CN110025318A - A kind of tumble real-time detecting system and method based on acceleration transducer - Google Patents
A kind of tumble real-time detecting system and method based on acceleration transducer Download PDFInfo
- Publication number
- CN110025318A CN110025318A CN201810027142.5A CN201810027142A CN110025318A CN 110025318 A CN110025318 A CN 110025318A CN 201810027142 A CN201810027142 A CN 201810027142A CN 110025318 A CN110025318 A CN 110025318A
- Authority
- CN
- China
- Prior art keywords
- data
- tumble
- module
- acceleration
- acceleration transducer
- 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
Classifications
-
- 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
- A61B5/1117—Fall detection
-
- 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
-
- 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
- A61B5/6813—Specially adapted to be attached to a specific body part
- A61B5/6823—Trunk, e.g., chest, back, abdomen, hip
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2503/00—Evaluating a particular growth phase or type of persons or animals
- A61B2503/08—Elderly
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physiology (AREA)
- Dentistry (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a kind of tumble real-time detecting system and method based on acceleration transducer.The system includes acceleration transducer, wireless transmitter, wireless receiver, data processing module, power module, warning module, system reset module and PC machine, and data processing module includes data pre-processing unit and tumble model.Method are as follows: the system is worn on the waist i.e. position of hipbone upper end of human body, acceleration transducer acquires the acceleration of three axis during human motion, and it is transferred to data processing module, data are pre-processed by data pre-processing unit, then tumble model is substituted into, judge whether to be in during tumble, starts warning module if falling, system is resetted by reseting module；Treated data and final testing result are emitted by wireless transmitter and are received by wireless receiver, are finally transmitted to PC machine and carry out data monitoring.The present invention is able to achieve the real-time detection to tumble, ensures the health and safety of institute's Monitoring Population.
Description
Technical field
The present invention relates to control system technical field, especially a kind of tumble real-time detection system based on acceleration transducer
System and method.
Background technique
Increasingly developed with China's economy, aging trend is got worse, and correlation study is studies have shown that the year two thousand thirty sky
Specific gravity shared by nest old man family will be up to 90%, and empty nesting will be presented in the Old Age Homes in China at that time, and simultaneously, the elderly
Tumble incidence is high, due to senile osteoporosis and is easy to fracture, the consequence fallen is serious.At present fall at
The 4th on the reason of causing injures and deaths for China list, at 65 years old and be then the 1st in the crowd of the above age bracket.With the age
Growth, Falls in Old People is easy to cause many diseases, gently then causes to scratch or even fracture, heavy then cause to go into a coma and the cerebrovascular
It is unexpected etc., it is also possible to cause paralysis even dead under more serious situation.The elderly can generate pavid psychology after tumble, this
The feared state of mind substantially reduces the mobility that will lead to the elderly, and then its scope of activities is caused to reduce, and influences the elderly's
Quality of life.
After Falls in Old People very big negative effect can be caused in physiology and psychological levels.Tested person is detected in real time
Member is falling, and issues warning signal, and convenient for a series of execution of subsequent safeguard procedures, effectively can reduce and prevent
It falls and the elderly's bring is injured.The algorithm of fall detection common pattern identification at present, but algorithm for pattern recognition is related to
The calculating of higher dimensional space causes calculation amount very big, so the requirement to hardware is very high, real-time is bad.
Summary of the invention
The purpose of the present invention is to provide a kind of real-times, and good, the easy reliably tumble based on acceleration transducer is real-time
Detection system and method,
The technical solution for realizing the aim of the invention is as follows: a kind of tumble real-time detection system based on acceleration transducer
System, including acceleration transducer, wireless transmitter, wireless receiver, data processing module, power module, warning module, system
Reseting module and PC machine, data processing module include data pre-processing unit and tumble model；Acceleration transducer accesses data
The data output end of processing module, data processing module is connect with wireless transmitter, and the output end and PC machine of wireless receiver connect
It connects, warning module is connect with data processing module；
The system is worn on the waist i.e. position of hipbone upper end of human body, is powered by electric power source pair of module whole system,
Acceleration transducer acquires the acceleration of tri- axis of X, Y, Z during human motion, and data are transferred to data processing by SPI
Module pre-processes data by the data pre-processing unit in data processing module, then substitutes into tumble model, and judgement is
It is no to be in during tumble, start warning module if falling, system is resetted by reseting module；Entire real-time
During detection, treated data and final testing result are emitted by wireless transmitter and are connect by wireless receiver
It receives, is finally transmitted to PC machine and carries out data monitoring.
Further, the acceleration transducer is ADXL345 acceleration transducer, and the Z axis of acceleration transducer
Perpendicular to ground.
Further, the main control chip of the data processing module be LPC1114, auxiliary circuit include reset circuit,
Power module, oscillating circuit and JTAG debugging circuit module.
Further, the power module uses power supply of the chargeable 5V lithium battery as whole system.
Further, the wireless transmitter and wireless receiver use the modulation demodulation system of GFSK to carry out data biography
It is defeated.
Further, the wireless receiver uses USB interface wireless data transfer module.
A kind of tumble real-time detection method based on acceleration transducer, comprising the following steps:
Step 1, by the tumble real-time detecting system based on acceleration transducer, it is worn on the waist i.e. hipbone upper end of human body
Position, acceleration transducer acquires the acceleration of tri- axis of X, Y, Z during human motion, and data is transferred to by SPI
Data processing module；
Step 2, the data pre-processing unit in data processing module pre-processes data, then substitutes into tumble mould
Type judges whether to be in during tumble, starts warning module if falling, is resetted by reseting module to system；
Step 3, during entire real-time detection, treated data and final testing result are by wireless hair
Emitter emits and is received by wireless receiver, is finally transmitted to PC machine and carries out data monitoring.
Further, the data pre-processing unit in data processing module described in step 2 pre-processes data, so
Tumble model is substituted into afterwards, judges whether to be in during tumble, specific as follows:
It is firstly introduced into a total acceleration A0As characteristic quantity:
Wherein ax、ay、azThe respectively acceleration of tri- axis of X, Y, Z of acceleration transducer acquisition；
Secondly, by continuously acquire 5~10 total acceleration A0It averages to obtain V0；
Finally, establishing tumble model according to following three criterions:
1) resultant acceleration threshold criterion
Threshold value A is set1=2g, as resultant acceleration A0Greater than A1When enter judgement 2), otherwise continue to acquire data, g attaches most importance to
Power acceleration；
2) slope criterion
Select S1Threshold value of=the 25g/s as slope, when detect resultant acceleration slope data S be greater than S1When, into 3)
Judgement, otherwise continue acquire data；
3) average acceleration criterion
Select V3=3.9g is as average value judgment basis, as the data V by average value processing0Greater than V3When, then finally sentence
Break to fall.
Compared with prior art, the present invention its remarkable advantage is: (1) being developed based on embedded, having can cut
Property, hard real-time, stiff stability, weak interactivity the characteristics of, it is easy to operate it is simple, friendly figure GUI and graphical interfaces are provided
With provide powerful network function be various mobile computing device spare interfaces, and have better hardware adaptive mechanism, have compared with
Long life cycle；(2) 3-axis acceleration sensor ADXL345 is used, has precision high, low in energy consumption, adjustable sampling is fast
The mode of rate and measurement, can measure dynamic acceleration under movement or even vibration environment, and portability, real-time are good；(3) fall
The method that reverse mould type uses threshold decision, can effectively realize and rapidly judge tumble, and accuracy is high, and can fall
It detects to fall during, is conducive to instant take safeguard procedures；(4) chargeable 5V lithium battery, energy-saving ring are used
It protects, it is easy to carry, be conducive to sustainable development；Compact and layout is simple, easy to carry, sexual valence is high；(5) it may be implemented pair
The detection of the elderly's motion state records the daily routines situation of the elderly and sounds an alarm in time when unsafe condition occurs,
So as to the execution of follow-up, is conducive to the health and safety for ensureing the elderly, improves the quality of life of the elderly.
Detailed description of the invention
Fig. 1 is the hardware block diagram of the tumble real-time detecting system the present invention is based on acceleration transducer.
Fig. 2 is micro controller module hardware block diagram in the present invention.
Fig. 3 is 3-axis acceleration sensor module hardware structural block diagram in the present invention.
Fig. 4 is serial ports sending and receiving data process figure in the present invention, wherein (a) is to send data flowchart, it is (b) reception data
Flow chart.
Fig. 5 is SPI communication agreement timing diagram in the present invention.
Fig. 6 is the schematic diagram of tumble model in the present invention.
Fig. 7 is software overview flow chart of the invention.
Specific embodiment
The present invention is based on the tumble real-time detecting systems of acceleration transducer, including acceleration transducer 1, wireless transmitter
2, wireless receiver 3, data processing module 4, power module 5, warning module 6, system reset module 7 and PC machine 8, data processing
Module 4 includes data pre-processing unit and tumble model；Acceleration transducer 1 accesses data processing module 4, data processing module
4 data output end is connect with wireless transmitter 2, and the output end of wireless receiver 3 is connect with PC machine 8, warning module 6 and data
Processing module 4 connects；
The system is worn on the waist i.e. position of hipbone upper end of human body, is powered by power module 6 to whole system,
Acceleration transducer 1 acquires the acceleration of tri- axis of X, Y, Z during human motion, and data are transferred at data by SPI
Module 4 is managed, data are pre-processed by the data pre-processing unit in data processing module 4, tumble model is then substituted into, sentences
It is disconnected whether to be in during tumble, start warning module 6 if falling, system is resetted by reseting module 7；Whole
During a real-time detection, treated data and final testing result are emitted by wireless transmitter 2 and by wireless
Receiver 3 receives, and is finally transmitted to PC machine 8 and carries out data monitoring.
Further, the acceleration transducer 1 is ADXL345 acceleration transducer, and the Z axis of acceleration transducer
Perpendicular to ground.
Further, the main control chip of the data processing module 4 be LPC1114, auxiliary circuit include reset circuit,
Power module, oscillating circuit and JTAG debugging circuit module.
Further, power supply of the power module 5 using chargeable 5V lithium battery as whole system.
Further, the wireless transmitter 2 and wireless receiver 3 use the modulation demodulation system of GFSK to carry out data
Transmission.
Further, the wireless receiver 3 uses USB interface wireless data transfer module.
A kind of tumble real-time detection method based on acceleration transducer, comprising the following steps:
Step 1, by the tumble real-time detecting system based on acceleration transducer, it is worn on the waist i.e. hipbone upper end of human body
Position, acceleration transducer 1 acquires the acceleration of tri- axis of X, Y, Z during human motion, and data is transferred to by SPI
Data processing module 4；
Step 2, the data pre-processing unit in data processing module 4 pre-processes data, then substitutes into tumble mould
Type judges whether to be in during tumble, starts warning module 6 if falling, is answered by reseting module 7 system
Position；
Step 3, during entire real-time detection, treated data and final testing result are by wireless hair
Emitter 2 emits and is received by wireless receiver 3, is finally transmitted to PC machine 8 and carries out data monitoring.
Further, the data pre-processing unit in data processing module 4 described in step 2 pre-processes data,
Then tumble model is substituted into, judges whether to be in during tumble, specific as follows:
It is firstly introduced into a total acceleration A0As characteristic quantity:
Wherein ax、ay、azThe respectively acceleration of tri- axis of X, Y, Z of acceleration transducer acquisition；
Secondly, by continuously acquire 5~10 total acceleration A0It averages to obtain V0；
Finally, establishing tumble model according to following three criterions:
1) resultant acceleration threshold criterion
Threshold value A is set1=2g, as resultant acceleration A0Greater than A1When enter judgement 2), otherwise continue to acquire data, g attaches most importance to
Power acceleration；
2) slope criterion
Select S1Threshold value of=the 25g/s as slope, when detect resultant acceleration slope data S be greater than S1When, into 3)
Judgement, otherwise continue acquire data；
3) average acceleration criterion
Select V3=3.9g is as average value judgment basis, as the data V by average value processing0Greater than V3When, then finally sentence
Break to fall.
With reference to the accompanying drawing and specific embodiment is further elaborated with the present invention.
Embodiment 1
The present embodiment is that the collected human body movement data of acceleration transducer ADXL345 institute is transferred to microcontroller
Several threshold values are obtained by data processing later, are compared with tumble model, to realize the quick reality to Falls in Old People
When detection, if it is detected that falling, buzzer sounds an alarm indication and falls, and is answered after the completion of whole flow process by reset system
Position starts the data acquisition of a new round and judges work.
1. the Falls in Old People real-time detecting system Design of Hardware Architecture based on acceleration transducer ADXL345
Old people's air defense falls the real-time hardware design structure of falling detection system block diagram as shown in Figure 1, being made of following module:
Acceleration transducer 1, wireless transmitter 2, wireless receiver 3, data processing module 4, power module 5, warning module 6, system
Reseting module 7 and PC machine 8, data processing module 4 include data pre-processing unit and tumble model；Acceleration transducer 1 accesses
The data output end of data processing module 4, data processing module 4 is connect with wireless transmitter 2, the output end of wireless receiver 3
It is connect with PC machine 8, warning module 6 is connect with data processing module 4.
Development platform of the microcontroller based on ARM technology and KEIL in acceleration transducer 1, by being based on MEMS
(MEMS) 3-axis acceleration sensor of technology carries out the acquisition of 3-axis acceleration data, and collected data are led to by SPI
Letter agreement is transferred to the main control chip LPC1114 of data processing module 4, carries out the processing of data simultaneously by main control chip LPC1114
Treated data are compared and are judged with the tumble model in Fig. 6, judging result is exported in a manner of alarm；It falls
And after early warning, the reset to whole system is realized by reseting module, restarts to work normally.It is built by wireless transceiver module
The vertical communication with PC machine, by whole process data and alarm record be transferred to PC machine, realize to the elderly's motion state
Detection and data storage.
2. acceleration transducer
3-axis acceleration sensor is ADXL345.ADXL345 is the ADI company three axis accelerometer released in 2008, it
Using MEMS technology, has the function of SPI and IIC numeral output.The hardware circuit of 3-axis acceleration sensor as shown in Figure 2
Figure, is powered by 3.3V power supply, is initialized by program to 3-axis acceleration device module, passes through past corresponding register
Write instruction configures it.Its measurement range is ± 16g, and interrupt signal is that high level is effective, and normal mode is measurement
Mode.ADXL345 can measure 3-axis acceleration and angle corresponding to 3-D walls and floor X, Y, Z axis (three weeks vertical two-by-two)
Information, but since ADXL345 angle collected is static angular information, it is not suitable for dynamical system, and ADXL345 can transported
Dynamic acceleration is measured under dynamic or even vibration environment, so the acceleration information using three axis acquires data as effective,
And it is inputted 3-axis acceleration information collected as the data of microcontroller.
3. data processing module
Data processing module 4 following components as shown in figure 3, be made of: main control chip LPC1114, reset circuit, vibration
Swing circuit and JTAG debugging circuit.Power module 5 is the power supply of data processing module 4.
1) main control chip
Main control chip uses LPC1114, and the kernel of LPC1100 series A RM is Cortex-M0, be with high-performance,
Easy to use and low-power consumption feature 32-bit microprocessor.LPC1100 is the solution of 32 most cheap 8-digit microcontrollers, it
Cost performance and ease for use it is more outstanding compared to existing 8/16 microcontroller, and can significantly shorten 8/
The length of the code of 16 applications.LPC1100 family chip can not only carry out the control on basis, can also realize complicated fortune
It calculates, the efficient running of the chip efficiently reduces the energy consumption of system.The frequency of LPC1100 chip is 50MHz, the chip
The electric current for only needing to be less than 10mA can work.
2) power module
Power module uses the chip of model SPX3819M5-3.3, which has low noise, low cost, efficiency
The advantage that high, static electric current is small, circuit is simple.The power module uses the power supply of 5V as input, is then converted to 3.3V
Output voltage give main control chip power supply.
3) reset circuit
The chip that reset circuit uses is CAT811TBI.LPC1114 chipReset pin low level is effective,
Need to guarantee that reset pin keeps a period of time low level when system electrification, CAT811TBI can satisfy this demand.This
Outside,Pin is the input pin of hand-reset, and the pin low level effectively (triggering reset signal), the signal of reset needs
Keep andLow level time consistency, simultaneouslyAfter pin restores high level, need to continue for 140ms.
4) oscillating circuit
As shown in Figure 3, XTALIN and XTALOUT pin connected be data processing module oscillating circuit, be
System oscillating circuit connects external 12M crystal oscillator.After system reset, main control chip takes the lead in using internal RC oscillator, which is to pass through
Software exchange, this characteristic can allow the boot loader of system not influenced by external crystal oscillator, run on given frequency it
Under.The clock of different peripheral and memory by the distribution of system clock, the timer of SPI0/1, SysTick and UART
Clock dividers can obtain suitable peripheral clock.
5) JTAG debugging circuit
J-Link is SEGGER company for the JTAG emulator for supporting emulation ARM core chip to be released.Cooperate IAR
The Integrated Development Environment of EWAR, WINARM, KEIL, ADS, RealView etc. support ARM7/ARM9/ARM11, Cortex M0/
The emulation of the core chips such as M1/M3/M4, Cortex A4/A8/A9 may be implemented with KEIL translation and compiling environment used in the present invention
Seamless connection, it is easy to connect, easy to operate and easy to learn.If J1 is the connection jaws of J-Link converter in Fig. 3, J- is used
Link converter can will compile errorless program and download in chip in KEIL runs.
5. data communication module
The communication of data is RS232 interface and SPI interface with interface in the present invention.
1) serial ports sending and receiving data process figure
In the present invention, using UART interface as the output channel of data, software flow pattern is as shown in Figure 4.Complete
After arriving at the acquisition and timing of data, pass through UART serial ports output data groups.The initial work of UART includes will
I/O pin function is set as UART serial ports, and baud rate and operating mode is arranged.UART serial ports is queue structure when transmitting data
(can be used as buffer area), when judge queue for it is empty or full when, UART serial ports informs that main control chip sends or connects using interrupt mode
Harvest data.
2) SPI communication agreement timing diagram
The standard four-wire system SPI hardware interface of LPC1114 used by system can be connected with most of microcontrollers, and
It is convenient to operate, and development efficiency is high.Such as the timing diagram that Fig. 5 is SPI communication.Determined with CPOL the clock signal of SCK it is idle when
The level of time, CPOL=0, level is low level when idle；When CPOL=1, level is high level when idle.It is determined with CPHA
The time of sampling, CPHA=0 are sampled on first clock edge of each period；CPHA=1, at each period second
Clock edge is sampled.
6. tumble model
Since human body is flexible structure, in the daily physical activity including falling, each joint and position are all
Have the opposite variation of position, thus the test point of tumble model need to be arranged to human body daily routines reflection it is the most steady it
Place.In the normal motion of human motion, waist (the most upper edge of hipbone) position is can be in the stable state of comparison
, posture changes acutely if falling, so measurement basic point is put in the position, is illustrated in fig. 6 shown below, is human 3d model
Figure, using the top (i.e. origin is in the position of waistband) of hipbone, is about the z axis yaw, is to bow around X-axis in the detection of this system
It faces upward, and is roll around Y-axis.Test point be arranged in waist location, can be relatively stable obtain test data, be conducive to subsequent
Data processing.
During daily walk, it is assumed that in the ideal case, that is, ignore human body and tilted as caused by body and core
The factors such as the measurement error of piece itself, when three axis of 3-axis acceleration sensor are completely coincident with the total balance of the body model established
When, human body X, Y-axis acceleration should be fluctuated near 0G, and the acceleration of Z axis should fluctuate near 1G.Due to the radian of human body
And the reasons such as displacement caused by loosening when wearing, human body have certain deviation when wearing, this deviation can pass through note
Then initial condition data when recording chip powers on (when just wearing) compensates data to eliminate, has no effect on number
According to analysis and modeling.
7. Falls in Old People real-time detecting system software flow
3-axis acceleration sensor is initialized by calling initialization function ADXL345_Init () first, and
Its output frequency (200Hz), resolution ratio (13) and range (± 16g) etc. are configured.System clock is set as 10ms,
Every 10ms, serial ports output identification just adds one and three axis accelerometer acquires a data, because the output frequency of serial ports is
40Hz, so serial ports output identification resets and export the number that accelerometer this time acquires when serial ports output identification is greater than 5
According to.Then data substitution tumble model is judged, when meeting the model, that is, tumble alarm is carried out, in order in next step
Protection；When be judged as do not meet model when, then repeated data sample the step for.
In conjunction with Fig. 7, the method that the tumble model uses threshold decision, by three summed up according to the experimental data
Criterion composition, is threshold value, slope and the average value of resultant acceleration respectively.
In the present system, by acceleration transducer can directly it is collected be three axis acceleration information, in ideal feelings
Under condition, i.e., using the top of hipbone (i.e. origin is in the position of waistband), it is about the z axis yaw, is pitching around X-axis, and is cross around Y-axis
When rolling.Test chip is when fully horizontally wearing, X under human body stationary state, Y-axis acceleration should be in 0G, and the acceleration of Z axis is answered
In 1G.When actual wear the case where 3-axis acceleration with ideal model under state when wearing there are certain error, but human body
Total acceleration will not be affected by these factors, though wear state, total acceleration is in 1G, institute when static
In addition to introduce a total acceleration A here0As characteristic quantity.
Secondly, by continuously acquire 5~10 total acceleration A0It averages to obtain V0；
On this basis, tumble model includes following three criterions: resultant acceleration threshold criterion, slope criterion in this system
With average acceleration criterion.
1) resultant acceleration threshold criterion
It is found by the analysis to experimental data, human body horizontal stationary stands up, sits down and lie down that these are more slow
The movable resultant acceleration A of daily behavior0All in 2g hereinafter, and corresponding, resultant acceleration A when tumble0All in 2g or more, mostly
Number tumble situation is generally in 2.5g or more, so passing through setting threshold value A1=2g judges to fall；As resultant acceleration A0Greater than A1When
Judge into next step, otherwise continues to acquire data.
2) slope criterion
It is found by the analysis to experimental data, it, all can be in pole after resultant acceleration often reaches a minimum point when tumble
There is the rising of certain amplitude in the short time (generally within 0.2s), so taking ascensional range and the ratio of time (i.e. tiltedly
Rate S1) as the foundation for further discriminating between and judging.After largely testing and being compared analysis to experimental data, choosing
Select S1Threshold value of=the 25g/s as slope, when detect resultant acceleration slope data S be greater than S1When, judge into next step,
Otherwise continue to acquire data.
3) average acceleration criterion
It is found by the analysis to experimental data, the average value of the acceleration in daily routines certain time should be one
A stationary value nearby fluctuates, but when falling, can be seen that resultant acceleration A according to its curve0Should have one it is obvious
Rising or downward trend and to continue for some time interior fluctuation larger, so the place averaged two-by-two of data will be acquired
Reason, draws corresponding data and curves after obtaining new data.By comparing and analyzing selection V3=3.9g judges as average value
Foundation, as the data V by average value processing0Greater than V3When, then finally it is judged as and is falling.
After tumble model is finally judged to be in tumble process, warning module will issue early warning alarm more to indicate
It falls, in order to the execution of subsequent safeguard measure.
Claims (8)
1. a kind of tumble real-time detecting system based on acceleration transducer, which is characterized in that including acceleration transducer (1),
Wireless transmitter (2), wireless receiver (3), data processing module (4), power module (5), warning module (6), system reset
Module (7) and PC machine (8), data processing module (4) include data pre-processing unit and tumble model；Acceleration transducer (1)
It accesses data processing module (4), the data output end of data processing module (4) is connect with wireless transmitter (2), wireless receiver
(3) output end is connect with PC machine (8), and warning module (6) is connect with data processing module (4)；
The system is worn on the waist i.e. position of hipbone upper end of human body, is powered, is added to whole system by power module (6)
Velocity sensor (1) acquires the acceleration of tri- axis of X, Y, Z during human motion, and data are transferred at data by SPI
It manages module (4), data is pre-processed by the data pre-processing unit in data processing module (4), then substitute into tumble mould
Type judges whether to be in during tumble, starts warning module (6) if falling, and is carried out by reseting module (7) to system
It resets；During entire real-time detection, treated data and final testing result are sent out by wireless transmitter (2)
It penetrates and is received by wireless receiver (3), be finally transmitted to PC machine (8) and carry out data monitoring.
2. the tumble real-time detecting system according to claim 1 based on acceleration transducer, which is characterized in that described
Acceleration transducer (1) is ADXL345 acceleration transducer, and the Z axis of acceleration transducer is perpendicular to ground.
3. the tumble real-time detecting system according to claim 1 based on acceleration transducer, which is characterized in that described
The main control chip of data processing module (4) be LPC1114, auxiliary circuit include reset circuit, power module, oscillating circuit and
JTAG debugging circuit module.
4. the tumble real-time detecting system according to claim 1 based on acceleration transducer, which is characterized in that described
Power supply of the power module (5) using chargeable 5V lithium battery as whole system.
5. the tumble real-time detecting system according to claim 1 based on acceleration transducer, which is characterized in that described
Wireless transmitter (2) and wireless receiver (3) are carried out data transmission using the modulation demodulation system of GFSK.
6. the tumble real-time detecting system according to claim 1 based on acceleration transducer, which is characterized in that described
Wireless receiver (3) uses USB interface wireless data transfer module.
7. a kind of tumble real-time detection method based on acceleration transducer, which comprises the following steps:
Step 1, by the tumble real-time detecting system based on acceleration transducer, it is worn on the waist i.e. position of hipbone upper end of human body
It sets, acceleration transducer (1) acquires the acceleration of tri- axis of X, Y, Z during human motion, and data are transferred to number by SPI
According to processing module (4)；
Step 2, the data pre-processing unit in data processing module (4) pre-processes data, then substitutes into tumble model,
Judge whether to be in during tumble, starts warning module (6) if falling, system is answered by reseting module (7)
Position；
Step 3, during entire real-time detection, treated data and final testing result are by wireless transmitter
(2) emit and received by wireless receiver (3), be finally transmitted to PC machine (8) and carry out data monitoring.
8. the tumble real-time detection method according to claim 7 based on acceleration transducer, which is characterized in that step 2
Described in data pre-processing unit in data processing module (4) data are pre-processed, then substitute into tumble model, judgement
Whether it is in during tumble, specific as follows:
It is firstly introduced into a total acceleration A0As characteristic quantity:
Wherein ax、ay、azThe respectively acceleration of tri- axis of X, Y, Z of acceleration transducer acquisition；
Secondly, by continuously acquire 5~10 total acceleration A0It averages to obtain V0；
Finally, establishing tumble model according to following three criterions:
1) resultant acceleration threshold criterion
Threshold value A is set1=2g, as resultant acceleration A0Greater than A1When enter judgement 2), otherwise continue to acquire data, g adds for gravity
Speed；
2) slope criterion
Select S1Threshold value of=the 25g/s as slope, when detect resultant acceleration slope data S be greater than S1When, into sentencing 3)
It is disconnected, on the contrary continue to acquire data；
3) average acceleration criterion
Select V3=3.9g is as average value judgment basis, as the data V by average value processing0Greater than V3When, then finally it is judged as
It is falling.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810027142.5A CN110025318A (en) | 2018-01-11 | 2018-01-11 | A kind of tumble real-time detecting system and method based on acceleration transducer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810027142.5A CN110025318A (en) | 2018-01-11 | 2018-01-11 | A kind of tumble real-time detecting system and method based on acceleration transducer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110025318A true CN110025318A (en) | 2019-07-19 |
Family
ID=67234739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810027142.5A Pending CN110025318A (en) | 2018-01-11 | 2018-01-11 | A kind of tumble real-time detecting system and method based on acceleration transducer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110025318A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113792360A (en) * | 2021-09-14 | 2021-12-14 | 陕西燕翼贻谋工业智能科技有限责任公司 | Industrial pipeline twin health monitoring method and system based on simulation technology |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103337132A (en) * | 2013-06-08 | 2013-10-02 | 山东师范大学 | Tumble detection method for human body based on three-axis acceleration sensor |
| CN103377542A (en) * | 2013-07-16 | 2013-10-30 | 中国科学院深圳先进技术研究院 | Human body fall-down preventing early warning method and human body fall-down preventing early warning device |
| CN105654666A (en) * | 2016-02-25 | 2016-06-08 | 福州大学 | Old people falling monitoring method based on MEMS sensor |
-
2018
- 2018-01-11 CN CN201810027142.5A patent/CN110025318A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103337132A (en) * | 2013-06-08 | 2013-10-02 | 山东师范大学 | Tumble detection method for human body based on three-axis acceleration sensor |
| CN103377542A (en) * | 2013-07-16 | 2013-10-30 | 中国科学院深圳先进技术研究院 | Human body fall-down preventing early warning method and human body fall-down preventing early warning device |
| CN105654666A (en) * | 2016-02-25 | 2016-06-08 | 福州大学 | Old people falling monitoring method based on MEMS sensor |
Non-Patent Citations (3)
| Title |
|---|
| 孙子文等: "基于加速度传感器的人体跌倒检测方法", 《计算机工程与科学》 * |
| 孙晓雯等: "基于阈值与PSO-SVM的人体跌倒检测研究", 《计算机工程》 * |
| 林木泉: "老人跌倒报警系统设计", 《智能计算机与应用》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113792360A (en) * | 2021-09-14 | 2021-12-14 | 陕西燕翼贻谋工业智能科技有限责任公司 | Industrial pipeline twin health monitoring method and system based on simulation technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104135911B (en) | Activity classification in multi-axial cord movement monitoring device | |
| CN105286874A (en) | System and method for predicting tumble risk of old people | |
| CN106178481B (en) | A kind of analytical equipment of billiard training, system and analysis method | |
| CN103377542A (en) | Human body fall-down preventing early warning method and human body fall-down preventing early warning device | |
| CN208677399U (en) | Intelligent switch joint angle measurement unit and system | |
| CN103616034A (en) | Network pedometer based on Bluetooth and step calculation method | |
| CN103543834A (en) | Gesture recognition device and method | |
| CN105243787B (en) | A kind of Falls in Old People warning system, alarm method and alarm | |
| CN104173057B (en) | Based on the wearable fall detection system of mobile communication | |
| CN108646316A (en) | A kind of wearing state recognition methods of wearable device and wearable device | |
| Popa | Hand gesture recognition based on accelerometer sensors | |
| CN103390327A (en) | Early warning method and device for preventing human bodies from falling | |
| CN204562127U (en) | A kind of Physical Examination System | |
| CN108186024A (en) | A kind of movement gait monitor and multi-parameter sensor data processing method | |
| CN103903391A (en) | System and method for tumble detection based on RFID | |
| CN106960542A (en) | A kind of low-power consumption falling detection device based on interruption driving | |
| CN110025318A (en) | A kind of tumble real-time detecting system and method based on acceleration transducer | |
| CN103542843A (en) | Apparatus and system for measuring swinging velocity of racket | |
| CN204808545U (en) | Human body detection device that tumbles based on dual sensor | |
| CN104680718A (en) | Intelligent bracelet | |
| CN102440782A (en) | Arm rehabilitation parameter extraction device based on accelerator sensor | |
| CN204192584U (en) | The intelligent individual soldier's motion monitoring of People's Armed Police and vital sign early warning salvage system | |
| CN202710070U (en) | Device and system for measuring racket swinging speed | |
| CN207380669U (en) | Electronic equipment automatically wakes up system and electronic equipment | |
| CN206147905U (en) | System based on multi -sensor fusion fall detection |
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 |
Application publication date: 20190719 |
|
| RJ01 | Rejection of invention patent application after publication |