CN109993119A - A kind of data collection and learning method based on wearable device - Google Patents
A kind of data collection and learning method based on wearable device Download PDFInfo
- Publication number
- CN109993119A CN109993119A CN201910254284.XA CN201910254284A CN109993119A CN 109993119 A CN109993119 A CN 109993119A CN 201910254284 A CN201910254284 A CN 201910254284A CN 109993119 A CN109993119 A CN 109993119A
- Authority
- CN
- China
- Prior art keywords
- data
- sequence
- wearable device
- data sequence
- collection
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
- G06V20/597—Recognising the driver's state or behaviour, e.g. attention or drowsiness
Abstract
A kind of data collection and learning method based on wearable device, first is that can using personal data carry out model conversion be sent to a part of server as big data, can be carried out for other termination mouth using;Second is that the gap of the traveling behavior (such as the mode of hand-held steering wheel, the mode of steering wheel rotation) due to everyone, causes many data uploading onto the server without reference value, but these data can be used as the data reference data of personal traveling next time;Third is that it is creative using F as the foundation calculated in the application, in a manner of abandoning using time point as calculation basis, better reflect the accuracy of above-mentioned reference data sequence and the directive significance to personal user;Fourth is that the mode that the present invention creatively proposes standard deviation and normal distribution combines, so that after obtaining personal data, the result of study and analysis is entirely capable of extremely accurate reflecting personal general behavior, it entirely eliminated special situation, so that carrying out driving behavior timing using reference data sequence when personal, can prevent abnormal condition completely does not remind situation.
Description
Technical field
The present invention relates to a kind of methods of data capture, collect in particular to a kind of driving data of wearable device
Method.
Background technique
Driving fatigue early warning system mainly passes through the perception of camera.It is substantially to capture and divide in the process of moving
Analyse the biobehavioral information of driver, such as the technology of eyes, face, heart, electrical activity of brain etc..However heart activity and brain
Pyroelectric monitor is due to being limited by contacting, currently without batch application in the car.Currently at most adopted fatigue detecting means are
Driver drives behavioural analysis, i.e., by recording and parsing driver turn steering wheel, the behavioural characteristics such as touch on the brake, differentiates and drive
Whether member is tired.But this mode is influenced greatly by driver's driving habit.Another other detection method of major class is: passing through figure
As analysis means carry out Fatigue Assessment to driver face and eye feature.This method is just received and is adopted by whole-car firm gradually
With.But it is also only limitted to fatigue driving, dangerous play is driven, drive when intoxicated etc. to judge very well.And this is general
Be belong to vehicle choose to install configuration, be unfavorable for transplanting between vehicle and universal use.
Also, user, often due to not paying close attention to traffic information in time, causes to pass by and gather around in daily way on and off duty
Stifled section, it is more likely that cause working late.
Summary of the invention
In view of this, the present invention provides a kind of data collection and learning method based on wearable device, by wearable
Equipment is collected and analyzes to various data, the body data of user of the driving behavior of user, forms personal data,
The personal data are personality data, become a reference data in big data by wearable device, can be for more
Other users refer to, and whether the relatively accurate driving behavior for judging user has the tendency of dangerous driving.
The specific technical solution of the present invention is as follows.
A kind of data collection and learning method based on wearable device, the wearable device can carry out drive parameter
It collects, which comprises
Step 1 is collected the data of first day commuter time section in user's a cycle:
Wearable device is acquired acceleration, formed acceleration information sequence A1 in the period=[a11,
A21 ... ..., ai1 ... ..., an1];(acceleration that serial response user holds steering wheel), i1 identifies the 1st day the i-th time
The data of point acquisition.
Wearable device is acquired amplitude, formed amplitude data sequence B 1 in the period=[b11,
B21 ... ..., bi1 ... ..., bn1];(movement range that serial response user holds steering wheel)
Wearable device is acquired heart rate data, formed heart rate data sequence C 1 in the period=[c11,
C21 ... ..., ci1 ... ..., cn1];(emotional stability of serial response user when driving)
Wearable device GPS data is acquired, formed vehicle driving acceleration information sequence D 1 in the period=
[d11, d21 ... ..., di1 ... ..., dn1], vehicle speed data sequence E1=[e11, e21 ... ..., ei1 ... ..., en1], vehicle
Position (coordinate) data sequence F1=[f11, f21 ... ..., fi1 ... ..., fn1];(the number of these serial response vehicle drivings
According to)
The data of above-mentioned collection, which are formed, collects data sequence W1=[A1, B1, C1, D1, E1, F1];
Step 2 is collected the data of second day commuter time section in user's a cycle:
Wearable device is acquired acceleration, formed acceleration information sequence A2 in the period=[a12,
A22 ... ..., ai2 ... ..., an2];
Wearable device is acquired amplitude, formed amplitude data sequence B 2 in the period=[b12, b22 ... ...,
Bi2 ... ..., bn2];
Wearable device is acquired heart rate data, formed heart rate data sequence C 2 in the period=[c12,
C22 ... ..., ci2 ... ..., cn2];
Wearable device GPS data is acquired, formed vehicle driving acceleration information sequence D 2 in the period=
[d12, d22 ... ..., di2 ... ..., dn2], vehicle speed data sequence E2=[e12, e22 ... ..., ei2 ... ..., en2], vehicle
Position (coordinate) data sequence F2=[f12, f22 ... ..., fi2 ... ..., fn2];
The data of above-mentioned collection form collection data sequence W2=[A2, B2, C2, D2, E2, the F2];
Step 3, F value is equal or close in arbitrary sequence of data in the collection data sequence obtained to first day and second day
Different time points connect, using F value it is equal or it is close as calculate basis, to except the vehicle position data sequence it
All data that outer remainder data sequence carries out time point corresponding with the F value carry out standard deviation calculating, obtain collecting data
Sequence W12=[A12, B12, C12, D12, E12, F12];
Step 4 is collected the data of the commuter time section in the third day in user's a cycle:
Wearable device is acquired acceleration, formed acceleration information sequence A3 in the period=[a13,
A23 ... ..., ai3 ... ..., an3];
Wearable device is acquired amplitude, formed amplitude data sequence B 3 in the period=[b13, b23 ... ...,
Bi3 ... ..., bn3];
Wearable device is acquired heart rate data, formed heart rate data sequence C 3 in the period=[c13,
C23 ... ..., ci3 ... ..., cn3];
Wearable device GPS data is acquired, formed vehicle driving acceleration information sequence D 3 in the period=
[d13, d23 ... ..., di3 ... ..., dn3], vehicle speed data sequence E3=[e13, e23 ... ..., ei3 ... ..., en3], vehicle
Position (coordinate) data sequence F3=[f13, f23 ... ..., fi3 ... ..., fn3];
The data of above-mentioned collection form collection data sequence W3=[A3, B3, C3, D3, E3, the F3];
Step 5, to the collection data sequence W12 and the data sequence W3 that collects also according to the mode of the step 3
Standard deviation calculating is carried out, obtains collecting data sequence W13=[A13, B13, C13, D13, E13, F13];
Step 6 repeats step 4 and step 5, until obtain collecting data sequence W1m=[A1m, B1m, C1m, D1m, E1m,
F1m], the m be a cycle in last day, i.e., the m days;The collection data sequence W1m is uploaded to server to make
For an one's share of expenses for a joint undertaking of big data;
Step 7 carries out just too distributed arithmetic to the data in one period of each time point of each data sequence,
Data within normal distribution curve 95% are used as just too distributed data WZ=[AZ, BZ, CZ, DZ, EZ, FZ], take the step 6
Obtained collection data sequence W1m and just too the intersection of distributed data WZ as user reference data sequence W=[A, B, C, D,
E,F].Preferably, further, can using the data within normal distribution curve 60% as just too distributed data WZ=[AZ,
BZ,CZ,DZ,EZ,FZ】。
Further, it analyzes the just too distribution curve and is located at the time point of data except described 95%, and be mapped to described
Vehicle position data sequence, when user travels to the position, the wearable device issues user and reminds.
Further, the period is 17 points to 19 points of 7 points to 9 points of every morning and afternoon.Certainly, according to individual
The difference of work hours is different with the time driven, and can be adjusted.
Further, one period is at least one month.
Further, the i was the i-th time point in the period, and n is the final time section of the period;
It is fixed in work hours section and quitting time section, the duration of i to the i+1.Such as, 7 points to 9 points of morning can it is per second or
It is used as within 0.5 second a time point, 5 points to 7 points of evening can also be per second or 0.5 second is used as a time point.Wherein, 1 to n it
Between certainly exist it is discontinuous because the last one period of going to work is to being incoherent between an earliest period of coming off duty.
Further, the wearable device obtains the speed-limiting messages S of the position road by big data Cloud Server,
When the speed at any point is more than the S in vehicle speed data sequence E in the reference data sequence W, then with S replacement
Fall the car speed sequence in the W, forms new reference data sequence W.
Further, when the data exceeded in the reference data sequence W in the data of wearable device measurement, then
The wearable device issues user and reminds.
Further, the wearable device includes buzzer and/or motor, it is described remind be buzzer work and/or
Motor vibrations.
Through the above technical solution, first is that personal data can be carried out to model conversion is sent to server as big number
According to a part, can be carried out for other termination mouth using;Second is that since everyone traveling behavior is (such as the side of hand-held steering wheel
Formula, the mode of steering wheel rotation) gap, cause many data uploading onto the server without reference value, but these
Data can be used as the data reference data of personal traveling next time;Third is that it is creative using F as the foundation calculated in the application,
In a manner of abandoning using time point as calculation basis, better reflects the accuracy of above-mentioned reference data sequence and individual is used
The directive significance at family;Fourth is that the mode that the present invention creatively proposes standard deviation and normal distribution combines, so that obtaining individual
After data, the result of study and analysis is entirely capable of extremely accurate reflecting personal general behavior, and it is special to entirely eliminated
The case where, so that carrying out driving behavior timing using reference data sequence when personal, abnormal condition can be prevented completely
Do not remind situation.
Specific embodiment
It in order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below will be in the embodiment of the present invention
Technical solution be clearly and completely described, it is clear that described embodiments are some of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall within the protection scope of the present invention.
The term used in embodiments of the present invention is only to be not intended to be limiting merely for for the purpose of describing particular embodiments
The present invention.In the embodiment of the present invention and the "an" of singular used in the attached claims, " described " and "the"
It is also intended to including most forms, unless the context clearly indicates other meaning, " a variety of " generally comprise at least two, but not
It excludes to include at least one situation.
It should be appreciated that term "and/or" used herein is only a kind of incidence relation for describing affiliated partner, indicate
There may be three kinds of relationships, for example, A and/or B, can indicate: individualism A, exist simultaneously A and B, individualism B these three
Situation.In addition, character "/" herein, typicallys represent the relationship that forward-backward correlation object is a kind of "or".
It will be appreciated that though may be described in embodiments of the present invention using term first, second, third, etc..,
But these ... it should not necessarily be limited by these terms.These terms be only used to by ... distinguish.For example, not departing from implementation of the present invention
In the case where example range, first ... can also be referred to as second ..., and similarly, second ... can also be referred to as the
One ....
Depending on context, word as used in this " if ", " if " can be construed to " ... when " or
" when ... " or " in response to determination " or " in response to detection ".Similarly, context is depended on, phrase " if it is determined that " or " such as
Fruit detection (condition or event of statement) " can be construed to " when determining " or " in response to determination " or " when detection (statement
Condition or event) when " or " in response to detection (condition or event of statement) ".
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
Include, so that commodity or system including a series of elements not only include those elements, but also including not clear
The other element listed, or further include for this commodity or the intrinsic element of system.In the feelings not limited more
Under condition, the element that is limited by sentence "including a ...", it is not excluded that in the commodity or system for including the element also
There are other identical elements.
In addition, the step timing in following each method embodiments is only a kind of citing, rather than considered critical.
With the increase of vehicle population, working drives to become the vehicles of most of office worker's selection, so also
Cause traffic blocking problem.An objective of the invention is to learn personal driving behavior data for a long time,
To obtain Behavior law, personal driving behavior is instructed and suggested.
Embodiment one
By observation, the hand-held position of steering wheel of each individual, the mode of steering wheel rotation, right crus of diaphragm are put for brake and throttle
The difference of individual can be had by setting habit all.The thing of these othernesses forms big data, can look for the demographic data for beating regularity,
The demographic data can be burning hot for existing market automated driving system provide foundation.
But for the stage that automated driving system is popularized not yet, it is also to rely on the manipulation of people, institute as above at present
It states, everyone driving behavior can have differences, and above-mentioned demographic data has the constraint and guidance of personal driving behavior very much
Personal Behavior law may be deviated, therefore, it is not especially big that the meaning of personal driving directions is carried out using demographic data.
One aspect of the present invention collects above-mentioned demographic data, provides reference to the automated driving system of subsequent automobile, second is that
Personal driving behavior data are learnt, the performance data of individual itself is formed, user is driven using the performance data
The behavior of sailing is instructed.
In one cycle, which is the long-term period, generally no less than one month (22 working days), when needs obtain
When taking more preferably data, it might even be possible to be 1 year.
Firstly, being collected to personal data.User holds the weared on wrist wearable device of steering wheel, preferably hand
Ring or wrist-watch, the wearable device include at least multi-shaft acceleration transducer and/or gyroscope, GPS unit.
For office worker, go to work daily, the next time has focused largely on the following period: 7 points to 9 points of the morning and
17 points to 19 points of afternoon.Certainly, this time can be customized, to ensure that the above-mentioned time can cover on and off duty complete of user
Journey.
Firstly, being collected to personal data, at first day, within the above-mentioned period, wearable device was to bracelet
Acceleration is acquired, and the acceleration information sequence A1 in the period of formation first day=[a11, a21 ... ..., ai1 ... ...,
an1];
Wearable device is acquired amplitude, formed amplitude data sequence B 1 in the period=[b11, b21 ... ...,
Bi1 ... ..., bn1];
Wearable device is acquired heart rate data, formed heart rate data sequence C 1 in the period=[c11,
C21 ... ..., ci1 ... ..., cn1].
Above-mentioned data sequence is the personal driving behavior parameter of user, be could be aware that at some time point, the driving of user
Behavior (such as acceleration and amplitude) and body mood (such as heart rate).
Then, wearable device GPS data is acquired, forms the vehicle driving acceleration information in the period
Sequence D 1=[d11, d21 ... ..., di1 ... ..., dn1], vehicle speed data sequence E1=[e11, e21 ... ..., ei1 ... ...,
En1], vehicle position data sequence F1=[f11, f21 ... ..., fi1 ... ..., fn1].
Some information of these data reaction vehicle.Wherein, vehicle position data sequence is critical data.
Arrangement modeling carried out to above-mentioned data, formed the collection data sequence W1 of same day user=[A1, B1, C1, D1, E1,
F1】。
Then, continue to collect data, above-mentioned first day collection mode is copied, to second day in user's a cycle
The commuter time data of section are collected:
Wearable device is acquired acceleration, formed acceleration information sequence A2 in the period=[a12,
A22 ... ..., ai2 ... ..., an2];
Wearable device is acquired amplitude, formed amplitude data sequence B 2 in the period=[b12, b22 ... ...,
Bi2 ... ..., bn2];
Wearable device is acquired heart rate data, formed heart rate data sequence C 2 in the period=[c12,
C22 ... ..., ci2 ... ..., cn2];
Wearable device GPS data is acquired, formed vehicle driving acceleration information sequence D 2 in the period=
[d12, d22 ... ..., di2 ... ..., dn2], vehicle speed data sequence E2=[e12, e22 ... ..., ei2 ... ..., en2], vehicle
Position (coordinate) data sequence F2=[f12, f22 ... ..., fi2 ... ..., fn2];
The data of above-mentioned collection form collection data sequence W2=[A2, B2, C2, D2, E2, the F2].
Then, the study and calculating for carrying out data, by learning and calculating, available effective data.
Most driving data all can be as a reference point with the time, still, since user waits traffic lights daily, congestion feelings
The difference of condition, the application is using critical data above-mentioned --- and vehicle position data sequence is as the basis calculated.It calculates
Mode is as follows:
Compare the data sequence F in W1 and W2, finds out the equal and similar point of wherein F value.In fact, since GPS module positions
Precision, the F value within the scope of 10 meters of diameter, we are still considered close.
Method particularly includes:
It is 1 any point into N there are FQ(Q in W1) value finds out equal with FQ value to phase in the F data sequence of W2
The time point of all points found above-mentioned in the corresponding time point of FQ in W1, W2 is mapped, is based on by close all points
The correspondence of F sequence, to the FQ of W1 sequence corresponding time point corresponding all sequences in addition to F data sequence, W2 sequence
All time points found, the corresponding all sequences in addition to F sequence carried out standard deviation calculating;
Repeat aforesaid operations, until all calculates all F values of F sequence in W1 and finishes, obtain collection data sequence W12=[A12,
B12,C12,D12,E12,F12】。
Alternatively, specific method are as follows:
It is 1 any point into N there are FP(P in W2) value finds out equal with FQ value to phase in the F data sequence of W1
The time point of all points found above-mentioned in the corresponding time point of FP in W2, W1 is mapped, is based on by close all points
The correspondence of F sequence, to the FP of W2 sequence corresponding time point corresponding all sequences in addition to F data sequence, W1 sequence
All time points found, the corresponding all sequences in addition to F sequence carried out standard deviation calculating;
Repeat aforesaid operations, until all calculates all F values of F sequence in W2 and finishes, obtain collection data sequence W12=[A12,
B12,C12,D12,E12,F12】。
Subsequently, the data of the commuter time section in the third day in a cycle are collected:
Wearable device is acquired acceleration, formed acceleration information sequence A3 in the period=[a13,
A23 ... ..., ai3 ... ..., an3];
Wearable device is acquired amplitude, formed amplitude data sequence B 3 in the period=[b13, b23 ... ...,
Bi3 ... ..., bn3];
Wearable device is acquired heart rate data, formed heart rate data sequence C 3 in the period=[c13,
C23 ... ..., ci3 ... ..., cn3];
Wearable device GPS data is acquired, formed vehicle driving acceleration information sequence D 3 in the period=
[d13, d23 ... ..., di3 ... ..., dn3], vehicle speed data sequence E3=[e13, e23 ... ..., ei3 ... ..., en3], vehicle
Position (coordinate) data sequence F3=[f13, f23 ... ..., fi3 ... ..., fn3];
The data of above-mentioned collection form collection data sequence W3=[A3, B3, C3, D3, E3, the F3].
In above manner, to the collection data sequence W12 and collection data sequence W3 also according to the step
Three mode carries out standard deviation calculating, obtains collecting data sequence W13=[A13, B13, C13, D13, E13, F13];
Aforesaid operations are repeated, until obtaining collecting data sequence W1m=[A1m, B1m, C1m, D1m, E1m, F1m], the m is one
Last day in a period, i.e., the m days;The collection data sequence W1m is uploaded to portion of the server as big data
Son.
In fact, what above-mentioned collection data sequence W1m had the been obtained more perfectly driving behavior data of user,
But there may be largely due to invalid data (having walked different road due to traffic congestion in such as user day) in the data,
Only this, above-mentioned invalid data must be weeded out in data statistics.Generally in statistics, can be had using normal distribution statistical
Effect proposes a large amount of invalid data.But as the data for instructing user to drive, it is necessary to give enough constraints to user and refer to
It leads, range cannot be too wide in range, and therefore, the present invention creatively proposes to carry out data processing such as under type:
Just too distributed arithmetic, normal distribution are carried out to the data in one period of each time point of each data sequence
Data within curve 95% are used as just too distributed data WZ=[AZ, BZ, CZ, DZ, EZ, FZ], the receipts for taking the step 6 to obtain
Collect data sequence W1m and just too reference data sequence W=[A, B, C, D, E, F] of the intersection of distributed data WZ as user.
Intersection using the choice being just distributed very much and with W1m data sequence can very well reject invalid data, and
And it can be by the driving behavior data constraint of user in very reasonable range.
Embodiment two
The period mentioned a kind of for above-described embodiment.We are configured advantageously according to the processing capacity of processor.
The i was the i-th time point in the period, and n is the final time section of the period;In work hours section and quitting time
Section, the fixation when duration of i to the i+1.Such as, 7 points to 9 points of morning can be per second or be used as within 0.5 second a time point, evening
Upper 5 points to 7 points can also per second or 0.5 second one time point of conduct.Wherein, 1 is discontinuous to certainly existing between n, because
The last one period go to work to being incoherent between an earliest period of coming off duty.
The section as locating for user is constantly in the section of more congestion, then when may be set to be U.S. 10 seconds as two
Between point between length.
Embodiment three
According to the specification that traffic travels, most of old driver can be turned on the edge ball of speed limit.Surpass in daily driving process
In the range of 10%, old driver also can less pay close attention to speed.But such case must remind driver.Therefore, this Shen
The speed-limiting messages S that the position road is please obtained by big data Cloud Server, when car speed in the reference data sequence W
The speed at any point is more than the S in data sequence E, then replaces the car speed sequence in the W with the S, is formed
New reference data sequence W.And when the number exceeded in the reference data sequence W in the data of wearable device measurement
According to then the wearable device issues user and reminds.
The wearable device includes buzzer and/or motor, and described remind is buzzer work and/or motor vibrations.
Through the above technical solution, first is that personal data can be carried out to model conversion is sent to server as big number
According to a part, can be carried out for other termination mouth using;Second is that since everyone traveling behavior is (such as the side of hand-held steering wheel
Formula, the mode of steering wheel rotation) gap, cause many data uploading onto the server without reference value, but these
Data can be used as the data reference data of personal traveling next time;Third is that it is creative using F as the foundation calculated in the application,
In a manner of abandoning using time point as calculation basis, better reflects the accuracy of above-mentioned reference data sequence and individual is used
The directive significance at family;Fourth is that the mode that the present invention creatively proposes standard deviation and normal distribution combines, so that obtaining individual
After data, the result of study and analysis is entirely capable of extremely accurate reflecting personal general behavior, and it is special to entirely eliminated
The case where, so that carrying out driving behavior timing using reference data sequence when personal, abnormal condition can be prevented completely
Do not remind situation.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features;
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (9)
1. a kind of data collection and learning method based on wearable device, the wearable device can receive drive parameter
Collection, which comprises
Step 1 is collected the data of first day commuter time section in user's a cycle:
Wearable device is acquired acceleration, formed acceleration information sequence A1 in the period=[a11,
A21 ... ..., ai1 ... ..., an1];Wherein, i1 indicates the data of the 1st day the i-th time point acquisition;
Wearable device is acquired amplitude, formed amplitude data sequence B 1 in the period=[b11, b21 ... ...,
Bi1 ... ..., bn1];
Wearable device is acquired heart rate data, formed heart rate data sequence C 1 in the period=[c11,
C21 ... ..., ci1 ... ..., cn1];
Wearable device GPS data is acquired, formed vehicle driving acceleration information sequence D 1 in the period=
[d11, d21 ... ..., di1 ... ..., dn1], vehicle speed data sequence E1=[e11, e21 ... ..., ei1 ... ..., en1], vehicle
Position data sequence F1=[f11, f21 ... ..., fi1 ... ..., fn1];
The data of above-mentioned collection, which are formed, collects data sequence W1=[A1, B1, C1, D1, E1, F1];
Step 2 is collected the data of second day commuter time section in user's a cycle:
Wearable device is acquired acceleration, formed acceleration information sequence A2 in the period=[a12,
A22 ... ..., ai2 ... ..., an2];
Wearable device is acquired amplitude, formed amplitude data sequence B 2 in the period=[b12, b22 ... ...,
Bi2 ... ..., bn2];
Wearable device is acquired heart rate data, formed heart rate data sequence C 2 in the period=[c12,
C22 ... ..., ci2 ... ..., cn2];
Wearable device GPS data is acquired, formed vehicle driving acceleration information sequence D 2 in the period=
[d12, d22 ... ..., di2 ... ..., dn2], vehicle speed data sequence E2=[e12, e22 ... ..., ei2 ... ..., en2], vehicle
Position data sequence F2=[f12, f22 ... ..., fi2 ... ..., fn2];
The data of above-mentioned collection form collection data sequence W2=[A2, B2, C2, D2, E2, the F2];
Step 3, F value is equal or close in arbitrary sequence of data in the collection data sequence obtained to first day and second day
Different time points connect, using F value it is equal or it is close as calculate basis, to except the vehicle position data sequence it
All data that outer remainder data sequence carries out time point corresponding with the F value carry out standard deviation calculating, obtain collecting data
Sequence W12=[A12, B12, C12, D12, E12, F12];
Step 4 is collected the data of the commuter time section in the third day in user's a cycle:
Wearable device is acquired acceleration, formed acceleration information sequence A3 in the period=[a13,
A23 ... ..., ai3 ... ..., an3];
Wearable device is acquired amplitude, formed amplitude data sequence B 3 in the period=[b13, b23 ... ...,
Bi3 ... ..., bn3];
Wearable device is acquired heart rate data, formed heart rate data sequence C 3 in the period=[c13,
C23 ... ..., ci3 ... ..., cn3];
Wearable device GPS data is acquired, formed vehicle driving acceleration information sequence D 3 in the period=
[d13, d23 ... ..., di3 ... ..., dn3], vehicle speed data sequence E3=[e13, e23 ... ..., ei3 ... ..., en3], vehicle
Position (coordinate) data sequence F3=[f13, f23 ... ..., fi3 ... ..., fn3];
The data of above-mentioned collection form collection data sequence W3=[A3, B3, C3, D3, E3, the F3];
Step 5, to the collection data sequence W12 and the data sequence W3 that collects also according to the mode of the step 3
Standard deviation calculating is carried out, obtains collecting data sequence W13=[A13, B13, C13, D13, E13, F13];
Step 6 repeats step 4 and step 5, until obtain collecting data sequence W1m=[A1m, B1m, C1m, D1m, E1m,
F1m], the m be a cycle in last day, i.e., the m days;The collection data sequence W1m is uploaded to server to make
For an one's share of expenses for a joint undertaking of big data;
Step 7 carries out just too distributed arithmetic to the data in one period of each time point of each data sequence,
Data within normal distribution curve 95% are used as just too distributed data WZ=[AZ, BZ, CZ, DZ, EZ, FZ], take the step 6
Obtained collection data sequence W1m and just too the intersection of distributed data WZ as user reference data sequence W=[A, B, C, D,
E,F】。
2. a kind of data collection and learning method based on wearable device according to claim 1, which is characterized in that point
It analyses the just too distribution curve and is located at the time point of data except described 95%, and be mapped to the vehicle position data sequence, when
When user travels to the position, the wearable device issues user and reminds.
3. a kind of data collection and learning method based on wearable device according to claim 1, which is characterized in that institute
Stating the period is 17 points to 19 points of 7 points to 9 points of every morning and afternoon.
4. a kind of data collection and learning method based on wearable device according to claim 1, which is characterized in that institute
A cycle is stated to be at least one month.
5. a kind of data collection and learning method based on wearable device according to claim 1, which is characterized in that institute
Stating the i was the i-th time point in the period, and n is the final time section of the period;The work hours section and it is next when
Between section, it is fixed when the duration of i to the i+1.
6. a kind of data collection and learning method based on wearable device according to claim 1, which is characterized in that institute
The speed-limiting messages S that wearable device obtains the position road by big data Cloud Server is stated, as the reference data sequence W
The speed at any point is more than the S in middle vehicle speed data sequence E, then the car speed in the W is replaced with the S
Sequence forms new reference data sequence W.
7. a kind of data collection and learning method based on wearable device according to claim 1, which is characterized in that when
Beyond the data in the reference data sequence W in the data of wearable device measurement, then the wearable device to
Family, which issues, reminds.
8. a kind of data collection and learning method based on wearable device according to claim 1, which is characterized in that
In the step 3, in W1, there are FQ values, in the F data sequence of W2, find out equal with FQ value and similar all
Point maps the time point of all points found above-mentioned in the corresponding time point of FQ in W1, W2, pair based on F sequence
Answer, to the FQ of the W1 sequence corresponding time point corresponding all sequences in addition to F data sequence, W2 sequence find it is all
Time point, the corresponding all sequences in addition to F sequence carried out standard deviation calculating;
Repeat aforesaid operations, until all calculates all F values of F sequence in W1 and finishes, obtain collection data sequence W12=[A12,
B12,C12,D12,E12,F12】。
9. a kind of data collection and learning method based on wearable device according to claim 1, which is characterized in that
It is 1 any point into N there are FP(P in W2) value finds out equal with FQ value and similar institute in the F data sequence of W1
It is some, the time point of all points found above-mentioned in the corresponding time point of FP in W2, W1 is mapped, F sequence is based on
Correspondence, the FP of W2 sequence corresponding time point corresponding all sequences in addition to F data sequence, W1 sequence are found
All time points, the corresponding all sequences in addition to F sequence carried out standard deviation calculating;
Repeat aforesaid operations, until all calculates all F values of F sequence in W2 and finishes, obtain collection data sequence W12=[A12,
B12,C12,D12,E12,F12】。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910254284.XA CN109993119A (en) | 2019-03-31 | 2019-03-31 | A kind of data collection and learning method based on wearable device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910254284.XA CN109993119A (en) | 2019-03-31 | 2019-03-31 | A kind of data collection and learning method based on wearable device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109993119A true CN109993119A (en) | 2019-07-09 |
Family
ID=67132004
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910254284.XA Pending CN109993119A (en) | 2019-03-31 | 2019-03-31 | A kind of data collection and learning method based on wearable device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109993119A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113392892A (en) * | 2021-06-08 | 2021-09-14 | 重庆大学 | Method and device for identifying driving habits of driver based on data fusion |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104952210A (en) * | 2015-05-15 | 2015-09-30 | 南京邮电大学 | Fatigue driving state detecting system and method based on decision-making level data integration |
US20160039424A1 (en) * | 2014-08-11 | 2016-02-11 | Lg Electronics Inc. | Wearable device and method of operating the same |
CN105996990A (en) * | 2016-04-29 | 2016-10-12 | 清华大学 | Fatigue driving early warning bracelet integrating heart rate and driving action and early warning method |
JP2016222136A (en) * | 2015-06-01 | 2016-12-28 | パイオニア株式会社 | Driver estimation device, control method, program and storage medium |
CN108765876A (en) * | 2018-05-31 | 2018-11-06 | 东北大学 | Driving fatigue depth analysis early warning system based on multimode signal and method |
-
2019
- 2019-03-31 CN CN201910254284.XA patent/CN109993119A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160039424A1 (en) * | 2014-08-11 | 2016-02-11 | Lg Electronics Inc. | Wearable device and method of operating the same |
CN104952210A (en) * | 2015-05-15 | 2015-09-30 | 南京邮电大学 | Fatigue driving state detecting system and method based on decision-making level data integration |
JP2016222136A (en) * | 2015-06-01 | 2016-12-28 | パイオニア株式会社 | Driver estimation device, control method, program and storage medium |
CN105996990A (en) * | 2016-04-29 | 2016-10-12 | 清华大学 | Fatigue driving early warning bracelet integrating heart rate and driving action and early warning method |
CN108765876A (en) * | 2018-05-31 | 2018-11-06 | 东北大学 | Driving fatigue depth analysis early warning system based on multimode signal and method |
Non-Patent Citations (3)
Title |
---|
张驰等: "驾驶疲劳检测技术概述", 《交通工程》 * |
邢彦涛等: "基于多传感器的可穿戴疲劳检测装置设计", 《国外电子测量技术》 * |
邢邗等: "利他情境对驾驶个体跟驰行为的影响", 《清华大学学报(自然科学版)》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113392892A (en) * | 2021-06-08 | 2021-09-14 | 重庆大学 | Method and device for identifying driving habits of driver based on data fusion |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP7032966B2 (en) | Vehicle safety support devices and methods | |
CN108694813B (en) | Method and device for judging fatigue driving based on driving behaviors | |
EP1997705B1 (en) | Drive behavior estimating device, drive supporting device, vehicle evaluating system, driver model making device, and drive behavior judging device | |
TWI447039B (en) | Driving behavior analysis and warning system and method thereof | |
JP4791874B2 (en) | Driving support device and driving action determination device | |
US6313749B1 (en) | Sleepiness detection for vehicle driver or machine operator | |
US8427326B2 (en) | Method and system for detecting the physiological onset of operator fatigue, drowsiness, or performance decrement | |
JP5045492B2 (en) | Awakening level estimation device | |
EP3730375B1 (en) | Method and system for analysing the control of a vehicle | |
KR20100048279A (en) | Method and system of driving safety index computing | |
CN110728824B (en) | Driver fatigue state detection and reminding method based on multi-source data | |
US11173919B2 (en) | Control device, system and method for determining a comfort level of a driver | |
JP2011022738A (en) | Device and method for estimation of driver's condition | |
CN109993119A (en) | A kind of data collection and learning method based on wearable device | |
JP2005092285A (en) | Vehicle driving status estimating device and driver's vehicle driving characteristic estimating device | |
JP2017194872A (en) | Determination program, determination method and information processing device | |
CN108711204B (en) | Driving abnormity detection system and method integrating human-vehicle-road multi-source information | |
CN114852088A (en) | Driver assistance system and method for identifying and warning dangerous and fatigue driving behaviors | |
JP2008301957A (en) | Mental state inference device | |
CN109249931A (en) | Control method and device for vehicle drive | |
CN115909752B (en) | Method for identifying and counting sharp turns based on historical data of vehicle users | |
EP1674375B1 (en) | Method for determining a measure for evaluating the behaviour of a driver of a vehicle | |
Găvrută et al. | Software solution for monitoring and analyzing driver behavior | |
CN109118125B (en) | Urban environment and economy comprehensive evaluation system | |
KR102176065B1 (en) | Method and program for calculation of sleep and driving planning based on sleep pattern and fatigue |
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 | ||
AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20230516 |
|
AD01 | Patent right deemed abandoned |