CN104461007B - A kind of driver assistance people's car mutual system based on EEG signals - Google Patents
A kind of driver assistance people's car mutual system based on EEG signals Download PDFInfo
- Publication number
- CN104461007B CN104461007B CN201410804459.7A CN201410804459A CN104461007B CN 104461007 B CN104461007 B CN 104461007B CN 201410804459 A CN201410804459 A CN 201410804459A CN 104461007 B CN104461007 B CN 104461007B
- Authority
- CN
- China
- Prior art keywords
- driver
- subsystem
- eeg signals
- ssvep
- eeg
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2203/00—Indexing scheme relating to G06F3/00 - G06F3/048
- G06F2203/01—Indexing scheme relating to G06F3/01
- G06F2203/011—Emotion or mood input determined on the basis of sensed human body parameters such as pulse, heart rate or beat, temperature of skin, facial expressions, iris, voice pitch, brain activity patterns
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Neurosurgery (AREA)
- General Health & Medical Sciences (AREA)
- Neurology (AREA)
- Health & Medical Sciences (AREA)
- Dermatology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The present invention relates to the people's car mutual system of auxiliary driver being controlled using EEG signals (EEG) a kind of and its Related Computational Methods.The task of people is divided into driving task (track keep and dangerous monitoring etc.) and non-driving task in people's Vehicular system (with the switch of car-mounted device of the driving task without direct relation etc., such as the switch of air-conditioning, the switch of music player etc.).Present invention aims at a kind of new interactive mode realized among non-driving task between driver and vehicle.With the method for the present invention, driver no longer carries out non-driving task with limbs, but uses brain.Driver only needs to carry out corresponding task according to the demand of oneself, and then system carries out Treatment Analysis to corresponding brain electric information, reads the demand of driver, and then realizes the execution of non-driving task, realizes a kind of new people's car mutual pattern.The invention belongs to the integrated application of Car design field, man-machine interaction science, Cognitive Neuroscience and automation field.
Description
Technical field
The present invention relates to the driver assistance people's car mutual system that a kind of utilization EEG signals are controlled.It is specific next
Say, the detection of non-driving task is carried out using parallel schema using EEG signals and the detection of this task, Ran Houyou whether is performed
Car-mounted device performs non-driving task.Specifically, carrying out non-drive using commonly used P300 current potentials among brain-computer interface
The detection of task, is made whether to perform the detection of this task using Steady State Visual Evoked Potential (SSVEP).P300 current potentials are related to
The time-domain information of EEG signals, Steady State Visual Evoked Potential (SSVEP) is related to the frequency domain information among EEG signals, so, enter
Row parallel processing is feasible.Method proposed by the present invention does not need driver to have any limbs fortune related to non-driving task
Correspondence EEG signals are analyzed by dynamic or language, it is only necessary to which driver carries out corresponding brain-computer interface task by system,
Realize the new interactive mode of people and vehicle among non-driving task.The invention belongs to Car design field, man-machine interaction science,
The integrated application of Cognitive Neuroscience and automation field.
Background technology
Among people-Vehicular system, the task of people generally comprises various driving tasks and non-driving task.Drive
Task is primarily referred to as the holding in track and the monitoring of road hazard, non-driving task generally include the use of various car-mounted devices with
And occur in the car among environment with driving task without directly related task.For non-driving task, at present, people's car mutual
System includes speech recognition system, gesture recognition system, eye and moves identifying system etc..Brain-computer interface is one kind in human brain and outside
Established direct links between equipment independent of nervus peripheralis and the information exchanging system of musculature, people can be straight by brain
Connect and enter exchanging and directly controlling external equipment for row information with external equipment.Brain-computer interface technology behaviour car interactive system is carried
May for a kind of new interactive mode.EEG signals are the summations of the electrical activity of all neurons in cerebral nervous system, be brain-
The carrier and medium of purpose identification are carried out in machine interfacing to human brain.Brain-computer interface passes through the extraction to EEG signals, processing
The order that EEG signals " translation " are grown up with identification, realizes control of the people to external equipment.At present, conventional EEG signals have
P300 current potentials and Steady State Visual Evoked Potential.
P300 current potentials are a compositions of event related potential, are that people is produced by the stimulation of the stimulus related to event
When raw " surprise ", the about 300ms time after, the crest of a positive potential occurs in EEG signals.If can know
Do not go out the crest of this positive potential, then just can determine corresponding stimulation, and then know the intention of people.Brain based on P300-
One important advantage of machine interface is the training that user needs not move through complexity, it becomes possible to obtain higher accuracy rate.Stable state is regarded
It is Evoked ptential (VEP) one kind to feel Evoked ptential (SSVEP), when people stares at the stimulation persistently flashed with certain frequency, greatly
Brain will produce EEG signals identical with frequency of stimulation or into resonant relationship, especially at brain occipitalia and top.If energy
Enough identify the EEG signals with this feature, then it may determine that whether people has stared at stimulation interface, and then read people's
It is intended to.
Man-machine interactive system based on EEG signals is exactly the correlation technique using brain-computer interface, is read using EEG signals
The intention of people is taken, so as to realize the control to car-mounted device.Such a new people's car mutual pattern is realized, is the main of the present invention
Purpose.
The content of the invention
According to application claims there is provided a kind of driver assistance people's car mutual system based on EEG signals, including stimulate
Display subsystem, brain wave acquisition subsystem, Processig of EEG information subsystem, as a result output subsystem, car-mounted device subsystem;Its
In, stimulate display subsystem application LCD (or CRT/HUD) to carry out visual stimulus to induce corresponding brain electric information to driver;
Brain wave acquisition subsystem carries out the real-time collection of brain electric information and carries out signal amplification and analog-to-digital conversion, then brain electric information
It is transferred to Processig of EEG information subsystem;Processig of EEG information subsystem receives brain electric information and it is handled;As a result
Output subsystem is judged and exported to the intention of driver most to terminate according to the testing result of Processig of EEG information subsystem
Really;Car-mounted device system enters the reception of line command and performs corresponding non-driving task.
A kind of driver assistance people's car mutual method based on EEG signals, methods described includes:Step 1, it will induce
The visual stimulus of P300 current potentials and the visual stimulus of SSVEP Steady State Visual Evoked Potentials is induced with LCD (or CRT/HUD etc.) display
Mode is supplied to driver;Step 2, driver's brain electric information is gathered in real time and carries out signal amplification and analog-to-digital conversion;Step
3, Processig of EEG information subsystem receives driver's brain electric information, and brain electric information is handled, and whether judges driver
Wish to perform non-driving task, while judging the non-driving task that driver is carried out;Step 4, as a result output subsystem according to brain
The result that power information processing subsystem judges determines last output;Step 5, car-mounted device system takes orders and performs phase
The non-driving task answered.
Described, visual stimulus subsystem includes inducing the stimulation interface of P300 current potentials and induces SSVEP stable state vision inductings
The stimulation interface of current potential.Described, the stimulation interface for inducing P300 current potentials is made up of 3*2 character matrixs, the character of each in matrix
Represent corresponding non-driving task.The stimulation interface of the induction SSVEP Steady State Visual Evoked Potentials chess is held for two by left and right
Disk lattice are constituted, and each rectangle is made up of 30*10 blockage (each square is the pixels of 20 pixel * 20).Visual stimulus
System shows driver by LCD (or CRT/HUD etc.).
Described, brain wave acquisition subsystem gathers EEG signals by the electrode for encephalograms being placed on scalp, and is put by brain electricity
Big device amplifies and exports pending EEG signals.
Described, step 3 includes:Step 31, the original brain electric information collected is handled, judges whether driver notes
Interface is stimulated depending on SSVEP, whether is ready to perform non-driving task with this determination driver;Step 32, the original brain to collecting
Power information is handled, and judge driver's selection is that P300 stimulates some in six targets in interface, is determined with this
Driver performs the non-driving task of a certain item.
It is described, step 4 elaborate for:If the result that Processig of EEG information subsystem judges, which is driver, wishes execution
Non- driving task, then the non-driving task that output driver performs;If the result that Processig of EEG information subsystem judges is
Driver is not intended to perform non-driving task, then do not export the non-driving task of driver's execution, and vehicle keeps original shape
State.
Described, step 31 includes:Step 311, Fast Fourier Transform (FFT) is carried out to EEG signals, obtains its power spectrum letter
Breath, extracts the relative power spectrum averag density using centered on 13Hz and is used as SSVEP features;Step 312, threshold selected offline is utilized
Value is classified, so that obtaining SSVEP stimulates the intention of the driver under interface, determines whether driver is ready to perform non-driving
Task.
Described, step 32 includes:Step 321, original EEG signals are overlapped to take out noise, be then filtered
And principal component analysis;Step 322, using the Time Domain Amplitude after Signal averaging as P300 current potential features, substitute into Fisher and linearly sentence
Other model is classified, to judge that the P300 of driver's selection stimulates some stimulation character among interface.
The present invention proposes a kind of new people's car mutual system based on EEG signals.The system can be applied to and vehicle phase
The people's car mutual design of pass.
Brief description of the drawings
Fig. 1 is work system block diagram of the invention;
Fig. 2 stimulates the display interface of display subsystem for the present invention's;
The corresponding channel position of EEG signals of Fig. 3 collections required for the present invention;
Fig. 4 is EEG Processing schematic flow sheet of the invention;
Fig. 5 is that user watches EEG signals power spectrum chart when SSVEP is stimulated attentively;
Fig. 6 is that user does not watch EEG signals power spectrum chart when SSVEP is stimulated attentively;
Embodiment
The method of the people's car mutual system based on EEG signals described by the invention can be widely applied to driving field
People's car mutual system, equally, can also be according to the basic equipment and principle of the invention, further other new people of development
Car interactive system.
The general principle of the present invention is when user needs to carry out non-driving task, it is not necessary to limbs or voice
Deng being operated, but use brain.For example, the A-F among interface, interface is stimulated to represent respectively as shown in Figure 2 different
Non- driving task, for inducing P300 current potentials, the gridiron pattern that both sides are flashed with fixed frequency lures for inducing SSVEP stable state visions
Generating position.By taking C as an example, such as C represents the opening air-conditioning of the task, and driver will open air-conditioning, then be accomplished by driver's output C
Order.Driver is accomplished by staring at region 201 with eyes, while with remaining light, that is, a part of notice is distributed to character C
On.C is often highlighted once, and driver is accomplished by within counting once.So count two or three times, SSVEP stable state vision inductings electricity
Position and P300 current potentials, which can just be induced, to be come.After being gathered through brain wave acquisition subsystem, Processig of EEG information subsystem is sequentially passed through
System, the processing of order output subsystem, finally open air-conditioning.If driver needs to export F, then driver then needs what is stared at
Region is 202.For other non-driving tasks, the corresponding order of output, driver need the region stared at determination method and
C, F method are similar above, it is therefore an objective to induce SSVEP Steady State Visual Evoked Potentials and P300 current potentials simultaneously.
A kind of driver assistance people based on EEG signals provided below in conjunction with the accompanying drawings with specific embodiment the present invention
Car interactive system is described in detail.
Meanwhile, do herein with explanation, in order that embodiment is more detailed, the following examples is most preferably, preferably
Embodiment, can also be implemented for some known technologies those skilled in the art using other alternatives;And accompanying drawing
Part is merely to more specifically describe embodiment, and be not intended as and the present invention is specifically limited.
The present invention covers any replacement, modification, equivalent method and scheme made in the spirit and scope of the present invention.For
Make the public have the present invention thoroughly to understand, concrete details is described in detail in present invention below preferred embodiment, and
Description without these details can also understand the present invention completely for a person skilled in the art.In addition, in order to avoid to this
The essence of invention cause it is unnecessary obscure, well-known method, process, flow, element and circuit are not described in detail
Deng.
In an embodiment of the present invention, it is proposed that a kind of driver assistance people's car mutual system based on EEG signals, should
System includes stimulating display subsystem, brain wave acquisition subsystem, Processig of EEG information subsystem, and as a result output subsystem, vehicle-mounted
Device subsystem, particular flow sheet refers to Fig. 1.It is described, stimulate display subsystem include induce P300 current potentials visual stimulus and
Two kinds of stimulus modalities of visual stimulus of SSVEP Steady State Visual Evoked Potentials are induced, they are with LCD (or CRT/HUD etc.) display side
Formula is supplied to user.
It is described, stimulate the stimulation interface of display subsystem as shown in Figure 2.The visual stimulus for inducing P300 current potentials contains 6
Individual goal stimulus, each goal stimulus represents a non-driving task, induces the vision thorn of SSVEP Steady State Visual Evoked Potentials
Swash and contain 2 same target stimulations.Wherein, according to existing induction P300 Potential Technologies, devised according to Oddball normal forms
The visual stimulus of P300 current potentials is induced, is arranged using 3*2 matrix forms, includes six flicker letters of A, B, C, D, E, F, each
Letter corresponds to a non-driving task respectively.According to existing induction SSVEP Steady State Visual Evoked Potential technologies, design induces
The visual stimulus of SSVEP Steady State Visual Evoked Potentials includes left and right two flicker gridiron patterns, and (each contains 30*10 small
Square).P300 Evoked ptential visual stimulus, which are distributed in, stimulates the middle part at interface;SSVEP Steady State Visual Evoked Potential visual stimulus
Being distributed in stimulates the left part and right part at interface.
Wherein, brain wave acquisition subsystem is used to gather real-time EEG signals and be amplified and analog-to-digital conversion, Zhi Houchuan
Processig of EEG information subsystem is defeated by be handled.P300 Evoked ptentials are primarily generated at the top area of brain, SSVEP stable states
VEP is primarily generated at the occipitalia region of brain, therefore, according to " 10-20 international standards lead ", by brain wave acquisition
Electrode is placed on Cz, Pz, Fz, Oz, P3, P4, P7, P8, O1, O2 position on user head, and reference electrode is placed on and used
A11, A12 position (each electrode position is as shown in Figure 3) on person's ear-lobe, grounding electrode ground connection.
Described, by stimulating display subsystem to run vision induced stimulation, two classes, which are stimulated, to be flashed simultaneously.User according to
Demand watches corresponding region attentively.
Wherein, P300 Evoked ptentials visual stimulus flicker rule is:Six letters of A-F all dodge one by one at random in each round
It is bright, and ensure that each letter only flashes once in being taken turns one, each letter flicker continues 90ms, two alphabetical blinking intervals
10ms, is 600ms (=(90+10) × 6) the time required to often wheel flicker.
Wherein, SSVEP Steady State Visual Evoked Potentials visual stimulus flicker rule is:Left and right side SSVEP stable state vision inductings
Current potential visual stimulus flicker per second 13 times.
Described, Processig of EEG information subsystem is used to receive EEG signals, and EEG signals are handled, and judges to drive
The intention for the person of sailing.The one section of EEG signals collected in real time are handled while carrying out two kinds, a kind of is that driver opens air-conditioning
Order detection, it is a kind of to be whether driver performs the detection for opening air conditioning demand.
Described, the processing carried out to brain electric information is lured including the processing related to P300 current potentials and with SSVEP stable state visions
The related processing in generating position, this two kinds processing are parallel to be carried out.
The processing related to P300 current potentials includes:Step 1, each stimulate when occurring of interception are starting point, to after stimulating and occur
The eeg data of respective channel is as the EEG signals to that should stimulate in the 512ms periods, and carries out eeg data pretreatment,
Including superposition denoising, filtering and principal component analysis;After step 2, pretreatment are finished, the amplitude for extracting EEG signals in time domain is folded
Plus as P300 current potential features, substitute into Fisher linear discriminant models and classified, to judge the stimulation interface of driver's selection
Some target in six P300 targets of upper display.
The processing related to SSVEP Steady State Visual Evoked Potentials includes:EEG signals are extracted, fast Fourier change is carried out
Change, obtain its power spectral information, the relative power spectral density for then extracting the frequency band using centered on 13Hz is used as SSVEP features;
Classification and Identification is carried out using threshold value, judges that driver watches SSVEP stimulations attentively and stimulated again without SSVEP is watched attentively.Specific flow
Figure is shown in Fig. 4.
Wherein, step 1 is specific as follows:
1) it is superimposed denoising
EEG signals are a kind of very faint seismic electrical signals, for the influence of Removing Random No, and in order to strengthen
Each sample, is respectively taken turns corresponding EEG signals and is overlapped, then averaged, it is possible to realize this purpose by P300 current potentials.
P300 current potentials take average by superposition many times, and its value does not still have too big change, and random noise, then can weaken very
It is many, so that, P300 current potentials will become readily apparent from, beneficial to identification.
2) filtering and noise reduction
Due to extraneous noise jamming, some eye movement interferences of driver in itself and action interference etc., brain telecommunications are added
Number it is filtered processing.EEG signals after the superposition of each passage need to carry out bandpass filtering, to eliminate low frequency action interference
And eye movement interference, cut-off frequency is 0.53-15Hz.
3) feature extraction
EEG signals still suffer from substantial amounts of data after superposition denoising and the pretreatment of filtering and noise reduction, substantial amounts of
Data can cause data redundancy, be unfavorable for calculating and handle.Therefore, principal component analytical method is applied in the present invention to EEG signals
Dimension-reduction treatment is carried out, to contain as far as possible many original information with as far as possible few packet.
Described, the specific calculation procedure of principal component analytical method is as follows:
I. standardized transformation
Xi is random sample variable in above formula,For sample average, SiFor sample standard deviation;
Ii. correlation matrix is calculated
In above formula, rij(i, j=1,2 ..., p) it is primal variable xiWith xjCoefficient correlation, its calculation formula is
Because R is real symmetric matrix (i.e. rij=rji), so need to only calculate triangle element thereon or lower triangle element i.e.
Can.
Iii. eigen vector is calculated
First solve characteristic equation | λ I-R |=0, obtain eigenvalue λi(i=1,2 ..., p), and characteristic value is suitable by size
Sequence is arranged, i.e. λ1≥λ2≥…≥λp≥0;Then obtain corresponding to each eigenvalue λiCharacteristic vector ei(i=1,2 ..., p).
Iv. the contribution rate and contribution rate of accumulative total of each principal component are obtained
Principal component ziContribution rate:
Contribution rate of accumulative total:
Generally select the eigenvalue λ that contribution rate of accumulative total reaches 85-95%1,λ2,…,λmCorresponding the first, the second ... ..., m
(m≤p) individual principal component.
V. principal component load is calculated
ekiFor λkK-th of component of the characteristic vector after correspondence standardization
Vi. after the load for obtaining each principal component, principal component scores can further be calculated according to formula (1):
By calculating the contribution rate of accumulative total of each principal component, preceding 50 principal components are selected as the feature of sample.Preceding 50 masters
The contribution rate of accumulative total of composition can reach more than 95%.
Afterwards, the sample of above-mentioned gained is substituted into the P300 discrimination models set up with Fisher linear discriminant methods, assert
That signal that maximum must be worth is the EEG signals for containing P300 information, it is possible to which determine driver's selection is which
P300 characters.
Described, in step 2, Fisher discriminating steps are as follows:
The two class samples to be classified are selected, two class samples are demarcated, such as judging in EEG signals
Whether P300 information is included.Assuming that the sample containing P300 compositions is X1Class, the sample without P300 compositions is X2Class;
A. sample mean vector m of the Different categories of samples in 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 matrix Sb;
C. criterion function is determined
A) average of the Different categories of samples in projector space:
B) within class scatter matrix S of the Different categories of samples in projector spacei, total within class scatter matrix SwAnd inter _ class relationship
Matrix Sb:
Sw=S1+S2
Sb=(m1-m2)(m1-m2)T
C) sample x and its project relation between y statistic:
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. determining projecting direction w criterion is:Making original sample, the projection of sample is as far as possible intensive in class in this direction, between class
The projection of sample is tried one's best separation, and best projection direction is just so that JFObtain the w of extreme value:
Threshold value w0Selection use ROC curve.ROC curve is that a kind of threshold function table for being used to detect two classification problems is bent
Line, is a series of different cut off value according to two classification problems, with kidney-Yang rate (True Positive Rate) for ordinate, with
False sun rate (False Positive Rate) is the function curve that abscissa is drawn.Can very easily it be found out by ROC curve
Choose classification performance during different cut off value (threshold values).In use, the analysis to practical problem can be combined, select optimal
Cut off value.
Described, among the processing related to SSVEP, using Welch classical spectrum estimate methods, the frequency spectrum for extracting EEG signals is special
Levy.The relative power spectrum averag density centered on 13Hz is calculated, specific computational methods are, by frequency band 12.5-13.5Hz's
The average value of the average value of power spectral amplitude ratio divided by frequency band 11-12.5Hz and 13.5-15Hz power spectral amplitude ratio, using this as with
Relative power spectrum averag density centered on 13Hz.Then it is compared with threshold value, if being greater than threshold value, assert driver's note
Interface is stimulated depending on SSVEP, that is to say, that driver wishes to open air-conditioning, conversely, assert that driver does not watch SSVEP stimulations attentively
Interface, that is, driver are not intended to open air-conditioning.By Fig. 5 and Fig. 6, it can clearly be seen that watching SSVEP interfaces attentively and not noting
Spectrogram depending on EEG signals in the case of two kinds of SSVEP interfaces is very different.
Finally, order output subsystem receives the judged result of Processig of EEG information subsystem and carried out according to judged result
Last order output judges.Open air-conditioning if the result that Processig of EEG information subsystem judges is driver and perform this
Order, then the order of order output subsystem output unlatching air-conditioning to air-conditioning port, last air-conditioning is opened.
Finally it should be noted that above example is only to describe technical scheme rather than to this technology method
Limited, the present invention application can above extend to other modifications, change, using and embodiment, and it is taken as that institute
Have such modification, change, using, embodiment all in the range of the spirit or teaching of the present invention.
Claims (9)
1. a kind of driver assistance people's car mutual system based on EEG signals, including stimulate display subsystem, brain wave acquisition
System, Processig of EEG information subsystem, as a result output subsystem, car-mounted device subsystem;Wherein, display subsystem application is stimulated
LCD or CRT or HUD carries out visual stimulus to induce corresponding brain electric information to driver;Brain wave acquisition subsystem carries out brain electricity
The real-time collection of information and signal amplification and analog-to-digital conversion are carried out, brain electric information is then transferred to Processig of EEG information subsystem
System;Processig of EEG information subsystem receive brain electric information simultaneously it is handled, to brain electric information carry out processing include with
P300 current potentials related processing and the processing related to SSVEP Steady State Visual Evoked Potentials, this two kinds processing are parallel to be carried out;Knot
Fruit output subsystem is judged the intention of driver according to the testing result of Processig of EEG information subsystem and exports final
As a result;Car-mounted device subsystem enters the reception of line command and performs corresponding non-driving task;Wherein, Processig of EEG information subsystem
The detection united to P300 current potentials and SSVEP Steady State Visual Evoked Potentials, using parallel fo, is that detection SSVEP current potentials induce
Whether while also judge the selected P300 of driver stimulate among character;Wherein, the visual stimulus of P300 Evoked ptentials is dodged
Bright rule is:Six letters of A-F all flash one by one at random in each round, and ensure that each letter only flashes one in being taken turns one
Secondary, each letter flicker continues 90ms, two alphabetical blinking intervals 10ms, is 600ms (=(90+ the time required to often wheel flicker
10)×6);As a result output subsystem according to the two class orders received from Processig of EEG information subsystem to non-driving task
Whether perform and judged;Among the processing related to SSVEP Steady State Visual Evoked Potentials, using Welch classical spectrum estimates
Method, extracts the spectrum signature of EEG signals;Calculate the relative power spectrum averag density centered on 13Hz, specific computational methods
For:By the average value of frequency band 12.5-13.5Hz power spectral amplitude ratio divided by frequency band 11-12.5Hz and 13.5-15Hz power spectrum
The average value of amplitude, using this as the relative power spectrum averag density centered on 13Hz, is then compared with threshold value, if
More than threshold value, then assert that driver has watched SSVEP attentively and stimulated interface, conversely, assert that driver does not watch SSVEP attentively and stimulates boundary
Face.
2. the driver assistance people's car mutual system according to claim 1 based on EEG signals, wherein, stimulate display
System is mainly used in P300 current potentials and SSVEP Steady State Visual Evoked Potentials among evoked brain potential information, wherein, induce P300 electricity
The visual stimulus of position and the visual stimulus of induction SSVEP Steady State Visual Evoked Potentials include multiple goal stimuluses respectively;Further,
Inducing the visual stimulus of P300 current potentials includes being distributed in six in the middle part of the stimulation interface representatives respectively flashed according to random sequence
The letter of six non-driving tasks;Inducing the visual stimulus of SSVEP Steady State Visual Evoked Potentials includes left and right two according to identical
The gridiron pattern of frequency scintillation, flicker frequency is set as 13Hz.
3. a kind of driver assistance people's car mutual method based on EEG signals, this method is applied to any one of claim 1-2
A kind of described driver assistance people's car mutual system based on EEG signals, is specifically included:
Step 1, the visual stimulus of P300 current potentials will be induced and induces the visual stimulus of SSVEP Steady State Visual Evoked Potentials with LCD
Or CRT or HUD display modes are supplied to driver;
Step 2, driver's brain electric information is gathered in real time and carries out signal amplification and analog-to-digital conversion;
Step 3, Processig of EEG information subsystem receives driver's brain electric information, and brain electric information is handled, and judges to drive
Whether the person of sailing wishes to perform non-driving task, while judging the non-driving task that driver is carried out;
Step 4, as a result output subsystem judges according to Processig of EEG information subsystem result determines last output;
Step 5, car-mounted device subsystem takes orders and performs corresponding non-driving task.
4. the driver assistance people's car mutual method according to claim 3 based on EEG signals, wherein, step 3 includes:
Step 31, the original brain electric information collected is handled, judges whether driver watches SSVEP attentively and stimulate interface, with
Whether this determination driver is ready to perform non-driving task;
Step 32, the original brain electric information collected is handled, judge driver's selection is that P300 is stimulated six in interface
Some in individual target, determines that driver performs the non-driving task of a certain item with this.
5. the driver assistance people's car mutual method according to claim 3 based on EEG signals, wherein, step 4 is detailed
It is illustrated as:If the result that Processig of EEG information subsystem judges is that driver wishes to perform non-driving task, then output is driven
The non-driving task that the person of sailing performs;If the result that Processig of EEG information subsystem judges is that driver is not intended to perform non-driving
Task, then do not export the non-driving task of driver's execution, vehicle keeps original state.
6. the driver assistance people's car mutual method according to claim 4 based on EEG signals, wherein, in step 31,
The step of processing, includes:
Step 311, Fast Fourier Transform (FFT) is carried out to EEG signals, obtains its power spectral information, extracted centered on 13Hz
Relative power spectrum averag density is used as SSVEP features;
Step 312, classified using threshold value, so that it is determined that whether driver is ready to perform non-driving task.
7. the driver assistance people's car mutual method according to claim 4 based on EEG signals, wherein, step 32 enters one
Step includes:
Step 321, original EEG signals are overlapped to take out noise, be then filtered and principal component analysis;
Step 322, using the Time Domain Amplitude after Signal averaging as P300 current potential features, Fisher linear discriminant models is substituted into and are carried out
Classification, to judge that the P300 of driver's selection stimulates some stimulation character among interface.
8. the driver assistance people's car mutual method according to claim 6 based on EEG signals, wherein:
In step 311, using Welch classical spectrum estimate methods, the spectrum signature of EEG signals is extracted, by frequency band 12.5Hz-
13.5Hz power spectral amplitude ratio average value divided by frequency band 11.0Hz-12.5Hz and 13.5Hz-15.0Hz power spectral amplitude ratio
Average value is as the relative power spectrum averag density centered on 13Hz, as feature;
In step 312, stimulate and do not watch attentively the sample of SSVEP stimulations to carry out threshold value for watching SSVEP attentively from driver, then
With obtained relative power spectrum averag density is handled in real time and threshold value compares size, so as to obtain result of determination.
9. the driver assistance people's car mutual method according to claim 7 based on EEG signals, wherein,
Step 321 further comprises:EEG signals comprising P300 information are overlapped, bandpass filtering are then carried out, afterwards
Dimension-reduction treatment is carried out to sample using the method for principal component analysis;
Step 322 further comprises:Obtained sample application Fisher linear discriminant methods are set up into P300 discrimination models, will be real
When brain electric information bring model into, the signal that must be worth maximum is identified as including the EEG signals of P300 information, so as to determine
The selected specific P300 of driver stimulates character.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410804459.7A CN104461007B (en) | 2014-12-19 | 2014-12-19 | A kind of driver assistance people's car mutual system based on EEG signals |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410804459.7A CN104461007B (en) | 2014-12-19 | 2014-12-19 | A kind of driver assistance people's car mutual system based on EEG signals |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104461007A CN104461007A (en) | 2015-03-25 |
CN104461007B true CN104461007B (en) | 2017-11-03 |
Family
ID=52907199
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410804459.7A Active CN104461007B (en) | 2014-12-19 | 2014-12-19 | A kind of driver assistance people's car mutual system based on EEG signals |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104461007B (en) |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107015489B (en) * | 2016-01-28 | 2020-03-31 | 长城汽车股份有限公司 | Brake control method and system for vehicle |
CN107015632A (en) * | 2016-01-28 | 2017-08-04 | 南开大学 | Control method for vehicle, system based on brain electricity driving |
CN106022291A (en) * | 2016-05-31 | 2016-10-12 | 北京理工大学 | Method of detecting braking intention of driver in emergency state based on neural signal |
CN107300969A (en) * | 2017-05-02 | 2017-10-27 | 昆明理工大学 | A kind of MP4 player devices and its control method based on Mental imagery |
CN108255293A (en) * | 2017-12-07 | 2018-07-06 | 中国航空工业集团公司西安航空计算技术研究所 | Eye moves-brain electricity mixing man-machine interface system framework |
CN108874137B (en) * | 2018-06-15 | 2021-01-12 | 北京理工大学 | General model for gesture action intention detection based on electroencephalogram signals |
CN109144277B (en) * | 2018-10-19 | 2021-04-27 | 东南大学 | Method for constructing intelligent vehicle controlled by brain based on machine learning |
KR20200129291A (en) * | 2019-05-08 | 2020-11-18 | 현대자동차주식회사 | Apparatus for controlling convenience device of vehicle and method thereof |
CN112783314B (en) * | 2019-11-07 | 2023-04-18 | 中国科学院上海高等研究院 | Brain-computer interface stimulation paradigm generating and detecting method, system, medium and terminal based on SSVEP |
CN112109718A (en) * | 2020-06-17 | 2020-12-22 | 上汽通用五菱汽车股份有限公司 | Vehicle control method, device and computer readable storage medium |
CN111643077B (en) * | 2020-06-19 | 2024-05-07 | 北方工业大学 | Identification method for complexity of traffic dynamic factors based on electroencephalogram data |
CN111913582B (en) * | 2020-08-18 | 2022-06-14 | 福州大学 | P300 brain-computer interface Chekerbard stimulation sequence generation method |
CN115390655A (en) * | 2021-05-24 | 2022-11-25 | 华为技术有限公司 | Man-machine interaction method, man-machine interaction device and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101571748A (en) * | 2009-06-04 | 2009-11-04 | 浙江大学 | Brain-computer interactive system based on reinforced realization |
CN102200833A (en) * | 2011-05-13 | 2011-09-28 | 天津大学 | Speller brain-computer interface (SCI) system and control method thereof |
CN102609090A (en) * | 2012-01-16 | 2012-07-25 | 中国人民解放军国防科学技术大学 | Electrocerebral time-frequency component dual positioning normal form quick character input method |
CN103472922A (en) * | 2013-09-23 | 2013-12-25 | 北京理工大学 | Destination selecting system based on P300 and SSVEP (Steady State Visual Evoked Potential) hybrid brain-computer interface |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20110072730A (en) * | 2009-12-23 | 2011-06-29 | 한국과학기술원 | Adaptive brain-computer interface device |
-
2014
- 2014-12-19 CN CN201410804459.7A patent/CN104461007B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101571748A (en) * | 2009-06-04 | 2009-11-04 | 浙江大学 | Brain-computer interactive system based on reinforced realization |
CN102200833A (en) * | 2011-05-13 | 2011-09-28 | 天津大学 | Speller brain-computer interface (SCI) system and control method thereof |
CN102609090A (en) * | 2012-01-16 | 2012-07-25 | 中国人民解放军国防科学技术大学 | Electrocerebral time-frequency component dual positioning normal form quick character input method |
CN103472922A (en) * | 2013-09-23 | 2013-12-25 | 北京理工大学 | Destination selecting system based on P300 and SSVEP (Steady State Visual Evoked Potential) hybrid brain-computer interface |
Also Published As
Publication number | Publication date |
---|---|
CN104461007A (en) | 2015-03-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104461007B (en) | A kind of driver assistance people's car mutual system based on EEG signals | |
CN104586387B (en) | Method for extracting and fusing time, frequency and space domain multi-parameter electroencephalogram characters | |
US20200367800A1 (en) | Method for identifying driving fatigue based on cnn-lstm deep learning model | |
CN102715911B (en) | Brain electric features based emotional state recognition method | |
CN103150023B (en) | A kind of cursor control system based on brain-computer interface and method | |
US8862581B2 (en) | Method and system for concentration detection | |
CN105877766A (en) | Mental state detection system and method based on multiple physiological signal fusion | |
CN107822623A (en) | A kind of driver fatigue and Expression and Action method based on multi-source physiologic information | |
CN109512442A (en) | A kind of EEG fatigue state classification method based on LightGBM | |
CN103472922A (en) | Destination selecting system based on P300 and SSVEP (Steady State Visual Evoked Potential) hybrid brain-computer interface | |
CN106909784A (en) | Epileptic electroencephalogram (eeg) recognition methods based on two-dimentional time-frequency image depth convolutional neural networks | |
CN107981997B (en) | A kind of method for controlling intelligent wheelchair and system based on human brain motion intention | |
CN109770924A (en) | A kind of tired classification method based on Hadamard product building brain function network and Method Using Relevance Vector Machine | |
CN106446811A (en) | Deep-learning-based driver's fatigue detection method and apparatus | |
CN106108894A (en) | A kind of emotion electroencephalogramrecognition recognition method improving Emotion identification model time robustness | |
CN110321783A (en) | A kind of MEG spike detection method and system based on 1D convolutional neural networks | |
CN103919565A (en) | Fatigue driving electroencephalogram signal feature extraction and identification method | |
CN103892829B (en) | A kind of eye based on common space pattern moves signal recognition system and recognition methods thereof | |
CN104887224A (en) | Epileptic feature extraction and automatic identification method based on electroencephalogram signal | |
CN103425249A (en) | Electroencephalogram signal classifying and recognizing method based on regularized CSP and regularized SRC and electroencephalogram signal remote control system | |
CN104490391B (en) | A kind of combatant's condition monitoring system based on EEG signals | |
CN110772249A (en) | Attention feature identification method and application | |
CN106128032A (en) | A kind of fatigue state monitoring and method for early warning and system thereof | |
CN105640500A (en) | Scanning signal feature extraction method based on independent component analysis and recognition method | |
CN106446849A (en) | Fatigue driving detection method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |