CN106022291A - Method of detecting braking intention of driver in emergency state based on neural signal - Google Patents
Method of detecting braking intention of driver in emergency state based on neural signal Download PDFInfo
- Publication number
- CN106022291A CN106022291A CN201610377967.0A CN201610377967A CN106022291A CN 106022291 A CN106022291 A CN 106022291A CN 201610377967 A CN201610377967 A CN 201610377967A CN 106022291 A CN106022291 A CN 106022291A
- Authority
- CN
- China
- Prior art keywords
- driver
- intention
- feature
- signal
- line
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/245—Classification techniques relating to the decision surface
- G06F18/2451—Classification techniques relating to the decision surface linear, e.g. hyperplane
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/15—Biometric patterns based on physiological signals, e.g. heartbeat, blood flow
Abstract
The invention relates to a method of detecting a braking intention of a driver in an emergency state based on a neural signal. According to the method provided by the invention, body movements and languages are not required, and only the electroencephalogram (EEG) signal of the driver is required to be recorded, and by analyzing the EEG signal, the braking intention of the driver is predicted, and early braking decision making is facilitated. The method belongs to an integrated application in a cognitive neuroscience field and an information technology field.
Description
Technical field
The present invention relates to a kind of method that the driver's of utilization nerve signal predicts that its brake is intended to.The present invention proposes
Method need not any limb motion and language, only need to record driver's EEG signals, by brain electricity
The analyses and prediction driver of signal brakes intention, assists it to make brake decision-making in advance.The invention belongs to cognition
The integrated application of neuroscience, areas of information technology and automation field.
Background technology
On-vehicle safety aid system is as alleviating driver's burden, it is ensured that driving safety, reduces accident casualty
Important ancillary technique has become as the research emphasis of current automobile industry.Wherein, auxiliary brake system is the party
To main direction of studying, the technical difficult points of this technology is the detection of Herba Plantaginis barrier.Currently, barrier
Hinder analyte detection technology to rely primarily on onboard sensor, such as laser, sonar, infrared, it addition, photographic head with
Image processing techniques is the most relatively commonly applied to detection of obstacles.But, these above-mentioned detection techniques would generally be subject to
Restriction to factors such as external environment, barrier outward appearances.
Brain-computer interface (BCI) can be set up a kind of direct between human brain and computer or other external devices
Communication and control passage, be that one is independent of and conventional brain output channel (nervus peripheralis and muscle
Tissue) brand-new information exchanging system.The ultimate principle of brain-computer interface is to make brain produce by certain mode
The raw EEG signals being suitable for identifying, and the EEG signals that will be produced by the method for signal processing and pattern recognition
" translate " one-tenth order, thus it is mutual to realize brain-machine.
Driver in driving procedure, vehicle front occur emergency, such as: pedestrian crosses suddenly horse
Road, front vehicles sails suddenly current lane into without the vehicle in intended braking or opposite side track, these feelings
Condition is required for driver and makes brake reaction in time.But, the emergency having occurred and that is found from driver,
To complete brake reaction, need experience visual information perception, Psychological Assessment, psychology decision-making, Motor preparation and
Execution process, this series of activity probably can take the time of 850 milliseconds, in the case of speed is higher,
It is likely to result in serious consequence.If able to by the EEG signals detecting driver, analysis obtains it and stops
Car is intended to, and skips Motor preparation and execution process, is transmitted directly to brake and drive system for vehicle with tow and completes brake, then may be used
To fulfil brake ahead of schedule, reduce accident casualty probability.By detection driver's EEG signals in emergency circumstances,
Predict that its brake intention becomes main purpose of the present invention.
Summary of the invention
The present invention relates to driver under a kind of state of emergency brake intention detection method, described method includes: step
Rapid 1, the EEG signals gathered is carried out pretreatment, it is thus achieved that the EEG signals after pretreatment;Step 2, by
Described pretreated EEG signals is calculated feature;Step 3, carries out feature extraction to described feature, looks for
Go out optimal characteristics combination;Step 4, sets up brake based on the combination of described optimal characteristics and is intended to detection model;Step
5, real-time judge driver brakes intention.
Described, the EEG signals gathered is carried out pretreatment and includes: step 11, original EEG signals is carried out
50Hz notch filter is to filter Hz noise;Step 12, is carried out the EEG signals after described notch filter
8 rank Butterworth lowpass ripples, cut-off frequency 60Hz;Step 13, to the brain electricity after described low-pass filtering
Signal carries out down-sampled process, original sampling frequency drop to 200Hz;Step 14, filters described down-sampled
Artefact nictation in EEG signals afterwards.
Described, step 14 includes: step 141, determines the mixed matrix of solution and filtering threshold by off-line data;
Step 142, filters artefact nictation in brain electricity in real time during application on site.
Described, step 141 includes: step 1411, and the EEG signals collecting off-line performs independent element
Analytic process (ICA) obtains independent element and described solution mixes matrix;Step 1412, in conjunction with described each independent one-tenth
The brain topological diagram divided, determines the independently composition relevant to artefact nictation, and calculates this described independent element
Approximation entropy (ApEn) is as described filtering threshold.
Described, step 142 includes: step 1421, the driver's EEG signals arrived by Real-time Collection and step
In 1411, gained solution mixes matrix multiple, it is thus achieved that the independent element of this segment signal;Step 1422, calculation procedure 1421
The approximation entropy of each independent element of gained, and compare with filtering threshold described in step 1412, entropy will be approximated little
Independent element in filtering threshold is set to zero;Step 1413, performs the inverse operation of independent component analysis, it is thus achieved that
Filtered EEG signals.
Described, step 2 farther includes: use fft Power estimation method, after the filtering described in calculation procedure 1
The spectrum signature of EEG signals, described feature includes the power spectrum of each passage 1Hz-60Hz (not including 50Hz)
Value, feature quantity=port number * 59.
Described, step 3 includes: step 31, the feature described in step 2 is divided according to the extraction time of sample
Becoming two groups, be normal group and emergency group respectively, described normal group is the sample extracted under driver's normal condition,
Described emergency group is the sample extracted after driver meets with emergency;Step 32, is grouped described in step 31
After feature carry out Wilcoxon rank test, find out the feature alternately feature pool of significant difference;Step
33, to feature pool execution sequence floating sweep forward algorithm (SFFS) described in step 32, extract optimum spy
Levy combination.
Described, step 4 farther includes: by the optimal characteristics combined training regularization line described in step 32
The correction factor of property grader (RLDA) and sorter model corresponding to described coefficient.
Described, step 5 includes: step 51, Real-time Collection driver's eeg data;Step 52, by up-to-date
The eeg data collected completes pretreatment according to the process described in step 1, and wherein, nictation, artefact filtered
Journey performs according to step 142;Step 53, according to feature combination calculation characteristic vector optimum described in step 32;
Step 54, by the sorter model set up described in characteristic vector input step 4 described in step 52 output category
As a result, and according to this output result judge that driver brakes intention.
The present invention proposes a kind of to have wide range of applications, driver easy to use, that accuracy rate is higher brakes intention
Forecasting Methodology, utilize driver to meet with in driving procedure EEG signals correlated characteristic that emergency causes
Change judges the state of driver.The method can apply in BAS.To most driver
Speech, after meeting with emergency, wrongful reaction will cause serious consequence, it is contemplated that urban transportation
The complexity of environment can affect the performance of other obstacle detection methods, therefore, by driver's experience
And the foundation that reaction is intended to as prediction brake, and brake reaction can be automatically made according to system judged result,
Will be substantially reduced Traffic Casualties, BAS advanced for development is significant.
Accompanying drawing explanation
Fig. 1 is the work system block diagram of the present invention;
Fig. 2 is that the present invention trains flow process;
Fig. 3 is that the present invention detects driver in real time and brakes intention flow chart.
Detailed description of the invention
Described by this invention based on EEG signals carry out brake be intended to monitoring method be particularly well-suited to improve drive
Safety, those skilled in the art according to the principle of this invention, can further extend other driver
Ancillary technique.
The ultimate principle of the present invention is when driver meets with emergency in driving procedure, is not required to make limbs
Or voice reacts, it is only necessary to driver's brain electricity analytical in emergency situations is processed, driver can be obtained
Brake whether is had to be intended to.
Under a kind of state of emergency provided the present invention with specific embodiment below in conjunction with the accompanying drawings, driver brakes meaning
Figure detection method is described in detail.
, here doing with explanation, in order to make embodiment more detailed, the following examples are the most meanwhile
Good, preferred embodiment, other alternative be may be used without for some known technologies those skilled in the art and
Implement;And accompanying drawing part is merely to more specifically describe embodiment, and it is not intended as the present invention
Carry out concrete restriction.
The present invention contain any make in the spirit and scope of the present invention replacement, amendment, equivalent method and
Scheme.Understand thoroughly to make the public that the present invention to be had, in present invention below preferred embodiment specifically
Understand concrete details, and do not have the description of these details to manage completely for a person skilled in the art
Solve the present invention.It addition, in order to avoid the essence of the present invention is caused unnecessary obscuring, the most specifically
Bright well-known method, process, flow process, element and circuit etc..
In an embodiment of the present invention, it is proposed that under a kind of state of emergency, driver brakes intention detection method,
With reference to Fig. 1, this system includes brain wave acquisition module, brain electricity analytical processing module.
Wherein, brain wave acquisition module is for Real-time Collection EEG signals and is amplified and analog digital conversion, logical
Cross data wire to carry out data transmission with processor.Wherein, it is considered to driver is meeting with emergency traffic situation time institute
The a series of mental activitys related to, relate generally to the visually-perceptible of traffic, emotion sudden change and promptly stop
Car motion planning, so according to " 10-20 international standard is led ", brain wave acquisition electrode is placed on user head
Cz, Pz, Fz, Oz, C3, C4, P3, P4, P7, P8, T7, T8, O1, O2, F3, F4 position in portion
Put, A11, A12 position reference electrode being placed on user ear-lobe.
Described, brain electricity analytical processing module is used for receiving EEG signals, and processes EEG signals,
Judge the state of driver.
Described, brain wave is carried out process and includes: step 1, disaggregated model training;Step 2, monitor in real time
Driver status.
Wherein, step 1 is specific as follows:
11) classification intercepts
Before first Application, need model training process.First driver completes in virtual driving scene
Driving task, be normal driving class and emergency class respectively, simultaneously record the two different experiments field
The eeg data that scape is corresponding.After completing task, the time occurred according to task in scene intercepts driver's brain electricity
Eeg data is also divided into two groups according to task kind by data, and carries out down-sampled by the eeg data after packet
To 200Hz.
12) filtering and noise reduction
Due to described down-sampled after EEG signals in there is different types of interference, therefore at filtering and noise reduction
In the stage, will perform according to procedure below: 121) data are carried out the 8 rank Barts fertile hereby 1Hz-60Hz logical filter of band
Ripple and 50Hz notch filter.122) filter in signal nictation interference, method therefor specifically: 1221)
To original eeg data application independent component analysis (ICA);1222) according to the approximation entropy of each independent element
Get rid of interference nictation in signal;1223) filtered brain electricity is obtained by the inverse operation of independent component analysis
Signal.123) described filtered EEG signals is performed grand mean with reference to (CAR) algorithm, it is therefore an objective to remove
Fall the common interference component existed in each people having a common goal.
Described, 1221) in, the detailed process of independent component analysis method is as follows:
During use, electrode sum is n, and this n electrode can get one group of data:
I represents the time sequencing of sampling, say, that there are the sampling of m group, and each group of sampling is all n dimension,
And owing to electrode allocation position is relatively near, cause obtained initial data to be really by several independent signal sources
Send the linear combination of signal.
The target of independent component analysis (ICA) is to isolate each signal source from this m group sampled data to send
Signal S, be expressed as:
S(S1,S2,...,Sn)T
The most one-dimensional signal being all an independent signal source and sending of S, so, can set up the relation of X Yu S.
X(i)=AS(i)
A is a unknown matrix, and X is known, and the process being released S by X is referred to as blind source signal separation.
Can do with down conversion:
W=A-1
S(i)=A-1X(i)=WX(i)
Can obtain:
W is described solution and mixes matrix, and how explained later obtains the mixed matrix W of solution.
It is all unknown due to W Yu S, on the premise of there is no priori, S can not be released by X.
Assuming that each SiThere is probability density PS, the Joint Distribution of the most a certain moment original signal is:
Thus can obtain:
Under there is no priori premise, need to choose a probability density function and be assigned to S, in theory of probability,
Density function p (x) obtained by Cumulative Distribution Function F (x) derivation.F (x) two attributes to be met is: monotonic increase
Fall in [0,1] with codomain.It is sigmoid function than better suited Certain function summary:
Obtain after derivation:
Here it is the probability density of S.
It is now know that ps(s), the most remaining W.According to the sample X obtained, its log-likelihood is estimated
As follows:
The iterative formula of W can be tried to achieve according to above formula:
After obtaining the mixed matrix W of solution by iteration, just can release S by X and send out to restore each independent signal source
The independent signal gone out, is the independent element performing to obtain after ICA algorithm shown in Fig. 3.
After completing independent component analysis computing, further work just determines that the threshold value of approximate entropy, by closely
Complete independent element is filtered like entropy.Described, 1223) in, the calculating process of approximate entropy is as follows:
A. set given length as N One-dimension Time Series u (i), i=1 ... N}, by formula:
Xi=u (i), u (i+1) ... u (i+m-1) }
Reconstruct m dimensional vector Xi, i=1 ... n, n=N-m+1;
B. arbitrarily vector X is calculatediWith itself and vector Xj(j=1,2 ..., N-m+1, j ≠ i) between distance:
dij=max | u (i+j)-u (j+k) |, k=0,1 ... m-1
Between i.e. two vector corresponding elements, the maximum of absolute difference is exactly the distance between two vectors;
C. threshold value r is given, between usual r=0.2~0.3, to each vector XiStatistics dij≤ r × SD, (SD is
The standard value of sequence) number with distance sum (N-m) ratio, be designated as
D. willTake the logarithm, more all of i is averaged, be designated as φm(r):
E.m increases by 1, repeats A-D step, tries to achieveAnd φm+1(r)。
F. by φm+1(r) and φmR () tries to achieve approximate entropy.
G. for there being limit for length's sequence of events, ApEn can estimate to obtain by statistical value:
ApEn=φm-φm+1
According to the process of approximate entropy algorithm, the periodicity of a segment signal is the highest, and its approximation entropy is the lowest,
Signal is the most complicated, and approximation entropy is the highest.The periodicity existed due to signal of blinking, can be according to its approximate entropy
Whether the residing scope of value determines the threshold value of signal of blinking, be then that nictation is dry according to each independent element of threshold determination
Disturbing, the nictation in removing independent element performs the inverse operation of independent component analysis after interference component, it is thus achieved that filter
Signal after ripple, can complete filtering.
13) feature extraction
Feature extracting method used by this method is Wilcoxon rank test and order forward floating searching algorithm
(SFFS) combining, find out optimal characteristics combination, its basic thought and calculating process be:
131) calculate the power spectral value of 16 passage 1Hz-60Hz in two class samples respectively, totally 960 original
Feature;
132) respectively to step 131) described in primitive character perform Wilcoxon rank test, confidence is set
The factor 0.05, finds out the significance level all features less than confidence factor, alternately feature pool;
133) to step 132) described in feature pool execution sequence floating sweep forward algorithm, find out optimum
Feature combination is combined as the feature used by on-line checking driver intention.
14) train classification models
The linear discriminant method that the feature of two groups of data of above-mentioned gained substitutes into regularization is set up driver status
Detection model.Described, driver status detection model establishment step is as follows:
Select the two class samples carrying out classifying, two class samples are demarcated, such as judging to drive
Whether member is in a state of emergency.Assume that the sample being in a state of emergency is X1Class, is in the sample of normal condition
For X2Class;
A. the Different categories of samples sample mean vector m at higher dimensional space is calculatedi;
B. the within class scatter matrix S of sample is calculatedi, total within class scatter matrix SwWith inter _ class relationship square
Battle array Sb;
C. criterion function is determined
A) Different categories of samples is in the average of projector space:
B) Different categories of samples is at the within class scatter matrix S of projector spacei, total within class scatter matrix Sw and class
Between scatter matrix Sb:
Sw=S1+S2
Sb=(m1-m2)(m1-m2)T
C) sample x and its project the relation between the statistic of y:
Sb=(m1-m2)(m1-m2)T=(wTm1-wTm2)(wTm1-wTm2)T
=wT(m1-m2)(m1-m2)TW=wTSbw
S1+S2=wT(S1+S2) w=wTSww
D. the criterion determining projecting direction w is: make the projection of sample in former state class the most in the direction the closeest
Collection, between class, the projection of sample separates as far as possible, and best projection direction is just so that JFThe w of acquirement extreme value:
In RLDA algorithm, need w is modified:
Wherein, λ is modifying factor, by 11) described two groups of off-line data rate of accurateness acquisition, value
Scope is [0,1], and I refers to w with the unit matrix of dimension, and trace (w) refers to the mark of w, and d (w) refers to the dimension of w
Degree.
Threshold value w0 choose employing ROC curve.ROC curve is a kind of threshold value for detecting two classification problems
Function curve, is a series of different cut off value according to two classification problems, with kidney-Yang rate (True Positive
Rate) it is vertical coordinate, is the function curve that abscissa is drawn with false sun rate (False Positive Rate).
Classification performance when choosing different cut off value (threshold value) can be found out very easily by ROC curve.Use
Time, optimal cut off value can be selected in conjunction with the analysis to practical problem.
Described, step 1 trains process as shown in Figure 2.
Wherein, step 2 is specific as follows:
21) calculation procedure 12) described in solution mix matrix and approximate entropy threshold value;
22) with window width 1s, step-length 0.1s Real-time Collection driver eeg data in driving procedure.
Described to this 1s data handling procedure it is:
The most down-sampled to 200Hz, 8 rank Barts fertile hereby 1Hz-60Hz bandpass filtering, 50Hz notch filter;
Ii. by step 21) solution that calculates mixes matrix, and described 1s eeg data is decomposed into independent element, and
Calculate the approximate entropy of each independent element;
Iii. by approximate entropy and the step 21 of described each passage) described approximate entropy threshold value compares, and little
It is set to same latitude null vector in the independent element that the approximate entropy of threshold value is corresponding;
Iv. perform inverse ICA computing, obtain filtering the eeg data after artefact nictation;
V. the eeg data filtering artefact nictation described in iv is performed grand mean with reference to (CAR) algorithm;
Vi. according to step 133) optimal combination of characters that obtains of described search calculates correlated characteristic vector, input
Step 14) described disaggregated model, output category result.
Described, the process of step 2, with reference to Fig. 3, wherein 301 refers to by 22) described 1s eeg data,
According to 133) characteristic vector that described optimal characteristics combination calculation obtains, 302 refer to 14) described train
The regularization linear discrimination classification device arrived, 303 fingers are dodged or it in order to assist driver to make in an emergency situation
The means of his decision-making or equipment, because of it not in covering scope of the present invention, be not described in detail.
Finally it should be noted that above example is only in order to describe technical scheme rather than to this skill
Art method limits, and the present invention can extend to other amendment in application, change, applies and implement
Example, and it is taken as that all such amendments, change, apply, embodiment all the present invention spirit and
In teachings.
Claims (3)
1. under the state of emergency based on nerve signal, driver brakes an intention detection method, including:
Step 1, off-line training process, including the pretreatment of off-line data, feature extraction and model training;
Step 2, on-line checking driver brakes intention process.
Under a kind of state of emergency based on nerve signal the most according to claim 1, driver brakes intention detection method, and wherein, step 1 includes:
Step 11, carries out pretreatment to the eeg data of off-line collection, specifically includes the fertile hereby 1Hz-60Hz bandpass filtering of 8 rank Barts, 50Hz notch filter, and nictation of based on independent component analysis and approximate entropy, artefact filtered and grand mean reference;
Step 12, characteristic extraction procedure particularly as follows: calculated the power spectral value of 16 passage 1Hz-60Hz by fft, then analyzed the significance of difference of each feature by Wilcoxon rank test, in the significant feature of all differences, finally searched for the feature combination obtaining optimum by order forward floating searching algorithm;
Step 13, according to off-line data, training obtains the linear discrimination classification device of regularization.
Under a kind of state of emergency based on nerve signal the most according to claim 1, driver brakes intention detection method, and wherein, step 2 on-line checking driver brakes intention process:
Step 21, with the eeg data of 1s window width Real-time Collection driver;
Step 22, carries out pretreatment to the eeg data collected according to step 11;
Step 23, according to the characteristic type comprised in the optimal characteristics combination that step 13 obtains, is calculated eeg data characteristic of correspondence vector described in step 21;
Step 24, by the grader described in the characteristic vector input step 13 described in step 23, obtains driver status information, is used for judging whether it has brake to be intended to.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610377967.0A CN106022291A (en) | 2016-05-31 | 2016-05-31 | Method of detecting braking intention of driver in emergency state based on neural signal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610377967.0A CN106022291A (en) | 2016-05-31 | 2016-05-31 | Method of detecting braking intention of driver in emergency state based on neural signal |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106022291A true CN106022291A (en) | 2016-10-12 |
Family
ID=57093149
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610377967.0A Pending CN106022291A (en) | 2016-05-31 | 2016-05-31 | Method of detecting braking intention of driver in emergency state based on neural signal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106022291A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108354605A (en) * | 2017-12-20 | 2018-08-03 | 佛山科学技术学院 | The extraordinary teenager's brain function of mathematics based on tranquillization state EEG signal connects network analysis method |
CN108693973A (en) * | 2018-04-17 | 2018-10-23 | 北京理工大学 | A kind of emergency detecting system of fusion EEG signals and environmental information |
CN108874137A (en) * | 2018-06-15 | 2018-11-23 | 北京理工大学 | A kind of gesture motion based on EEG signals is intended to the universal model of detection |
CN112905001A (en) * | 2021-01-19 | 2021-06-04 | 武汉理工大学 | Driving sound mode switching method and system based on electroencephalogram signals |
CN113212410A (en) * | 2021-04-28 | 2021-08-06 | 三峡大学 | Brain wave intelligent driving system |
CN113940688A (en) * | 2021-10-21 | 2022-01-18 | 北京理工大学 | Driving emergency steering intention detection method fusing friction nano generator and electroencephalogram |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103085815A (en) * | 2013-01-17 | 2013-05-08 | 北京理工大学 | Method for recognizing lane changing intention of driver |
CN103434502A (en) * | 2013-09-04 | 2013-12-11 | 厦门大学 | Emergency brake system based on brain-computer interface technique |
CN104461007A (en) * | 2014-12-19 | 2015-03-25 | 北京理工大学 | Driver-car interactive system assisting driver based on electroencephalograms |
CN104504400A (en) * | 2015-01-08 | 2015-04-08 | 中国科学院重庆绿色智能技术研究院 | Detection method of driver's abnormal behavior modeled based on online behavior |
CN104490391A (en) * | 2014-12-19 | 2015-04-08 | 北京理工大学 | Combatant state monitoring system based on electroencephalogram signals |
CN105468138A (en) * | 2015-07-15 | 2016-04-06 | 武汉理工大学 | Intelligent vehicle obstacle avoidance and navigation method based on brain-computer interface technology and lidar |
-
2016
- 2016-05-31 CN CN201610377967.0A patent/CN106022291A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103085815A (en) * | 2013-01-17 | 2013-05-08 | 北京理工大学 | Method for recognizing lane changing intention of driver |
CN103434502A (en) * | 2013-09-04 | 2013-12-11 | 厦门大学 | Emergency brake system based on brain-computer interface technique |
CN104461007A (en) * | 2014-12-19 | 2015-03-25 | 北京理工大学 | Driver-car interactive system assisting driver based on electroencephalograms |
CN104490391A (en) * | 2014-12-19 | 2015-04-08 | 北京理工大学 | Combatant state monitoring system based on electroencephalogram signals |
CN104504400A (en) * | 2015-01-08 | 2015-04-08 | 中国科学院重庆绿色智能技术研究院 | Detection method of driver's abnormal behavior modeled based on online behavior |
CN105468138A (en) * | 2015-07-15 | 2016-04-06 | 武汉理工大学 | Intelligent vehicle obstacle avoidance and navigation method based on brain-computer interface technology and lidar |
Non-Patent Citations (8)
Title |
---|
A J BELL 等: "An information-maximization approach to blind separation and blind deconvolution", 《NEURAL COMPUTATION》 * |
TENG TENG 等: "Using EEG to Recognize Emergency Situations for Brain-controlled Vehicles", 《2015 IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV)》 * |
周昌乐: "《智能系统与技术丛书 智能科学技术导论》", 30 November 2015 * |
施克 等著: "《BCI2000与脑机接口》", 30 June 2011 * |
梁静坤 等: "基于听觉感知任务下的脑电驾驶实验研究", 《中国生物医工程学报》 * |
牛清宁: "基于信息融合的疲劳驾驶检测方法研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》 * |
王忠诚 主编: "《王忠诚神经外科学》", 31 January 2015 * |
贾民谊 等主编: "《现代临床医学理论与实践》", 31 August 1995 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108354605A (en) * | 2017-12-20 | 2018-08-03 | 佛山科学技术学院 | The extraordinary teenager's brain function of mathematics based on tranquillization state EEG signal connects network analysis method |
CN108693973A (en) * | 2018-04-17 | 2018-10-23 | 北京理工大学 | A kind of emergency detecting system of fusion EEG signals and environmental information |
CN108693973B (en) * | 2018-04-17 | 2020-09-11 | 北京理工大学 | Emergency condition detection system fusing electroencephalogram signals and environmental information |
CN108874137A (en) * | 2018-06-15 | 2018-11-23 | 北京理工大学 | A kind of gesture motion based on EEG signals is intended to the universal model of detection |
CN112905001A (en) * | 2021-01-19 | 2021-06-04 | 武汉理工大学 | Driving sound mode switching method and system based on electroencephalogram signals |
CN112905001B (en) * | 2021-01-19 | 2022-08-16 | 武汉理工大学 | Driving sound mode switching method and system based on electroencephalogram signals |
CN113212410A (en) * | 2021-04-28 | 2021-08-06 | 三峡大学 | Brain wave intelligent driving system |
CN113940688A (en) * | 2021-10-21 | 2022-01-18 | 北京理工大学 | Driving emergency steering intention detection method fusing friction nano generator and electroencephalogram |
CN113940688B (en) * | 2021-10-21 | 2022-05-17 | 北京理工大学 | Driving emergency steering intention detection method fusing friction nano generator and electroencephalogram |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106022291A (en) | Method of detecting braking intention of driver in emergency state based on neural signal | |
Hou et al. | Driver intention recognition method using continuous hidden Markov model | |
CN110575163B (en) | Method and device for detecting driver distraction | |
Bi et al. | A novel method of emergency situation detection for a brain-controlled vehicle by combining EEG signals with surrounding information | |
CN108693973A (en) | A kind of emergency detecting system of fusion EEG signals and environmental information | |
CN103824420A (en) | Fatigue driving identification system based on heart rate variability non-contact measuring | |
CN105468138A (en) | Intelligent vehicle obstacle avoidance and navigation method based on brain-computer interface technology and lidar | |
CN104490391A (en) | Combatant state monitoring system based on electroencephalogram signals | |
CN112644506A (en) | Method for detecting driver driving distraction based on model long-time memory neural network LSTM-NN | |
CN113487863A (en) | Multi-factor traffic risk real-time quantitative evaluation system and method | |
CN209258112U (en) | Vehicle DAS (Driver Assistant System) and vehicle | |
CN212009589U (en) | Video identification driving vehicle track acquisition device based on deep learning | |
CN108874137B (en) | General model for gesture action intention detection based on electroencephalogram signals | |
CN113633296A (en) | Reaction time prediction model construction method, device, equipment and readable storage medium | |
Dehzangi et al. | Unobtrusive driver drowsiness prediction using driving behavior from vehicular sensors | |
CN113276867A (en) | Brain wave emergency control system and method under automatic driving situation | |
Wowo et al. | Towards sub-maneuver selection for automated driver identification | |
CN111222477B (en) | Vision-based method and device for detecting departure of hands from steering wheel | |
Wang et al. | EEG-based emergency braking intention prediction for brain-controlled driving considering one electrode falling-off | |
CN115444432A (en) | Driver emergency braking intention detection method based on electroencephalogram signals | |
CN111098709B (en) | Unlocking and starting method and system for safe driving system | |
CN202067537U (en) | Chinese Voice control system with sonogram and spectrogram in cross-correlation for driver | |
CN112957044A (en) | Driver emotion recognition system based on double-layer neural network model | |
Zhang et al. | Emergency-braking distance prediction using deep learning | |
CN110956064A (en) | Driver behavior analysis and prediction system based on big data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161012 |