CN106251583A - Fatigue driving discrimination method based on driving behavior Yu eye movement characteristics - Google Patents
Fatigue driving discrimination method based on driving behavior Yu eye movement characteristics Download PDFInfo
- Publication number
- CN106251583A CN106251583A CN201610869724.9A CN201610869724A CN106251583A CN 106251583 A CN106251583 A CN 106251583A CN 201610869724 A CN201610869724 A CN 201610869724A CN 106251583 A CN106251583 A CN 106251583A
- Authority
- CN
- China
- Prior art keywords
- fatigue
- driving
- standard deviation
- analysis
- parameter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/06—Alarms for ensuring the safety of persons indicating a condition of sleep, e.g. anti-dozing alarms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a kind of fatigue driving discrimination method based on driving behavior Yu eye movement characteristics, comprise the following steps: 1) design fatigue driving simulation experiment, driver's driving behavior data under different fatigue state, and the eye movement data of driver is gathered by experiment；2) experimental data is carried out cutting screening, set up fatigue driving sample database；3) use the method for one factor analysis of variance that the driving behavior parameter of driver under different fatigue state and the significance of eye movement characteristics parameter differences are carried out statistical analysis, and have found the optimal time window of various characteristic parameter；4) characteristic parameter going out Preliminary screening carries out bivariate Spearman correlation analysis；5) optimal characteristics parameter is filtered out；6) fatigue driving BP neural network identification model is set up；7) utilize Matlab software programming model program, randomly select training sample and test sample the set pair analysis model is trained and fatigue state identification.
Description
Technical field
The present invention relates to a kind of fatigue driving discrimination method based on driving behavior Yu eye movement characteristics.
Background technology
Along with the fast development of communication, constantly riseing of vehicle guaranteeding organic quantity, road traffic accident quantity is increasingly
Many, traffic safety problem has become as a serious social problem.Fatigue driving is cause vehicle accident important former
One of because of, in recent years, the correlational study problem about fatigue driving gets more and more, and fatigue-driving detection technology have also been obtained rapidly
Development, has studied and fatigue detecting has had been achieved with good effect, but still in place of Shortcomings:
1) driver is a progressive formation from regaining consciousness to fatigue, and this process is often ignored in present research, to driving
When member enters fatigue state and cannot judge, also cannot give rational information for the practical situation of driver, therefore,
Reasonably the fatigue state of driver is divided and accurate recognition will be an emphasis of research；
2) existing based on single finger object detection method, it is impossible to overcome the environment such as space, illumination, weather to accuracy of detection
Impact, therefore, uses detection method based on information fusion to improve detection accuracy and the important channel of reliability by being.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of fatigue driving identification based on driving behavior Yu eye movement characteristics
Method.
Fatigue driving discrimination method based on driving behavior Yu eye movement characteristics, comprises the following steps:
1) design fatigue driving simulation experiment, gathers driver's driving behavior number under different fatigue state by experiment
According to, including the eye movement data of steering wheel angle, tarnsition velocity, speed, vehicle acceleration and driver, including blinking, watch attentively, sweeping
Depending on data；
2) experimental data is carried out cutting screening, set up fatigue driving sample database；
3) driving behavior parameter and the eye of driver under different fatigue state are moved by the method using one factor analysis of variance
The significance of characteristic parameter difference carries out statistical analysis, and have found the optimal time window of various characteristic parameter；Preliminary screening goes out
Driving behavior parameter includes steering wheel absolute mean SAM, corner standard deviation SASTD, steering wheel angular velocity absolute mean
SWM, angular velocity standard deviation SWSTD, velocity standard difference Vstd, acceleration average Am and standard deviation Astd；Eye movement characteristics parameter bag
Include frequency of wink BF, duration of fixation average FIXT_mean and standard deviation FIXT_std, pan average speed standard deviation
SACV_std and pupil diameter coefficient of variation CVPLD；
4) characteristic parameter going out Preliminary screening carries out bivariate Spearman correlation analysis, specifically comprises the following steps that
4-1) characteristic parameter in source with same index is analyzed, the significant feature of dependency, one can only be selected；
4-2) to the characteristic parameter filtered out, move parametric classification by driving behavior parameter and eye and carry out correlation analysis, with
Sample, index significant to dependency can only select one；
4-3) remaining index after above two screenings being carried out correlation analysis, index significant to dependency is adopted
Take step 4-1), 4-2) same processing mode；
5) filter out optimal characteristics parameter to include: steering wheel angle standard deviation SASTD, vehicle acceleration standard deviation Astd,
Pan average speed standard deviation SACV_std and pupil diameter coefficient of variation CVPLD；
6) fatigue driving BP neural network identification model is set up, it is first determined the input neuron number of this network structure, defeated
Go out neuron number, the implicit number of plies, hidden layer neuron number；The transmission function of next Internet, the power of each layer initial value is set
Value, finally determines the optimal learning efficiency, chooses suitable anticipation error, specific as follows:
6-1) input neuron, selects step 5) in optimal characteristics parameter and the input as model, and input data are entered
Row normalized；
6-2) output neuron, sets and is output as that 1-is clear-headed, 2-tired, 3-is the most tired, and represented by a neuron,
In the predicting the outcome of network, with 1.5 and 2.5 as boundary, if predictive value 1≤k≤1.5, then it is clear-headed；If predictive value 1.5
≤ k≤2.5, if being then tired predictive value 2.5≤k≤3, are then the most tired；
6-3) the implicit number of plies, single hidden layer, node in hidden layer is 9；
6-4) transmission function, the neutral net of BP is internal uses tansig to use purelin as transmission function, output layer
As transmission function；
6-5) initial weight, initial weight is set to the non-zero random value between (-1,1)；
6-6) the learning efficiency and anticipation error, uses LM learning algorithm, and the learning efficiency is automatically adjusted according to anticipation error, nothing
Need to set；Anticipation error is set as 0.001；
7) utilize Matlab software programming model program, randomly select training sample and test sample the set pair analysis model is instructed
Practice and fatigue state identification.
Further, described fatigue driving simulation experiment is specific as follows:
1) experimental facilities:
A, driving analog system, by emulation vehicle, control station district, visual system, computer system, network system and phase
The software system composition answered；Primary operational equipment is all real vehicles equipment, by the input action of sensor capture user, and will
The input of user and other relevant be simulated simulation calculation, simulation result is shown by visual system, sound is imitated simultaneously
Corresponding informance is fed back to user by true system and signal output system；System also includes emulating this car phase in vehicle operation
Close the export interface of data, different sample frequencys can be set according to Research Requirements, derive data and carry out secondary analysis；
B, Eye Link II type eye tracker, uses pupil pattern to be tracked；
2) experiment scene: using dull beltway as experiment scene, described dull beltway is that an annular is double
To the suburb dullness high-grade highway in 6 tracks, lane width 3.75m, road total length 20km, evenness of road surface, weather is set to fine
My god；
3) experimenter: healthy, rule of sleeping, do not allow to drink in 24h before test, do not allow in 12h before test
Coffee for drinking, tea；
4) experimental program: keeping the speed of 100km/s to travel at middle lane, staff every 5min record is the most tested
The Subjective fatigue grade (being demarcated by sleep yardstick table Karolinska sleepiness scale, KSS) of personnel, until travelling
60min, experiment terminates.
The invention has the beneficial effects as follows:
The present invention, in existing achievement in research, is extracted new fatigue driving Testing index, and is verified by variance analysis
The effectiveness of index, sets up fatigue driving identification model based on information fusion technology, overcomes based on single piece of information source tired
The limitation of labor discrimination method.
Detailed description of the invention
The present invention is further elaborated for specific examples below, but not as a limitation of the invention.
1, design fatigue driving simulation experiment, gathers driver's driving behavior number under different fatigue state by experiment
According to, specific as follows:
1) experimental facilities:
A, driving analog system, by emulation vehicle, control station district, visual system, computer system, network system and phase
The software system composition answered；Primary operational equipment is all real vehicles equipment, by the input action of sensor capture user, and will
The input of user and other relevant be simulated simulation calculation, simulation result is shown by visual system, sound is imitated simultaneously
Corresponding informance is fed back to user by true system and signal output system；System also includes emulating this car phase in vehicle operation
Close the export interface of data, different sample frequencys can be set according to Research Requirements, derive data and carry out secondary analysis；
B, Eye Link II type eye tracker, uses pupil pattern to be tracked；
2) experiment scene: using dull beltway as experiment scene, described dull beltway is that an annular is double
To the suburb dullness high-grade highway in 6 tracks, lane width 3.75m, road total length 20km, evenness of road surface, weather is set to fine
My god；
3) experimenter: 9 tested personnel (man: 7, female: 2), age 23-27 year, driving age 2-5, healthy,
Sleep rule, does not allows to drink in 24h before test, does not allow coffee for drinking, tea before test in 12h；
4) experimental program: keeping the speed of 100km/s to travel at middle lane, staff every 5min record is the most tested
The Subjective fatigue grade (being demarcated by sleep yardstick table Karolinska sleepiness scale, KSS) of personnel, until travelling
60min, experiment terminates.
2, experimental data is carried out cutting screening, set up fatigue driving sample database；
Be that an interval carries out cutting by 5min, i.e. 5min data are as a sample, and 9 subjectss participate in reality altogether
Test, totally 108 samples；To driving behavior data, reject containing bend and the sample data of thread-changing operation, for eye movement data,
By Eye Link II type eye tracker itself limit, intelligence to eyeball opposing headers in the range of horizontal vertical direction [-30 °, 30 °]
Eye movement data carry out record, therefore the data beyond [-30 °, 30 °] scope to be rejected.
After screening, remaining 102, valid data sample, according to the evaluation methodology of driving fatigue, by 102 samples
Carry out fatigue state division, set up fatigue driving sample database, as shown in table 1 below.
Table 1 fatigue driving sample data information
Driving behavior parameter and the eye of driver under different fatigue state are moved by the method 3, using one factor analysis of variance
The significance of characteristic parameter difference carries out statistical analysis, and have found the optimal time window of various characteristic parameter；Preliminary screening goes out
Driving behavior parameter includes steering wheel absolute mean SAM, corner standard deviation SASTD, steering wheel angular velocity absolute mean
SWM, angular velocity standard deviation SWSTD, velocity standard difference Vstd, acceleration average Am and standard deviation Astd；Eye movement characteristics parameter bag
Include frequency of wink BF, duration of fixation average FIXT_mean and standard deviation FIXT_std, pan average speed standard deviation
SACV_std and pupil diameter coefficient of variation CVPLD；
1) analysis directly perceived to steering wheel angle, can show that the fatigue characteristic of driver typically lasts for 5-15s, 5s,
Separately design steering wheel angle in tri-time windows of 10s, 15s to be correlated with each fatigue characteristic parameter.Utilize one factor analysis of variance method
Quantitative study driver steering wheel angle absolute value under different fatigue degree and the diversity of standard deviation, at significance level
In the case of α=0.05, the results of analysis of variance is as shown in table 2.
The results of analysis of variance of the diversity of table 2 steering wheel angle absolute value and standard deviation
Steering wheel angle absolute value SAM, when time window is 10s, F=3.893 is maximum, steering wheel angle absolute value SAM
Optimal time window be 10s；Steering wheel angle standard deviation SASTD, when time window is 10s, F=4.801 is maximum, and steering wheel turns
The optimal time window of mean angular deviation SASTD is 10s.
2) randomly select waking state, fatigue state and unusual each 10 sample datas under fatigue state, press respectively 5s,
10s, 15s tri-time window calculated direction dish angular velocity absolute mean SWM, tarnsition velocity standard deviations SWSTD, utilize single factor test side
Difference analysis standard measure research driver's diversity under different fatigue degree, in the case of level of significance α=0.05, side
Difference analysis result is as shown in table 3.
Table 3 steering wheel angular velocity absolute mean and the results of analysis of variance of standard deviation
Steering wheel angular velocity absolute mean SWM, when time window is 5s, F=4.923 is maximum, and steering wheel angular velocity is absolute
The optimal time window of average SWM is 5s；Tarnsition velocity standard deviation SWSTD, when time window is 5s, F=4.907 is maximum, corner
The optimal time window of velocity standard difference SWSTD is 5s.
3) calculate speed average Vm and vehicle speed standard deviation Vstd at tri-time windows of 40s, 80s, 120s, utilize single factor test
Method of analysis of variance quantitative study driver diversity under different fatigue degree, in the case of level of significance α=0.05,
The results of analysis of variance is as shown in table 4.
Table 4 speed average and the results of analysis of variance of standard deviation
There is not the significance difference opposite sex between speed average Vm different fatigue level, therefore speed average is unsuitable for fatigue driving
State-detection；Vehicle speed standard deviation Vstd, when time window is 80s, F=46.424 value is maximum, vehicle speed standard deviation Vstd
Optimal time window is 80s.
4) calculate acceleration absolute value and standard deviation at tri-time windows of 25s, 50s, 75s, utilize one factor analysis of variance
Standard measure research driver's diversity under different fatigue degree, in the case of level of significance α=0.05, variance analysis
Result is as shown in table 5.
Table 5 acceleration absolute value and the results of analysis of variance of standard deviation
For acceleration absolute value Am and acceleration standard deviation Astd, when time window is 25s, the two is at different fatigue shape
Difference between state is the most notable, and therefore, acceleration absolute value Am and acceleration standard deviation Astd optimal time window are 25s.
5) from sample database, randomly select 10 sample datas of fatigue state in 3 carry out one factor analysis of variance,
Time window is 30s and 60s, and frequency of wink is in the case of (α=0.05), and the results of analysis of variance is as shown in table 6 below.
The results of analysis of variance of table 6 frequency of wink
No matter time window is 30s or 60s, and frequency of wink BF all exists significant difference, and access time, window was 30s.
6) time window is 30s and 60s, enters average and the standard deviation of driver's duration of fixation under different fatigue state
Row one factor analysis of variance, result is as shown in table 7 below.
Table 7 duration of fixation average and the results of analysis of variance of standard deviation
All there is significant difference in duration of fixation average and standard deviation, upon extracting between window when being 30s, F value is maximum, because of
This, optimal time window is 30s.
7) time window is 30s and 60s, and driver's sweep amplitude average under different fatigue state and standard deviation are carried out Dan Yin
Element variance analysis, result is as shown in table 8 below.
Table 8 sweep amplitude average and the results of analysis of variance of standard deviation
No matter time window is 30s or 60s, and sweep amplitude average and standard deviation are in given significant level α=0.05 feelings
Under condition, there is not significant difference, therefore sweep amplitude cannot function as the characteristic parameter that fatigue driving differentiates.
8) time window is 30s and 60s, and average and standard deviation that driver under different fatigue state sweeps average speed are entered
Row one factor analysis of variance, result is as shown in table 9 below.
The results of analysis of variance of average speed average and standard deviation swept by table 9
No matter time window is 30s or 60s, pan average speed average in the case of given significant level α=0.05,
There is not significant difference, therefore pan average speed average cannot function as the characteristic parameter that fatigue driving differentiates；The average speed of pan
Degree standard deviation is when 30s, and F is maximum, and therefore, optimal time window is 30s.
9) time window is 30s and 60s, and driver's pupil diameter coefficient of variation under different fatigue state is carried out single factor test side
Difference analysis, result is as shown in table 10 below.
The results of analysis of variance of the table 10 pupil diameter coefficient of variation
When time window is 30s, F is maximum, and therefore, pupil diameter coefficient of variation optimal time window is 30s.
10) select Bonferroni to check as multiple comparisons method, draw the multiple comparisons of fatigue driving characteristic parameter
Result is as shown in table 11.
The result of multiple comparisons (α=0.05) of table 11 driving fatigue characteristic parameter
Can be seen that from upper table, in the case of given significant level α=0.05, between different fatigue level, all there is system
The characteristic parameter of significant difference in meter meaning have steering wheel angle standard deviation SASTD, steering wheel angular velocity absolute mean SWM,
Acceleration standard deviation Astd, pan average speed standard deviation SACV_std, pupil diameter coefficient of variation CVPLD；
Steering wheel absolute mean SAM between clear-headed and the most tired and tired and the most tired between there is significance
Difference, the identification of seven pairs of fatigue drivings still has certain contribution, therefore, it can the characteristic parameter as fatigue driving identification.
4, the characteristic parameter going out Preliminary screening carries out bivariate Spearman correlation analysis, specifically comprises the following steps that
4-1) characteristic parameter in source with same index is analyzed, the significant feature of dependency, one can only be selected；
4-2) to the characteristic parameter filtered out, move parametric classification by driving behavior parameter and eye and carry out correlation analysis, with
Sample, index significant to dependency can only select one；
4-3) remaining index after above two screenings being carried out correlation analysis, index significant to dependency is adopted
Take step 4-1), 4-2) same processing mode；
5, filter out optimal characteristics parameter to include: steering wheel angle standard deviation SASTD, vehicle acceleration standard deviation Astd,
Pan average speed standard deviation SACV_std and pupil diameter coefficient of variation CVPLD；Optimal characteristics parameter set is as shown in table 12 below.
Table 12 optimal characteristics parameter set
| Characteristic parameter | Implication | Optimal time window |
| CVPLD | The pupil diameter coefficient of variation | 30s |
| SACV_std | Pan average speed standard deviation | 30s |
| Astd | Vehicle acceleration standard deviation | 25s |
| SASTD | Steering wheel angle standard deviation | 10s |
6, fatigue driving BP neural network identification model is set up, it is first determined the input neuron number of this network structure, defeated
Go out neuron number, the implicit number of plies, hidden layer neuron number；The transmission function of next Internet, the power of each layer initial value is set
Value, finally determines the optimal learning efficiency, chooses suitable anticipation error, specific as follows:
6-1) input neuron, selects step 5) in optimal characteristics parameter and the input as model, and input data are entered
Row normalized；
6-2) output neuron, sets and is output as that 1-is clear-headed, 2-tired, 3-is the most tired, and represented by a neuron,
In the predicting the outcome of network, with 1.5 and 2.5 as boundary, if predictive value 1≤k≤1.5, then it is clear-headed；If predictive value 1.5
≤ k≤2.5, if being then tired predictive value 2.5≤k≤3, are then the most tired；
6-3) the implicit number of plies, single hidden layer, node in hidden layer is 9；
6-4) transmission function, the neutral net of BP is internal uses tansig to use purelin as transmission function, output layer
As transmission function；
6-5) initial weight, initial weight is set to the non-zero random value between (-1,1)；
6-6) the learning efficiency and anticipation error, uses LM learning algorithm, and the learning efficiency is automatically adjusted according to anticipation error, nothing
Need to set；Anticipation error is set as 0.001；
7, utilize Matlab software programming model program, randomly select training sample and test sample the set pair analysis model is instructed
Practice and fatigue state identification.
150 groups of sample datas are randomly selected from the fatigue driving sample database built, wherein clear-headed, tired, very
Tired sample each 50 groups as training sample, training sample data importing model is trained.
For the checking of model validity, the correct recognition rata of model can be as assessment indicator, and its computing formula is:
Wherein, n represents fatigue state, and 1-regains consciousness, and 2-is tired, and 3-is the most tired；
PnRepresent that model is to n-th grade of tired correct recognition rata；
qnRepresent in n-th grade of testing fatigue sample, the sample number being correctly validated；
QnRepresent n-th grade of testing fatigue total number of samples.
As the another one judging quota of model validity, its computing formula is as follows:
N=1,2,3；K=1,2,3；n≠k
Wherein, n, k represent fatigue state, and 1-regains consciousness, and 2-is tired, and 3-is the most tired；
PnkRepresent that model is identified as the correct recognition rata that k level is tired to n-th grade of fatigue；
rnkRepresent in n-th grade of testing fatigue sample, be mistaken for k level fatigue state number of samples.
Recruit 8 drivers, carry out above-mentioned identical experiment, record their driving behavior data and eye movement data, everyone
Extract each 30 groups of sample datas clear-headed, tired, the most tired to be used for the identification model trained is tested, test result
Accuracy rate and False Rate are as shown in table 13.
The table 13 fatigue driving identification model testing result to different drivers
Result shows that driver is regained consciousness by this model, three kinds of states tired, the most tired have arrived separately at 83.3%,
The average accuracy of identification of 69.6% and 79.6%, may be used for the fatigue state detection of driver.
The driver that the present invention selects has certain limitation, and individuality in fatigue machine posture detection is not completely eliminated
Difference, if expanding the sample size of fatigue driving sample database, accuracy of identification returns bigger raising, and model also can be made to have
There is adaptability widely.
Claims (2)
1. fatigue driving discrimination method based on driving behavior Yu eye movement characteristics, it is characterised in that comprise the following steps:
1) design fatigue driving simulation experiment, gathers driver's driving behavior data under different fatigue state by experiment,
Including the eye movement data of steering wheel angle, tarnsition velocity, speed, vehicle acceleration and driver, including blinking, watch attentively, sweeping
Data；
2) experimental data is carried out cutting screening, set up fatigue driving sample database；
3) use the method for one factor analysis of variance to the driving behavior parameter of driver under different fatigue state and eye movement characteristics
The significance of parameter differences carries out statistical analysis, and have found the optimal time window of various characteristic parameter；Preliminary screening goes out to drive
Cybernetics control number includes steering wheel absolute mean SAM, corner standard deviation SASTD, steering wheel angular velocity absolute mean SWM, angle
Velocity standard difference SWSTD, velocity standard difference Vstd, acceleration average Am and standard deviation Astd；Eye movement characteristics parameter includes nictation
Frequency BF, duration of fixation average FIXT_mean and standard deviation FIXT_std, pan average speed standard deviation SACV_std and
Pupil diameter coefficient of variation CVPLD；
4) characteristic parameter going out Preliminary screening carries out bivariate Spearman correlation analysis, specifically comprises the following steps that
4-1) characteristic parameter in source with same index is analyzed, the significant feature of dependency, one can only be selected；
4-2) to the characteristic parameter filtered out, move parametric classification by driving behavior parameter and eye and carry out correlation analysis, equally, right
The significant index of dependency can only select one；
4-3) remaining index after above two screenings being carried out correlation analysis, index significant to dependency takes step
Rapid 4-1), 4-2) same processing mode；
5) filter out optimal characteristics parameter to include: steering wheel angle standard deviation SASTD, vehicle acceleration standard deviation Astd, pan
Average speed standard deviation SACV_std and pupil diameter coefficient of variation CVPLD；
6) fatigue driving BP neural network identification model is set up, it is first determined the input neuron number of this network structure, output god
Through unit's number, the implicit number of plies, hidden layer neuron number；The transmission function of next Internet, the weights of each layer initial value are set,
After determine the optimal learning efficiency, choose suitable anticipation error, specific as follows:
6-1) input neuron, selects step 5) in optimal characteristics parameter and the input as model, and input data are returned
One change processes；
6-2) output neuron, sets and is output as that 1-is clear-headed, 2-tired, 3-is the most tired, and represented, at net by a neuron
In the predicting the outcome of network, with 1.5 and 2.5 as boundary, if predictive value 1≤k≤1.5, then it is clear-headed；If predictive value 1.5≤k≤
2.5, if being then tired predictive value 2.5≤k≤3, then it is the most tired；
6-3) the implicit number of plies, single hidden layer, node in hidden layer is 9；
6-4) transmission function, the neutral net of BP is internal uses tansig to use purelin conduct as transmission function, output layer
Transmission function；
6-5) initial weight, initial weight is set to the non-zero random value between (-1,1)；
6-6) the learning efficiency and anticipation error, uses LM learning algorithm, and the learning efficiency is automatically adjusted according to anticipation error, it is not necessary to set
Fixed；Anticipation error is set as 0.001；
7) utilize Matlab software programming model program, randomly select training sample and test sample the set pair analysis model is trained and
Fatigue state identification.
Method for detecting fatigue driving the most according to claim 1, it is characterised in that described fatigue driving simulation experiment is concrete
As follows:
1) experimental facilities:
A, driving analog system, by emulation vehicle, control station district, visual system, computer system, network system and corresponding
Software system forms；Primary operational equipment is all real vehicles equipment, by the input action of sensor capture user, by user
Input and other relevant simulation calculation that is simulated, simulation result is shown by visual system, sound simulation system simultaneously
Corresponding informance is fed back to user by system and signal output system；System also includes emulating this car dependency number in vehicle operation
According to export interface, different sample frequencys can be set according to Research Requirements, derive data and carry out secondary analysis；
B, Eye Link II type eye tracker, uses pupil pattern to be tracked；
2) experiment scene: using dull beltway as experiment scene, described dull beltway is two-way 6 cars of annular
The suburb dullness high-grade highway in road, lane width 3.75m, road total length 20km, evenness of road surface, weather is set to fine day；
3) experimenter: healthy, rule of sleeping, do not allow to drink in 24h before test, do not allow to drink in 12h before test
Coffee, tea；
4) experimental program: keeping the speed of 100km/s to travel at middle lane, the every 5min of staff records the most tested personnel
Subjective fatigue grade (by sleep yardstick table Karolinska sleepiness scale, KSS demarcate), until travel
60min, experiment terminates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610869724.9A CN106251583B (en) | 2016-09-30 | 2016-09-30 | Fatigue driving discrimination method based on driving behavior and eye movement characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610869724.9A CN106251583B (en) | 2016-09-30 | 2016-09-30 | Fatigue driving discrimination method based on driving behavior and eye movement characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106251583A true CN106251583A (en) | 2016-12-21 |
| CN106251583B CN106251583B (en) | 2018-09-25 |
Family
ID=57612122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610869724.9A Active CN106251583B (en) | 2016-09-30 | 2016-09-30 | Fatigue driving discrimination method based on driving behavior and eye movement characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106251583B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106611169A (en) * | 2016-12-31 | 2017-05-03 | 中国科学技术大学 | Dangerous driving behavior real-time detection method based on deep learning |
| CN107239895A (en) * | 2017-05-27 | 2017-10-10 | 西南交通大学 | The method and system examined for Sustained attention power |
| CN107301497A (en) * | 2017-05-27 | 2017-10-27 | 西南交通大学 | The method and system of working memory ability evaluation |
| CN107316436A (en) * | 2017-07-31 | 2017-11-03 | 努比亚技术有限公司 | Dangerous driving state processing method, electronic equipment and storage medium |
| CN108312964A (en) * | 2018-03-21 | 2018-07-24 | 华南理工大学 | A kind of fatigue driving state identification early warning system and method |
| CN108711016A (en) * | 2018-06-05 | 2018-10-26 | 合肥湛达智能科技有限公司 | A kind of driving behavior methods of marking based on BP neural network |
| CN109901446A (en) * | 2017-12-08 | 2019-06-18 | 广州汽车集团股份有限公司 | Controlling passing of road junction, apparatus and system |
| CN111292548A (en) * | 2020-02-06 | 2020-06-16 | 温州大学 | Safe driving method based on visual attention |
| CN111353636A (en) * | 2020-02-24 | 2020-06-30 | 交通运输部水运科学研究所 | Multi-mode data based ship driving behavior prediction method and system |
| CN111932829A (en) * | 2020-08-14 | 2020-11-13 | 北京智华通科技有限公司 | Fatigue driving prevention facility utility testing method and system |
| CN112052905A (en) * | 2020-09-11 | 2020-12-08 | 重庆科技学院 | Method for extracting multi-operation fatigue features of driver based on recurrent neural network |
| CN112750533A (en) * | 2020-11-03 | 2021-05-04 | 中国人民解放军63919部队 | Exoskeleton assistance effect evaluation method based on neural network |
| CN114170588A (en) * | 2021-12-13 | 2022-03-11 | 西南交通大学 | Railway dispatcher bad state identification method based on eye features |
| CN114424941A (en) * | 2022-01-26 | 2022-05-03 | 广东电网有限责任公司 | Fatigue detection model construction method, fatigue detection method, device and equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090040054A1 (en) * | 2007-04-11 | 2009-02-12 | Nec Laboratories America, Inc. | Real-time driving danger level prediction |
| CN101872171A (en) * | 2009-04-24 | 2010-10-27 | 中国农业大学 | Driver fatigue state recognition method and system based on information fusion |
| CN104207791A (en) * | 2014-08-26 | 2014-12-17 | 江南大学 | Fatigue driving detection method |
| CN204332007U (en) * | 2015-01-12 | 2015-05-13 | 江苏省交通规划设计院股份有限公司 | A kind of driving behavior early warning car-mounted terminal based on information fusion |
| CN105719431A (en) * | 2016-03-09 | 2016-06-29 | 深圳市中天安驰有限责任公司 | Fatigue driving detection system |
-
2016
- 2016-09-30 CN CN201610869724.9A patent/CN106251583B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090040054A1 (en) * | 2007-04-11 | 2009-02-12 | Nec Laboratories America, Inc. | Real-time driving danger level prediction |
| CN101872171A (en) * | 2009-04-24 | 2010-10-27 | 中国农业大学 | Driver fatigue state recognition method and system based on information fusion |
| CN104207791A (en) * | 2014-08-26 | 2014-12-17 | 江南大学 | Fatigue driving detection method |
| CN204332007U (en) * | 2015-01-12 | 2015-05-13 | 江苏省交通规划设计院股份有限公司 | A kind of driving behavior early warning car-mounted terminal based on information fusion |
| CN105719431A (en) * | 2016-03-09 | 2016-06-29 | 深圳市中天安驰有限责任公司 | Fatigue driving detection system |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106611169A (en) * | 2016-12-31 | 2017-05-03 | 中国科学技术大学 | Dangerous driving behavior real-time detection method based on deep learning |
| CN107239895A (en) * | 2017-05-27 | 2017-10-10 | 西南交通大学 | The method and system examined for Sustained attention power |
| CN107301497A (en) * | 2017-05-27 | 2017-10-27 | 西南交通大学 | The method and system of working memory ability evaluation |
| CN107316436B (en) * | 2017-07-31 | 2021-06-18 | 努比亚技术有限公司 | Dangerous driving state processing method, electronic device and storage medium |
| CN107316436A (en) * | 2017-07-31 | 2017-11-03 | 努比亚技术有限公司 | Dangerous driving state processing method, electronic equipment and storage medium |
| US11348455B2 (en) | 2017-12-08 | 2022-05-31 | Guangzhou Automobile Group Co., Ltd. | Intersection traffic control method, apparatus and system |
| CN109901446A (en) * | 2017-12-08 | 2019-06-18 | 广州汽车集团股份有限公司 | Controlling passing of road junction, apparatus and system |
| CN108312964A (en) * | 2018-03-21 | 2018-07-24 | 华南理工大学 | A kind of fatigue driving state identification early warning system and method |
| CN108711016A (en) * | 2018-06-05 | 2018-10-26 | 合肥湛达智能科技有限公司 | A kind of driving behavior methods of marking based on BP neural network |
| CN111292548A (en) * | 2020-02-06 | 2020-06-16 | 温州大学 | Safe driving method based on visual attention |
| CN111353636A (en) * | 2020-02-24 | 2020-06-30 | 交通运输部水运科学研究所 | Multi-mode data based ship driving behavior prediction method and system |
| CN111932829A (en) * | 2020-08-14 | 2020-11-13 | 北京智华通科技有限公司 | Fatigue driving prevention facility utility testing method and system |
| CN111932829B (en) * | 2020-08-14 | 2022-04-05 | 北京智华通科技有限公司 | Fatigue driving prevention facility utility testing method and system |
| CN112052905A (en) * | 2020-09-11 | 2020-12-08 | 重庆科技学院 | Method for extracting multi-operation fatigue features of driver based on recurrent neural network |
| CN112052905B (en) * | 2020-09-11 | 2023-02-03 | 重庆科技学院 | Method for extracting multi-operation fatigue features of driver based on recurrent neural network |
| CN112750533A (en) * | 2020-11-03 | 2021-05-04 | 中国人民解放军63919部队 | Exoskeleton assistance effect evaluation method based on neural network |
| CN114170588A (en) * | 2021-12-13 | 2022-03-11 | 西南交通大学 | Railway dispatcher bad state identification method based on eye features |
| CN114170588B (en) * | 2021-12-13 | 2023-09-12 | 西南交通大学 | Eye feature-based bad state identification method for railway dispatcher |
| CN114424941A (en) * | 2022-01-26 | 2022-05-03 | 广东电网有限责任公司 | Fatigue detection model construction method, fatigue detection method, device and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106251583B (en) | 2018-09-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106251583A (en) | Fatigue driving discrimination method based on driving behavior Yu eye movement characteristics | |
| CN107092881B (en) | A kind of human pilot changes system and method | |
| Daza et al. | Drowsiness monitoring based on driver and driving data fusion | |
| CN106960613B (en) | The assessment of non-intrusion type driver's potential danger identification capability and training system and method | |
| CN105620480B (en) | Intelligent vehicle independence lane-change opportunity decision-making technique | |
| CN113743471B (en) | Driving evaluation method and system | |
| CN103247091A (en) | Driving evaluation system and driving evaluation method | |
| Ābele et al. | Consistency between subjectively and objectively measured hazard perception skills among young male drivers | |
| Rakha et al. | Modeling driver behavior within a signalized intersection approach decision–dilemma zone | |
| CN103996287A (en) | Vehicle forced lane changing decision-making method based on decision-making tree model | |
| CN113753059B (en) | Method for predicting takeover capacity of driver under automatic driving system | |
| CN106057212A (en) | Driving fatigue detection method based on voice personality characteristics and models | |
| WO2024021871A1 (en) | Vehicle-road collaboration-based sensed roadside test data quality evaluating method | |
| Koh et al. | Smartphone-based modeling and detection of aggressiveness reactions in senior drivers | |
| CN103268426A (en) | Method for evaluating safe driving level of automobile driver | |
| CN106373390A (en) | Road traffic state evaluation method based on adaptive neuro fuzzy inference system | |
| Rhoads et al. | Observing our world | |
| CN106408032A (en) | Fatigue driving detection method based on corner of steering wheel | |
| Farah et al. | Association of risk proneness in overtaking maneuvers with impaired decision making | |
| CN117238126A (en) | Traffic accident risk assessment method under continuous flow road scene | |
| Vuckovic et al. | A sequential sampling account of response bias and speed–accuracy tradeoffs in a conflict detection task. | |
| CN115904091A (en) | Method and system for determining front distance of warning sign in operation area | |
| CN103927437B (en) | The method measuring space headway in non-rectilinear section | |
| CN108537161A (en) | A kind of driving of view-based access control model characteristic is divert one's attention detection method | |
| CN109649399A (en) | A kind of driver distraction's method of discrimination based on lateral running speed |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20180816 Address after: 214000 canal road 555-3908, Wuxi, Jiangsu Applicant after: SOL ELECTRONICS TECHNOLOGY CO., LTD. Address before: 538000 No. 2, Qiao Hing Street, Fangcheng District, Fangchenggang, the Guangxi Zhuang Autonomous Region. Applicant before: FANGCHENGGANG PORT DISTRICT GAOCHUANG INFORMATION TECHNOLOGY CO., LTD. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |