CN104457781B - A kind of adaptive step number detection method based on single-axis accelerometer - Google Patents
A kind of adaptive step number detection method based on single-axis accelerometer Download PDFInfo
- Publication number
- CN104457781B CN104457781B CN201410806143.1A CN201410806143A CN104457781B CN 104457781 B CN104457781 B CN 104457781B CN 201410806143 A CN201410806143 A CN 201410806143A CN 104457781 B CN104457781 B CN 104457781B
- Authority
- CN
- China
- Prior art keywords
- step number
- window
- pedestrian
- accelerometer
- peak
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C22/00—Measuring distance traversed on the ground by vehicles, persons, animals or other moving solid bodies, e.g. using odometers, using pedometers
- G01C22/006—Pedometers
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Measurement Of Distances Traversed On The Ground (AREA)
Abstract
A kind of adaptive step number detection method based on single-axis accelerometer, using single acceleration transducer and it is fixed on rear side of pedestrian's waist, with this accelerometer measures pedestrian day to acceleration, according to gravity center of human body in the periodic undulations variation characteristic of vertical direction, detection to human body walking step number can be achieved by the crest and trough number that identify this accelerometer output signal.Cause step number verification and measurement ratio to decline problem effectively to solve the factors such as vibration, propose a kind of Dual-window step number detection method, and propose that length of window adaptively adjusts criterion according to human motion state, realize accurate detection of the pedestrian to step number under the different speeds of travel.The present invention is easy and effective, easily realizes, and has certain robustness to interference such as resistance vibrations.
Description
Technical field
The present invention relates to a kind of adaptive step number detection method based on single-axis accelerometer, be effective against vibration etc. because
Disturbed caused by plain, realize the accurate counting to human body traveling step number, belong to pedestrian navigation field, available for improving pedestrian navigation
Middle pedestrian's positioning precision.
Background technology
Pedestrian navigation based on MEMS (MEMS) inertial sensor be in recent years in field of inertia technology one it is new
Emerging technology, generally indoors, mountains and rivers, valley, under the gps signal invalid situation such as tunnel, applied to fire-fighting, the disaster relief, blind person's guiding,
The occasion such as personal scheduling in building.In having radio field conjunction, employing wireless electricity positions comparative maturity to pedestrian,
And it is widely applied.And when radio signal can not be obtained, usually rely on motion sensor and combine inertial technology, people
Body dynamics realize pedestrian's autonomous positioning, and MEMS inertial sensor is conventional motion sensor.Due to microcomputer
Be present larger device error and the complexity of pedestrian's walking motion in electric system inertial sensor so that be used to based on MEMS
Pedestrian's autonomous positioning of property sensor has certain difficulty.Enter every trade due to MEMS inertial sensor is placed on into loins
People's positioning has the characteristics of easy to use so that waist formula pedestrian positioning turns into study hotspot.
Step number problem how is accurately detected for waist formula pedestrian positioning, existing most of research is to utilize weight during pedestrian's walking
Fluctuations feature is descended in the heart to detect step number, i.e., step number is detected to accelerometer output signal by day.But walked
Pin lands in journey and other disturbances can produce various vibrations, so as to cause day to introduce various interference signals to accelerometer so that
The accuracy rate for detecting step number declines.In being positioned for waist formula pedestrian, how to resist the interference vibrated during pedestrian's walking and walked with improving
Number verification and measurement ratio, this is a difficult point and and urgent problem in pedestrian navigation field.
The content of the invention
The purpose of the present invention：Step number verification and measurement ratio decline in pedestrian navigation is caused to be asked effectively to solve the disturbing factors such as vibration
Topic, proposes a kind of Dual-window step number detection method under study movement state on gravity center of human body's fluctuations regular basis, and according to
Human motion state proposes that length of window adaptively adjusts criterion, realizes accurate inspection of the pedestrian to step number under the different speeds of travel
Survey.
The present invention technical solution be：By analyzing gravity center of human body's fluctuations rule under motion state, finding out can
Characterize the characteristic quantity of walking step number.The disturbance factor being subject to during this characteristic quantity moves in practice is further analyzed, is proposed a kind of
Adaptive step number detection method based on single-axis accelerometer, the method are characterised by：Introduce two be used for extract pedestrian day to
The one-dimension array (abbreviation window) of acceleration peak value, length of window L is exactly array length, that is, the sampling number included,
Two windows head and the tail only overlapping sampling number；By Dual-window come identify pedestrian walk when crest number from day to acceleration,
And length of window is adaptively adjusted by human motion state, interference signal caused by vibration during walking is effective against, is realized
The accurate detection of pedestrian's step number.
Using the adaptive step number detection method based on single-axis accelerometer, comprise the following steps：
(1) single-axis accelerometer is fixed on rear side of human lumbar and for measuring pedestrian day to acceleration；
(2) gait cycle T (0), accelerometer signal maximum crest number N1=0, accelerometer signal are set when initializing
Minimum trough number N2=0, traveling step number N=0；
(3) the length L (N) of two windows is adaptively adjusted using gait cycle T (N) and system sampling frequency fs, wherein two
Window only overlapping 1 sampled point；
(4) two windows ask for the peak-peak of contained accelerometer output signal in each window respectively, judge two windows most
Whether big peak value is corresponding same number under same sequential, is, then N1=N1+1 that otherwise N1 is constant, and it is maximum to record appearance
The sequential j of peak value, and it is stored in maxpeak_tim (e N) 1；
(5) two windows ask for the minimum peak of contained accelerometer output signal in each window respectively, judge two windows most
Whether small leak is corresponding same number under same sequential, is, then N2=N2+1 that otherwise N2 is constant, and it is minimum to record appearance
The sequential j of peak value, and it is stored in minpeak_tim (e N) 2；
(6) as N1=N2 and N1>When 0, step number N=N+1, and gait cycle T (N) is calculated, otherwise N and T (N) are constant, from
And realize adaptive step number detection.
The principle of the present invention：Wen Zhonghang, which makes a move, refers to that single pin steps forward a step, and day when az represents pedestrian's walking is to adding
Speed.In the case where not considering the disturbed conditions such as vibration, pedestrian's walking step state is shown in accompanying drawing 1 with corresponding day to acceleration az corresponding relations.
When across the width maximum of foot of human body two, now for gravity center of human body in extreme lower position, az reaches minimum valley value；People is further continued for
When taking a step forward, position of human center starts up to move, and az is increased monotonically in the time；When human body is in erectility, this
When both legs be nearly at vertical state, center of gravity reaches extreme higher position,azReach maximum crest value；Then gravity center of human body is with step
Cut down from extreme higher position and reach extreme lower position, complete the action that row makes a move forward, form a period of change.
One skilled in the art's center of gravity is made a move at extreme higher position in vertical direction fluctuations rule, row by center of gravity, az
Always accordingly there is a crest in signal.Do not consider that pedestrian's row makes a move during the interference such as vibration, once maximum ripple accordingly occurs in az
Peak.Therefore there is crest times N um_crest in az and step number Num_step meets relationship below：
Num_crest=Num_step (1)
If do not vibrated etc. when influenceing, pedestrian's row is calculated by detecting az crest occurrence number Num_crest cans
The step number walked.But various interference signals are constantly present in actual walking process, wherein vibration when landing is to accelerometer
Signal output influences maximum, and these interference signals make it that false crest occurs in az in often walking, and these false crests to utilize az
Signal detection step number accuracy rate declines, and how to be effective against vibration interference and improves step number verification and measurement ratio, is asked for the vibrational perturbation
Topic, propose a kind of adaptive Dual-window step number detection method.
Two adaptive overlaid windows are quoted to detect the peak value that day occurs to acceleration az in each step, including maximum ripple
Peak value and minimum valley value.Detect general principle：Window refers to the one-dimension array for extracting az peak values；Length of window L is exactly
Array length, that is, the sampling number included；Two windows head and the tail only overlapping sampling number, if two windows are most
Big crest value is the same number under same sequential, then it is assumed that the unique maximum of two windows is the maximum crest of az in this step
Value, it can similarly obtain az minimum valley value.
But length of window L is long to make detected step number on the low side, similarly window is too short can make step number bigger than normal, and it is accurate not reach
Calculate the effect of step number.The step number and az peak-peak walked in order to self-adapting detecting different people with friction speed and minimum
Peak value.Experimental data is further analyzed, it is found that length of window is relevant with pedestrian walking speed, can be according to pedestrian walking speed not
Equally adaptively adjust length of window.For ease of simplifying, walking is characterized with gait cycle T (time that row makes a move) herein
Speed.It can show that length of window L and gait cycle T relational expression are as follows through analysis：
L=T*fs (2)
Wherein fs is system sampling frequency, is all relatively applicable when formula (2) is walked to different people with friction speed, passes through walking
The difference of speed carrys out the adaptive length L for adjusting two overlaid windows.By many experiments, different pedestrians are gathered with friction speed row
The true number walked, it is found that it is per second pedestrian in the cadence (step number in the unit interval) for just starting the starting stage is usually no more than 5 steps,
To test result of the different pedestrians under different starting speeds it is further concluded that following rule-of-thumb relation：
L=T (0) * fs (3)
Because of pedestrian, in ground zero, walking speed is small, and cadence further becomes on this basis afterwards within 5 steps are per second
Change, so formula (3) can be used to be grown again with formula (1) come the adaptive window that adjusts in walking process to setting home window length
Degree.
In often being walked using the method detection az maximums crest value and minimum valley value and the two corresponding sequential j (appearance
Time point).The sequential occurred by maximum crest value and minimum valley value can obtain pedestrian gait cycle T, and then use gait
Cycle T adjusts length of window L, realizes based on the adaptive step number detection of single-axis accelerometer.
The present invention compared with prior art the advantages of be：
(1) quote Dual-window and carry out identification feature signal, be effective against the interference such as vibration, to reach accurate detection step number.
(2) the adaptively changing length of window according to the difference of pedestrian's speed of travel, and then self-adapting detecting step number.
Brief description of the drawings
Fig. 1 is gait and accelerometer output signal az relation schematic diagrams；
Fig. 2 is the present invention based on the adaptive step number overhaul flow chart of single-axis accelerometer.
Embodiment：
Fig. 2 is the adaptive step number overhaul flow chart of the present invention, divides following steps：
(1) single-axis accelerometer is fixed on rear side of human lumbar and for measuring pedestrian day to acceleration；
(2) gait cycle T (0), accelerometer signal maximum crest number N1=0, accelerometer signal are set when initializing
Minimum trough number N2=0, traveling step number N=0；
(3) the length L (N) of two windows is adaptively adjusted using gait cycle T (N) and system sampling frequency fs, wherein two
Window only overlapping 1 sampled point；
(4) two windows ask for the peak-peak of contained accelerometer output signal in each window respectively, judge two windows most
Whether big peak value is corresponding same number under same sequential, is, then N1=N1+1 that otherwise N1 is constant, and it is maximum to record appearance
The sequential j of peak value, and it is stored in maxpeak_tim (e N) 1；
(5) two windows ask for the minimum peak of contained accelerometer output signal in each window respectively, judge two windows most
Whether small leak is corresponding same number under same sequential, is, then N2=N2+1 that otherwise N2 is constant, and it is minimum to record appearance
The sequential j of peak value, and it is stored in m inpeak_tim (e N) 2；
(6) as N1=N2 and N1>When 0, step number N=N+1, and gait cycle T (N) is calculated, otherwise N and T (N) are constant, from
And realize adaptive step number detection.
The content not being described in detail in this specification belongs to prior art known to professional and technical personnel in the field.
Claims (1)
1. a kind of adaptive step number detection method based on single-axis accelerometer, it is characterised in that comprise the following steps：
(1) single-axis accelerometer is fixed on rear side of human lumbar and for measuring pedestrian day to acceleration；
(2) gait cycle is set when initializing：The time T (0)=0.2 second that row makes a move, accelerometer signal maximum crest number
N1=0, accelerometer signal minimum trough number N2=0, traveling step number N=0；
(3) the length L (N) of two windows is adaptively adjusted using gait cycle T (N) and system sampling frequency fs, wherein two windows
Only overlapping 1 sampled point；
(4) two windows ask for the peak-peak of contained accelerometer output signal in each window respectively, judge two window maximum peaks
Whether value is corresponding same number under same sequential, is, then N1=N1+1 that no, then N1 is constant, and records and peak-peak occur
Sequential j, and be stored in maxpeak_time (N1)；
(5) two windows ask for the minimum peak of contained accelerometer output signal in each window respectively, judge two window smallest peaks
Whether value is corresponding same number under same sequential, is, then N2=N2+1 that otherwise N2 is constant, and records and minimum peak occur
Sequential j, and be stored in minpeak_time (N2)；
(6) as N1=N2 and N1>When 0, step number N=N+1, and gait cycle T (N) is calculated, otherwise N and T (N) are constant, so as to real
Now adaptive step number detection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410806143.1A CN104457781B (en) | 2014-12-22 | 2014-12-22 | A kind of adaptive step number detection method based on single-axis accelerometer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410806143.1A CN104457781B (en) | 2014-12-22 | 2014-12-22 | A kind of adaptive step number detection method based on single-axis accelerometer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104457781A CN104457781A (en) | 2015-03-25 |
| CN104457781B true CN104457781B (en) | 2018-01-30 |
Family
ID=52904214
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410806143.1A Active CN104457781B (en) | 2014-12-22 | 2014-12-22 | A kind of adaptive step number detection method based on single-axis accelerometer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104457781B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105004349B (en) * | 2015-06-30 | 2018-08-31 | 深圳还是威健康科技有限公司 | A kind of step-recording method |
| CN106767807B (en) * | 2015-11-20 | 2020-07-03 | 北京航空航天大学 | Pedestrian step length comprehensive measurement method based on height and motion characteristics |
| CN105674989B (en) * | 2016-01-27 | 2018-07-24 | 西北大学 | A kind of indoor objects movement locus method of estimation based on mobile phone built-in sensors |
| CN106767888A (en) * | 2016-11-15 | 2017-05-31 | 皖西学院 | A kind of meter based on Wave crest and wave trough detection walks algorithm |
| CN106888006B (en) * | 2017-03-17 | 2020-11-03 | 华自科技股份有限公司 | Signal peak value detection device |
| CN108362282B (en) * | 2018-01-29 | 2020-10-27 | 哈尔滨工程大学 | Inertial pedestrian positioning method based on self-adaptive zero-speed interval adjustment |
| CN115280101A (en) * | 2020-05-29 | 2022-11-01 | 北京嘀嘀无限科技发展有限公司 | Indoor navigation |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1770368A1 (en) * | 2005-10-03 | 2007-04-04 | STMicroelectronics S.r.l. | Pedometer device and step detection method using an algorithm for self-adaptative computation of acceleration thresholds |
| CN102419180A (en) * | 2011-09-02 | 2012-04-18 | 无锡智感星际科技有限公司 | Indoor positioning method based on inertial navigation system and WIFI (wireless fidelity) |
| CN102954803A (en) * | 2012-08-09 | 2013-03-06 | 益体康(北京)科技有限公司 | Adaptive step-counting processing system and method |
| CN103727959A (en) * | 2013-12-31 | 2014-04-16 | 歌尔声学股份有限公司 | Step counting method and step counting device |
| WO2014191803A1 (en) * | 2013-05-27 | 2014-12-04 | Tata Consultancy Services Limited | Acceleration-based step activity detection and classification on mobile devices |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140074431A1 (en) * | 2012-09-10 | 2014-03-13 | Apple Inc. | Wrist Pedometer Step Detection |
-
2014
- 2014-12-22 CN CN201410806143.1A patent/CN104457781B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1770368A1 (en) * | 2005-10-03 | 2007-04-04 | STMicroelectronics S.r.l. | Pedometer device and step detection method using an algorithm for self-adaptative computation of acceleration thresholds |
| CN102419180A (en) * | 2011-09-02 | 2012-04-18 | 无锡智感星际科技有限公司 | Indoor positioning method based on inertial navigation system and WIFI (wireless fidelity) |
| CN102954803A (en) * | 2012-08-09 | 2013-03-06 | 益体康(北京)科技有限公司 | Adaptive step-counting processing system and method |
| WO2014191803A1 (en) * | 2013-05-27 | 2014-12-04 | Tata Consultancy Services Limited | Acceleration-based step activity detection and classification on mobile devices |
| CN103727959A (en) * | 2013-12-31 | 2014-04-16 | 歌尔声学股份有限公司 | Step counting method and step counting device |
Non-Patent Citations (1)
| Title |
|---|
| Modeling Human Walking for Step Detection and Stride Determination by 3-Axis;Kashyap Tumkur，Suneeth Subbiah;《2012 Fourth International Conference on Computational Intelligence, Modelling and Simulation》;20121025;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104457781A (en) | 2015-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104457781B (en) | A kind of adaptive step number detection method based on single-axis accelerometer | |
| CN101354265B (en) | Method and device for counting steps | |
| EP3099229B1 (en) | Improvements in the detection of walking in measurements of the movement of a user | |
| EP2748093B1 (en) | Method and apparatus for elevator motion detection | |
| US8756032B2 (en) | Method and apparatus for fall detection | |
| CN102353383B (en) | Method for step counting and mileage reckoning based on single-axis gyroscope | |
| CN106705968A (en) | Indoor inertial navigation algorithm based on posture recognition and step length model | |
| US20150185044A1 (en) | Step counting method and a pedometer based on a 3-axis accelerometer | |
| CN104197952A (en) | User walk step counting method and device and mobile terminal | |
| CN104406603B (en) | A kind of step-recording method and device based on acceleration transducer | |
| CN106813676B (en) | A kind of step counting, localization method and device | |
| CN101881625A (en) | Stride correction method, ranging method and step-counting device | |
| WO2017000563A1 (en) | Real-time location method and system for intelligent device, and determination method for movement posture of mobile phone | |
| US20210093917A1 (en) | Detecting outdoor walking workouts on a wearable device | |
| CN104613965B (en) | A kind of step-by-step movement pedestrian navigation method based on bidirectional filtering smoothing technique | |
| CN105758404A (en) | Real-time positioning method and system of intelligent equipment | |
| WO2019033587A1 (en) | Smart watch-based gps drifting filtering method, and smart watch | |
| CN109938740A (en) | A kind of gait cycle detecting method, device and computer readable storage medium | |
| TWI439947B (en) | Method for pedestrian behavior recognition and the system thereof | |
| CN104605859B (en) | Indoor navigation gait detection method based on mobile terminal sensor | |
| CN105953794B (en) | A kind of meter step air navigation aid based on MEMS sensor | |
| CN105357753B (en) | A kind of indoor orientation method based on multimodality fusion recursive iteration | |
| CN106643785B (en) | A kind of adaptive step number detection method of multi-source information based on MEMS Inertial Measurement Units | |
| US20150032033A1 (en) | Apparatus and method for identifying movement in a patient | |
| CN106767807B (en) | Pedestrian step length comprehensive measurement method based on height and motion characteristics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200407 Address after: 215123 Room 403, Wenchang College, National University Science and Technology Park, Southeast University, No. 399 Linquan Street, Suzhou Industrial Park, Jiangsu Province Patentee after: Suzhou Miaohang Technology Co.,Ltd. Address before: 100191 Haidian District, Xueyuan Road, No. 37, Patentee before: BEIHANG University |
|
| TR01 | Transfer of patent right |