microprocessor module needs to be additionally provided with a timer and a timer, the timer sets an expiration event, and the only trigger condition of the event is that the input value of the sensor module exceeds a certain threshold value;
the mobile phone terminal receives the posture and physiological signals of the user sent by the microprocessor module through Bluetooth communication, and reminds the mobile phone terminal holder of the falling old man to cure the falling old man through sound and character information at the first time; the mobile phone terminal holder is a family or a care person.
All the hardware are installed in a wrist strap with good man-machine performance, the wrist strap is worn by the old, the body temperature, the heart rate and the blood oxygen level of the old can be measured in all weather, and whether a falling event happens or not is detected.
Has the advantages that:
the invention provides wearable human body sensing monitoring equipment based on a nine-axis inertia measurement unit, which is worn on the wrist position of a user by utilizing the nine-axis inertia measurement unit fused with a three-axis accelerometer, a three-axis gyroscope and a three-axis magnetometer and used for detecting the posture conditions of the user in three directions of a preset three-dimensional coordinate system X, Y and Z and judging the normal state and the accidental falling event of a human body by resolving a multi-source data combination mode identification method; because the data of various measuring units are fused and the error is corrected, the data can be accurately detected.
Drawings
Fig. 1 is a structural block diagram of a wearable human body sensing monitoring device based on a nine-axis inertial measurement unit according to an embodiment of the present invention;
fig. 2 is a flow chart of a fall detection algorithm provided in an embodiment of the invention;
FIG. 3 is a schematic diagram of a coordinate system of a wearing position of the device and a posture of a human body according to an embodiment of the present invention;
FIG. 4 is a calculation process of the present invention for converting the output data of the nine-axis inertial measurement unit into an arm attitude angle;
FIG. 5 is a diagram of a microprocessor and its peripheral circuits according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
According to one embodiment of the invention, the wearable human body sensing monitoring equipment based on the nine-axis inertial measurement unit is provided, the old people wear the equipment on the wrist or above the elbow through the wrist strap, the heart rate, the body temperature and the blood oxygen level of the old people can be recorded in real time by starting the equipment, the walking condition of the old people is monitored, and whether a falling event happens or not is tracked constantly. The microprocessor module is programmed based on the single chip technology, so that the equipment has a real-time operating system. The device is provided with a corresponding Android application program, and the microcontroller is communicated with the mobile phone terminal through Bluetooth.
The embodiment provides old person's family rehabilitation training check out test set based on intelligent wearable sensor, including power module, sensor module, inertia measurement module, microprocessor module, bluetooth communication module, cell-phone terminal.
The power module provides dc power to the sensor module, the inertia measurement module, the microcontroller module, and the bluetooth communication module, and controls the voltage and power of the system using a PMIC (power management integrated circuit), as shown in fig. 1.
The inertial measurement module is also worn on the wrist of a user, is a nine-axis inertial measurement unit which integrates a three-axis accelerometer, a three-axis gyroscope and a three-axis magnetometer and is used for detecting the posture conditions of the user in three directions of X, Y and Z in a preset three-dimensional coordinate system; the module mainly collects relevant parameters of human hand motion, and the posture of the moving limb in the space is restored through processing the motion parameters; the motion parameters describing the moving limbs comprise acceleration, rotation angular velocity of corresponding joints and magnetic field information of the limbs in corresponding spaces. The specific use of the inertial measurement module is as follows:
firstly, establishing a geographic coordinate system (n system) and a human body posture coordinate system (b system), wherein the geographic coordinate system OxnynznSelecting a northeast coordinate system, OxnThe axis pointing to the north, OynThe axis pointing east, OznThe axis points to the sky along the reverse direction of the plumb line, and the original point is the wearing position of the equipment; human body posture coordinate system OxbybzbThe origin of coordinates of (a) is also the device wearing position, OxbThe axis is directed forward, OybAxis pointing to the left, OzbThe axis points upward. The attitude angle parameters of the moving limb include the pitch angle theta (limb around Oy)bAxial motion), roll angle gamma (limbs wound by Ox)bShaft motion) and yaw angle(limbs wound around Oz)bThe shaft rotates). As defined below:
r is the cosine of the true attitude direction,cosine of attitude direction, R, calculated for complementary filtering0For the attitude matrix, mu, observed by the accelerometer and the three-axis magnetometerHIs R0Is observed as noise R0=R+μH,RCAttitude matrix, mu, calculated for gyroscope measurement dataLAccumulating error R for a gyroscopec=R+μLS denotes an intermediate operator, 1/s denotes integral and s denotes differential in the block diagram.
By which a high degree of accuracy is to be achievedThe high frequency components of the accelerometer and the magnetometer need to be filtered out, the low frequency components of the gyroscope need to be filtered out, wherein:
GL(s) and GH(s) is the transfer function of the complementary filter, GL(s) having a first-order low-pass filtering characteristic, GH(s) has a first-order high-pass filtering characteristic, and the accumulation of high-frequency noise and low-frequency errors can be eliminated through complementary filtering.
GL(s) and GH(s) complementary to obtain:
referring to fig. 2, a specific flow of a fall detection algorithm based on a nine-axis inertial measurement unit is described below with reference to a calculation process of a limb movement posture angle:
designing and manufacturing an arm posture data set of the old people in a falling state, wherein the data set is divided into a training set and a testing set; the arm posture data set comprises relevant feature vectors for fall detection and is provided with fall/non-fall labels, the fall labels can be subdivided into six categories of forward fall, backward fall, left side fall, right side fall, supine fall and prone fall, and the data set is used for training the fall detection model.
Considering that the falling experiment has certain danger to the old, the making of the arm posture data set is simulated by the young. The simulation experiments of six items of forward falling, backward falling, left falling, right falling, supine falling and prone falling were respectively carried out on 10cm protective mats by recruiting 20 volunteers (20-40 years old), and each item was carried out three times. Each volunteer also completed a set of daily movements of walking (5 steps), standing (3s), sitting (5s), standing (3s), lying (10s), and standing up as data in a non-falling state. The attitude angle change data and the corresponding time change data when the suspected fall occurs (the total acceleration is zero) are collected in the fall experiment, the proceeding process of each group of actions is divided into 20 intervals in the non-fall experiment, and the attitude angle change quantity and the corresponding time length of each interval are taken as elements of the feature vector.
Step two, constructing a falling detection model based on a deep neural network, wherein the input of the model is a feature vector related to falling, and the output is the probability of falling types and the probability of occurrence of various non-falling situations; training a falling detection model by using the training set in the arm posture data set to obtain model parameters, testing the performance of the model by using the test set in the arm posture data set and adjusting the model parameters.
According to the falling detection model based on the deep neural network, a 3-layer system is selected as a standard BP network, 8 neurons are arranged on an input layer of the network, and the input layer corresponds to 8 input characteristic values; the middle layer has 32 neurons; the output layer has 7 neurons, and corresponds to six fall types of forward fall, backward fall, left side fall, right side fall, supine fall and prone fall and non-fall situations. For the multi-classification problem, a Softmax function is adopted as an activation function, and a cross entropy loss function is used in the gradient descending process.
Acquiring output data of the nine-axis inertia measurement unit in a falling state, and calculating to obtain a feature vector; the output data of the nine-axis inertia measurement unit is as follows: accelerometer measurement ax,ay,az(ii) a The gyroscope measured value is omegax,ωy,ωz(ii) a The three-axis magnetometer has m measured valuesx,my,mz. The treatment process comprises the following steps:
(1) initializing quaternions
Wherein,theta and gamma are initial attitude angles of the moving limbs, and the arms of the middle-aged and old people are in a natural sagging state under the static state by combining the actual situation, so that the three initial attitude angles are assumed to be zero.
(2) Converting the measured value of the accelerometer and the measured value of the triaxial magnetometer into a unit vector;
(3) obtaining a gravity vector (v) using a quaternionx,vy,vz)TAnd magnetic field vector (w)x,wy,wz)T
In the attitude calculation process, a transformation matrix from an n system to a b system is set as follows:
the transformation matrix from b to n is:
The magnetometer measures the magnitude and direction of the earth's magnetic field at an angle to each of the coordinate axes of the n-system, denoted as (b)x,by,bz)TThe x-axis, which has been previously set to n, points to the north, so by0, i.e. (b)x,0,bz)T。
Calculating the magnetic field vector needs to be pushed from b to n systems, such thatWherein (m)x,my,mz)TIs the output of the magnetometer in the b-system. Since the geomagnetism meters have the same vector size on the XOY plane of the n system and the b system, the geomagnetism meters can be used for measuring the geomagnetism of the geomagnetismFor bzWithout change, bz=hzThereby obtaining (b)x,0,bz)T。
Obtaining the following result after converting the n system to the b system through the conversion matrix of the b system:
(4) calculating the error (e)x,ey,ez)T
(ex,ey,ez)T=(vx,vy,vz)Tⅹ(ax,ay,az)T+(mx,my,mz)Tⅹ(wx,wy,wz)T
(5) Correction of gyroscope data using errors
=Kp*e+Ki*∫e,ω=ωg+, where KpDenotes the proportional gain, KiRepresenting integral gain, for controlling the rate of convergence, omega, of the gyro deviationg=(ωx,ωy,ωz)T。
(6) Updating quaternion and converting quaternion to euler angles
Solving this differential equation yields:
through the calculation, the attitude angle data of the arm at any moment can be obtained, and the two change values of the attitude angle and the corresponding time variation thereof when the falling event occurs are taken as the characteristic vector. That is, when a suspected fall event occurs, the timer is turned on, the size of the attitude angle reaches the peak/trough for the first time to serve as a first section, and the variation delta phi of the attitude angle of the first section is recorded1、Δθ1、Δγ1And corresponding time variation Δ t1(ii) a The change quantity delta phi of the attitude angle of the section is recorded from the first time of reaching the peak/trough to the second time of reaching the trough/peak as a second section2、Δθ2、Δγ2And corresponding time variation Δ t2. The recorded data constitute an 8-dimensional feature vector.
The suspected fall event is the condition that the combined acceleration is zero and is detected by the accelerometer.
Step four, a pattern matching process; and after normalization preprocessing is carried out on the obtained characteristic vectors, the characteristic vectors are input into the falling detection model, the probability of six falling types of forward falling, backward falling, left falling, right falling, supine falling and prone falling and non-falling situations is obtained through model calculation, and whether falling occurs and the type of falling occurrence are judged according to the probability.
The sensor module is worn on the wrist of a user and is in contact with the skin of the user, comprises a pulse oximeter, a heart rate sensor and a body temperature sensor and is used for measuring the heart rate, the blood oxygen level and the body temperature of the user in real time; maxim's MAX30102 is a high sensitivity pulse oximeter and heart rate sensor that the present invention uses to measure heart rate and blood oxygen levels. Meanwhile, a MAX30205 body temperature sensor was used to measure real-time body temperature.
During the operation of the device, the memory mapping and the initialization of the peripherals are first performed, and the microprocessor can determine the reason for the wake-up of the device when a corresponding event (timer or external interrupt) occurs.
1. If the wake-up source is an inertial measurement module, then it is considered whether it is caused by a fall event. Firstly, the output vector of the fall detection model is obtained, and if the probability of a certain fall event exceeds 50%, the fall event can be judged to occur. When a fall event occurs, the microcontroller sends related information to the mobile phone terminal through the communication module thereof to inform the occurrence of the fall event and display the fall category.
2. If the wake-up source is not an inertial measurement module, this is because the timer measuring one of the three parameters (heart rate, body temperature, blood oxygen level) has expired. The only trigger condition for the expiration of the timer is that the measured values of the three physiological parameters exceed the normally set thresholds. First it is checked which timers have expired and then the measurement procedure is performed again and the data is stored.
And after the data processing is finished, the relevant parameter values are sent to the mobile phone terminal. It is emphasized that the data sent to the handset terminal also comprises measurement intervals for each parameter, so that these measurement intervals are modified in the application. If the transmission fails, it is stored for the next transmission. At the end of each cycle, all measurement timers go to sleep until either the timer expires or an interrupt event occurs.
The microprocessor module includes a single chip microcomputer and peripheral circuits thereof, as shown in fig. 5, the model of the single chip microcomputer in this embodiment is ESP 32. The input end of the microprocessor is connected with the output ends of the sensor module and the inertia measurement module, and the microprocessor is used for establishing an attitude model of a human body falling process according to the acquired sensor data, researching the human body falling process, and extracting characteristic parameters of hand motion changes in several falling attitudes, so that the normal state of the human body and an accidental falling event are accurately distinguished; the microprocessor module needs to be additionally provided with a timer, the timer sets an expiration event, and the only trigger condition of the event is that the input value of the sensor module exceeds a certain threshold value;
in this embodiment, the bluetooth communication module adopts bluetooth 5.0.
The holder of the mobile phone terminal is a family or a care person of the user of the monitoring equipment, the mobile phone terminal receives the posture and the physiological signal of the user sent by the microprocessor module through Bluetooth communication, and if an accident happens, related persons can be reminded to cure the old through sound and character information at the first time.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions and scope of the present invention as defined in the appended claims.
Claims (2)
1. The utility model provides a wearable human sensing monitoring facilities based on nine inertial measurement units which characterized in that includes: the device comprises a power module, a sensor module, a nine-axis inertia measurement unit, a microprocessor module, a Bluetooth communication module and a mobile phone terminal;
the power supply module provides a direct current power supply for the sensor module, the nine-axis inertia measurement unit, the microcontroller module and the Bluetooth communication module;
the nine-axis inertial measurement unit is integrated with a three-axis accelerometer, a three-axis gyroscope and a nine-axis inertial measurement unit of a three-axis magnetometer, is worn on the wrist position of a user, is used for detecting the posture conditions of the user in three directions of a preset three-dimensional coordinate system X, Y and Z, and judges the normal state of the human body and the accidental falling event by resolving multi-source data and combining a mode identification method;
the nine-axis inertia measurement module collects relevant parameters of human hand motion, and the motion parameters are processed to restore the posture of the moving limb in space; describing the movement of a moving limbThe dynamic parameters comprise acceleration, rotation angular velocity of a corresponding joint and magnetic field information of limbs in a corresponding space; firstly, a geographical coordinate system n system and a human body posture coordinate system b system are established, wherein the geographical coordinate system n system is OxnynznSelecting a northeast coordinate system, OxnThe axis pointing to the north, OynThe axis pointing east, OznThe axis points to the sky along the reverse direction of the plumb line, and the original point is the wearing position of the equipment;
the nine-axis inertia measurement module presets the wearing position of the equipment as a human body posture coordinate system b system, namely OxbybzbOrigin of coordinates of, OxbThe axis is directed forward, OybAxis pointing to the left, OzbThe axis points upward; as defined below:
r is the cosine of the true attitude direction,cosine of attitude direction, R, calculated for complementary filtering0For the attitude matrix, mu, observed by the accelerometer and the three-axis magnetometerHIs R0Is observed as noise R0=R+μH,RCAttitude matrix, mu, calculated for gyroscope measurement dataLAccumulating error R for a gyroscopec=R+μLS represents an intermediate operator, 1/s represents integration and s represents differentiation in the circuit diagram;
computingHigh-frequency components of the accelerometer and the magnetometer are filtered out firstly, and low-frequency components of the gyroscope are filtered out again, wherein:
GL(s) and GH(s) is the transfer function of the complementary filter, GL(s) having a first-order low-pass filtering characteristic, GH(s) having a first order high pass filtering characteristic, capable of eliminating the accumulation of high frequency noise and low frequency error by complementary filtering;
GL(s) and GH(s) complementary to give:
the sensor module is worn on the wrist of a user and is in contact with the skin of the user, comprises a high-sensitivity pulse oximeter, a heart rate sensor and a body temperature sensor and is used for measuring the heart rate, the blood oxygen level and the body temperature of the user in real time;
the microprocessor module comprises a control circuit, the input end of the control circuit is connected with the output ends of the sensor module and the nine-axis inertia measurement unit, and the control circuit is used for establishing an attitude model of a human body falling process according to the acquired sensor data, researching the human body falling process, extracting characteristic parameters of hand motion changes in several falling attitudes, and processing the parameters through a falling detection model so as to accurately distinguish the normal state of the human body from accidental falling events; the microprocessor module is additionally provided with a timer and a timer, the timer is used for setting an expiration event, and the only trigger condition of the event is that the input value of the sensor module exceeds a certain threshold value;
the mobile phone terminal receives the posture and physiological signals of the user sent by the microprocessor module through Bluetooth communication, and reminds the mobile phone terminal holder of the falling old man to cure the falling old man through sound and character information at the first time.
2. The wearable human body sensing monitoring device based on the nine-axis inertial measurement unit according to claim 1, wherein the fall detection algorithm based on the nine-axis inertial measurement unit comprises the following specific steps:
designing and manufacturing an arm posture data set of the old people in a falling state, wherein the data set is divided into a training set and a testing set; the arm posture data set comprises relevant feature vectors for fall detection and is provided with fall/non-fall labels, the fall labels can be subdivided into six categories of forward fall, backward fall, left side fall, right side fall, supine fall and prone fall, and the data set is used for training the fall detection model;
step two, constructing a falling detection model based on a deep neural network, wherein the input of the model is a feature vector related to falling, and the output is the probability of falling types and the probability of occurrence of various non-falling situations; training a falling detection model by using the training set in the arm posture data set to obtain model parameters, testing the performance of the model by using the test set in the arm posture data set and adjusting the model parameters;
acquiring output data of the nine-axis inertia measurement unit in a falling state, and calculating to obtain a feature vector; the output data of the nine-axis inertia measurement unit is as follows: accelerometer measurement ax,ay,az(ii) a The gyroscope measured value is omegax,ωy,ωz(ii) a The three-axis magnetometer has m measured valuesx,my,mz(ii) a The treatment process comprises the following steps:
(1) initializing quaternions
Wherein,theta and gamma are initial attitude angles of the moving limbs, the arms of the user are in a natural sagging state in a static state, and the three initial attitude angles are assumed to be zero;
(2) converting the measured value of the accelerometer and the measured value of the triaxial magnetometer into a unit vector;
(3) obtaining a gravity vector (v) using a quaternionx,vy,vz)TAnd magnetic field vector (w)x,wy,wz)T
the three-axis magnetometer measures the magnitude and direction of the earth's magnetic field at an angle to each coordinate axis of a geographic coordinate system n, denoted (b)x,by,bz)TThe x-axis of the n-system of the geographic coordinate system points to the north, so by0, i.e. (b)x,0,bz)T;
Calculating the magnetic field vector needs to be pushed from b to n systems, such thatWherein (m)x,my,mz)TIs the output of the magnetometer in system b; since the geomagnetism meters have the same vector size on the XOY plane of the n system and the b system, the geomagnetism meters can be used for measuring the geomagnetism of the geomagnetismFor bzWithout change, bz=hzThereby obtaining (b)x,0,bz)T;
Obtaining the following result after converting the n system to the b system through the conversion matrix of the b system:
wherein,a transformation matrix between a geographic coordinate system and a carrier coordinate system;
(4) calculation of error eI.e. (e)x,ey,ez)T
(ex,ey,ez)T=(vx,vy,vz)Tⅹ(ax,ay,az)T+(mx,my,mz)Tⅹ(wx,wy,wz)T
(5) Correction of gyroscope data using errors
=Kp*e+Ki*∫e,ω=ωg+, where KpDenotes the proportional gain, KiRepresenting integral gain, for controlling the rate of convergence, omega, of the gyro deviationg=(ωx,ωy,ωz)T;
(6) Updating quaternion and converting quaternion to euler angles
Solving this differential equation yields:
through the calculation, the attitude angle data of the arm at any moment can be obtained, and the two change values of the attitude angle and the corresponding time variation thereof when a falling event occurs are taken as the characteristic vectors; that is, from the occurrence of a suspected fall event to the first arrival of the peak/trough of the attitude angle as a first segment, the variation Δ φ of the attitude angle of the segment is recorded1、Δθ1、Δγ1And when it is in responseAmount of variation between delta t1(ii) a The change quantity delta phi of the attitude angle of the section is recorded from the first time of reaching the peak/trough to the second time of reaching the trough/peak as a second section2、Δθ2、Δγ2And corresponding time variation Δ t2(ii) a The recorded data form an 8-dimensional feature vector;
the suspected fall event is the condition that the combined acceleration is zero and is detected by an accelerometer;
step four, a pattern matching process; and after normalization preprocessing is carried out on the obtained characteristic vectors, the characteristic vectors are input into the falling detection model, the probability of different falling types and the probability of non-falling situations are obtained through model calculation, and whether falling occurs and the types of falling are judged according to the probability.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010530351.9A CN111693024A (en) | 2020-06-11 | 2020-06-11 | Wearable human body sensing monitoring equipment based on nine-axis inertia measurement unit |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010530351.9A CN111693024A (en) | 2020-06-11 | 2020-06-11 | Wearable human body sensing monitoring equipment based on nine-axis inertia measurement unit |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111693024A true CN111693024A (en) | 2020-09-22 |
Family
ID=72480484
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010530351.9A Pending CN111693024A (en) | 2020-06-11 | 2020-06-11 | Wearable human body sensing monitoring equipment based on nine-axis inertia measurement unit |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111693024A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113244595A (en) * | 2021-04-23 | 2021-08-13 | 南京布袋健康管理有限公司 | Action explosive force analysis method and system |
CN113287805A (en) * | 2021-05-27 | 2021-08-24 | 郑州大学 | Fall protection method and wearable fall protection device |
CN113520397A (en) * | 2021-07-23 | 2021-10-22 | 吉林大学 | Driving distraction behavior identification method based on wearable inertial measurement unit |
CN114111777A (en) * | 2021-11-24 | 2022-03-01 | 中国矿业大学(北京) | Underground personnel state sensing system based on head posture monitoring |
CN114343641A (en) * | 2022-01-24 | 2022-04-15 | 广州熠华教育咨询服务有限公司 | Learning difficulty intervention training guidance method and system thereof |
CN114967942A (en) * | 2022-03-31 | 2022-08-30 | 河北迥然科技有限公司 | Motion attitude analysis method, terminal and storage medium |
CN115462782A (en) * | 2022-09-06 | 2022-12-13 | 四川康谷康业科技有限公司 | Human body falling dynamic monitoring method and system based on multi-dimensional characteristic parameters |
CN116189382A (en) * | 2023-02-01 | 2023-05-30 | 观云(山东)智能科技有限公司 | Fall detection method and system based on inertial sensor network |
WO2023221524A1 (en) * | 2022-05-20 | 2023-11-23 | 北京航天时代光电科技有限公司 | Human movement intelligent measurement and digital training system |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8762091B1 (en) * | 2010-06-08 | 2014-06-24 | Thales Visionix, Inc. | Inertial measurement system |
CN104473648A (en) * | 2014-09-24 | 2015-04-01 | 上海大学 | Physiological parameter monitoring-combined human body tumble warning and detecting method |
CN104567931A (en) * | 2015-01-14 | 2015-04-29 | 华侨大学 | Course-drifting-error elimination method of indoor inertial navigation positioning |
CN105590409A (en) * | 2016-02-26 | 2016-05-18 | 江苏大学 | Human body tumble detection method and human body tumble detection system based on big data |
CN106679649A (en) * | 2016-12-12 | 2017-05-17 | 浙江大学 | Hand movement tracking system and tracking method |
CN107169512A (en) * | 2017-05-03 | 2017-09-15 | 苏州大学 | The construction method of HMM SVM tumble models and the fall detection method based on the model |
CN108061855A (en) * | 2017-11-30 | 2018-05-22 | 天津大学 | A kind of globular motor rotor position detecting method based on MEMS sensor |
CN110047248A (en) * | 2019-05-23 | 2019-07-23 | 河南工业大学 | It is a kind of based on wearable Falls Among Old People detection method and equipment |
CN110889330A (en) * | 2019-10-30 | 2020-03-17 | 西北工业大学 | BP neural network-based old people tumbling detection method and system |
-
2020
- 2020-06-11 CN CN202010530351.9A patent/CN111693024A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8762091B1 (en) * | 2010-06-08 | 2014-06-24 | Thales Visionix, Inc. | Inertial measurement system |
CN104473648A (en) * | 2014-09-24 | 2015-04-01 | 上海大学 | Physiological parameter monitoring-combined human body tumble warning and detecting method |
CN104567931A (en) * | 2015-01-14 | 2015-04-29 | 华侨大学 | Course-drifting-error elimination method of indoor inertial navigation positioning |
CN105590409A (en) * | 2016-02-26 | 2016-05-18 | 江苏大学 | Human body tumble detection method and human body tumble detection system based on big data |
CN106679649A (en) * | 2016-12-12 | 2017-05-17 | 浙江大学 | Hand movement tracking system and tracking method |
CN107169512A (en) * | 2017-05-03 | 2017-09-15 | 苏州大学 | The construction method of HMM SVM tumble models and the fall detection method based on the model |
CN108061855A (en) * | 2017-11-30 | 2018-05-22 | 天津大学 | A kind of globular motor rotor position detecting method based on MEMS sensor |
CN110047248A (en) * | 2019-05-23 | 2019-07-23 | 河南工业大学 | It is a kind of based on wearable Falls Among Old People detection method and equipment |
CN110889330A (en) * | 2019-10-30 | 2020-03-17 | 西北工业大学 | BP neural network-based old people tumbling detection method and system |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113244595A (en) * | 2021-04-23 | 2021-08-13 | 南京布袋健康管理有限公司 | Action explosive force analysis method and system |
CN113287805A (en) * | 2021-05-27 | 2021-08-24 | 郑州大学 | Fall protection method and wearable fall protection device |
CN113520397A (en) * | 2021-07-23 | 2021-10-22 | 吉林大学 | Driving distraction behavior identification method based on wearable inertial measurement unit |
CN114111777A (en) * | 2021-11-24 | 2022-03-01 | 中国矿业大学(北京) | Underground personnel state sensing system based on head posture monitoring |
CN114343641A (en) * | 2022-01-24 | 2022-04-15 | 广州熠华教育咨询服务有限公司 | Learning difficulty intervention training guidance method and system thereof |
CN114967942A (en) * | 2022-03-31 | 2022-08-30 | 河北迥然科技有限公司 | Motion attitude analysis method, terminal and storage medium |
WO2023221524A1 (en) * | 2022-05-20 | 2023-11-23 | 北京航天时代光电科技有限公司 | Human movement intelligent measurement and digital training system |
CN115462782A (en) * | 2022-09-06 | 2022-12-13 | 四川康谷康业科技有限公司 | Human body falling dynamic monitoring method and system based on multi-dimensional characteristic parameters |
CN116189382A (en) * | 2023-02-01 | 2023-05-30 | 观云(山东)智能科技有限公司 | Fall detection method and system based on inertial sensor network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111693024A (en) | Wearable human body sensing monitoring equipment based on nine-axis inertia measurement unit | |
Brigante et al. | Towards miniaturization of a MEMS-based wearable motion capture system | |
CN106815857B (en) | Gesture estimation method for mobile auxiliary robot | |
US20180132761A1 (en) | Use of epidermal electronic devices to measure orientation | |
Kong et al. | Development of a real-time IMU-based motion capture system for gait rehabilitation | |
Liu et al. | A wearable human motion tracking device using micro flow sensor incorporating a micro accelerometer | |
US20130116602A1 (en) | Automatic orientation calibration for a body-mounted device | |
CN208677399U (en) | Intelligent switch joint angle measurement unit and system | |
CN106618584B (en) | A method of for monitoring user's lower extremity movement | |
CN111895997B (en) | Human body action acquisition method based on inertial sensor without standard posture correction | |
Yi et al. | Wearable sensor data fusion for remote health assessment and fall detection | |
CN103417217A (en) | Joint mobility measuring device and measuring method thereof | |
CN205003738U (en) | Healthy management system of self -service community | |
CN106308810A (en) | Human motion capture system | |
Sabatini | Inertial sensing in biomechanics: a survey of computational techniques bridging motion analysis and personal navigation | |
Wang et al. | A new physiological signal acquisition patch designed with advanced respiration monitoring algorithm based on 3-axis accelerator and gyroscope | |
Yang et al. | Smart wearable monitoring system based on multi-type sensors for motion recognition | |
Cotton et al. | Wearable monitoring of joint angle and muscle activity | |
Johnson et al. | Estimation of three-dimensional thoracoabdominal displacements during respiration using inertial measurement units | |
Wong et al. | Measurement of postural change in trunk movements using three sensor modules | |
Madrigal et al. | Hip and lower limbs 3D motion tracking using a double-stage data fusion algorithm for IMU/MARG-based wearables sensors | |
CN113229806A (en) | Wearable human body gait detection and navigation system and operation method thereof | |
WO2024124638A1 (en) | Wearable single-node device for leg posture estimation and measurement method | |
Kumar et al. | A wearable inertial sensing device for fall detection and motion tracking | |
Alonge et al. | Use of accelerometers and gyros for hip and knee angle estimation |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200922 |