CN103770780B - A kind of active safety systems of vehicles alarm shield device - Google Patents
A kind of active safety systems of vehicles alarm shield device Download PDFInfo
- Publication number
- CN103770780B CN103770780B CN201410017718.1A CN201410017718A CN103770780B CN 103770780 B CN103770780 B CN 103770780B CN 201410017718 A CN201410017718 A CN 201410017718A CN 103770780 B CN103770780 B CN 103770780B
- Authority
- CN
- China
- Prior art keywords
- driver
- vehicle
- mind
- active safety
- attention
- 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
- 238000001514 detection method Methods 0.000 claims abstract description 32
- 230000004438 eyesight Effects 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000012360 testing method Methods 0.000 claims abstract description 8
- 238000000034 method Methods 0.000 claims description 44
- 210000001508 eye Anatomy 0.000 claims description 27
- 230000008569 process Effects 0.000 claims description 26
- 230000009471 action Effects 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 230000000007 visual effect Effects 0.000 claims description 7
- 230000004044 response Effects 0.000 claims description 4
- 238000013459 approach Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000007664 blowing Methods 0.000 claims description 2
- 210000005252 bulbus oculi Anatomy 0.000 claims description 2
- 238000009987 spinning Methods 0.000 claims 1
- 230000008901 benefit Effects 0.000 abstract description 2
- 230000000875 corresponding effect Effects 0.000 description 9
- 230000006872 improvement Effects 0.000 description 8
- 208000003443 Unconsciousness Diseases 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000009434 installation Methods 0.000 description 4
- 206010039203 Road traffic accident Diseases 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 206010003830 Automatism Diseases 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010009 beating Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W50/08—Interaction between the driver and the control system
- B60W50/14—Means for informing the driver, warning the driver or prompting a driver intervention
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2540/00—Input parameters relating to occupants
- B60W2540/26—Incapacity
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2552/00—Input parameters relating to infrastructure
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of active safety systems of vehicles alarm shield device, including:Driver attention's detection module, predicts the outcome for comparing human eye sight data with notice, and combines the state of mind that car body signal carrys out comprehensive descision driver;Vehicle-mounted active safety system module, for active analysis, judges the operation safe condition of vehicle, and point out driver;Comprehensive warning control module, for being judged to open or shield vehicle-mounted active safety system module according to the testing result of the current driver's state of mind, or is actively alerted to driver.Alarm mode can be the alarm of acousto-optic or other modes.The present invention has the advantages that realization principle is simple, good reliability, intelligence degree are high, can actively carry out security control in time.
Description
Technical field
The invention mainly relates to active safety systems of vehicles field, refer in particular to a kind of based on eye tracking and notice calculating phase
The active safety systems of vehicles alarm shield device of matching, for realizing active safety system(Or equipment)Warning strategies or
Alarm shield strategy.
Background technology
With the development of auto industry, automobile increasingly becomes social production and the important component in daily life;
At the same time, automotive safety problem is increasingly becoming focus of concern.According to statistics, about 75% in road traffic accident
Caused by being due to human error, and in accidents caused by these are artificial, it is due to that unconscious lane departure detection is led to have 19% again
Cause, 26% is due to preceding to caused by knocking into the back, and the generation of this two classes accident is all relevant directly with the state of mind of driver, most
Driver is shown as eventually estimates vehicle running environment response lag of inaccurate and driver etc..Therefore, driver will not note
It is to occur traffic accident main cause that power, which is concentrated on the traffic events closely related with current driving task,.
In order to avoid due to the not good caused accident of driver's state of mind, thering is practitioner to invent or have developed polytype
Deviation warning system, forward direction collision alarm system and pedestrian impact warning system, so as to accident generation before
The alarm or prompting of certain mode are carried out to driver, and then efficiently reduces the probability of accident generation.
Deviation warning system is relative to the method that the operation relation of road carries out accident judgement based on vehicle.Have from
Dealer discloses a kind of deviation alarm set based on single-chip.Carried out by the improper more road traveling action to vehicle
Detection detects come the driving condition to driver.
It is mainly before vehicle to collision system and judges relative distance and relative velocity between Current vehicle and front vehicles
Collision time is judged etc. index, system provides alert to driver when collision time is less than certain threshold value.There is practitioner public
A kind of forward direction crash device based on multi-sensor information fusion has been opened, millimetre-wave radar, imaging sensor and vehicle is used
Body sensors judge that the relativeness between Current vehicle and front truck carries out danger judgement and early warning.Separately there is practitioner's proposition
Prevent the thinking that the alarm method of rear-end collision and system are taken from falling within such method.
Pedestrian active protection technique is that front pedestrian is detected using sensor and is analyzed with the relativeness of this car
The risk of pedestrian is collided, so as to carry out early warning.Separately there is practitioner's human hair to understand a kind of pedestrian's anticollision detection method, by using meter
The method of calculation machine vision estimates vehicle and the probability of pedestrian collision come the pedestrian position in front of detecting.If probability is big
In certain threshold value, system can be alerted.
Above-described three classes method is for trial and overcome because driver is unconscious(Energy do not concentrate, fatigue, anti-
Should it is slow, doze off etc.), it is non-active caused by unsafe condition generation.Such as occurs unconscious deviation, with before
Car pedestrian it is too fast etc. apart from too small or closing speed.When above-mentioned situation occurs, carrying out alarm to driver can be to a certain degree
The generation of upper reduction accident.However, the appearance of above-mentioned situation is not necessarily because driver is unconscious caused, and it is likely to be and drives
The person of sailing(Especially experienced driver)Normal operating, such as can also produce deviation when not beating steering indicating light lane-change, overtake other vehicles
When potential range leading vehicle distance it is too small.
Therefore deviation, with front truck pedestrian apart from it is too small or close to it is too fast situations such as, might not all be dangerous shape
Condition.Therefore need not all it alert.Otherwise, this can cause driver to be sick of and produce distrust to corresponding active safety system
Sense, so as to cause driver to close active safety system or not install such active safety system additional.The hair of such a situation
Life will seriously limit the popularization of active safety system and use.Solving the key of this problem, to be to discriminate between above-mentioned condition be due to drive
Caused by the conscious active operation of the person of sailing, or caused by the unconscious operation of driver.There is practitioner to propose by being
It is no to play steering indicating light to judge whether driver is conscious operation.In fact, this is a kind of simple, indirect determination methods.
If driver plays steering indicating light, judgement is rational consciously in fact.But do not beat steering indicating light judge its it is unconscious be irrational.
Therefore, following two conditions be must simultaneously meet when active safety system is alerted:First, occurring in that to cause danger
Feature(Deviation, with front truck pedestrian apart from too small or close to too fast);2nd, these situations are the unconscious, non-of driver
Caused by the operation of active(Driver does not recognize imminent danger).When exist a kind of mode can with direct measurement or
When judging that driver has appreciated that As-Is, system should mask alarm signal.That is, being in this case
Make to occur in that deviation, with Qian Che pedestrian apart from it is too small or close to it is too fast situations such as, system will not also be alerted.Meet so
The active safety alarm of condition could really be played a part of improving drive safety, while allowing driver to trust.
The content of the invention
The technical problem to be solved in the present invention is that:The technical problem existed for prior art, the present invention provides one
Plant simple realization principle, good reliability, intelligence degree height, can actively carry out the active safety systems of vehicles of security control in time
Alarm shield device, so as to reduce the false alarm rate of current active safety system.The strategy and device are based on to driver
The current state of mind be monitored and judge, and this method is contactless and does not disturb driver normally to drive work
It is dynamic.
In order to solve the above technical problems, the present invention uses following technical scheme:
A kind of active safety systems of vehicles alarm shield device, including:
Driver attention's detection module, predicts the outcome, and combine car body for comparing human eye sight data with notice
Signal carrys out the state of mind of comprehensive descision driver;
Vehicle-mounted active safety system module, for active analysis, judges the operation safe condition of vehicle, and points out to drive
Member;
Comprehensive warning control module, is used for when active safety system judges to cause danger according to current driver's essence
The testing result of refreshing state judges to open or shield vehicle-mounted active safety system module alarm signal, or is actively accused to driver
It is alert.Alarm mode can be the alarm of acousto-optic or other modes.
As a further improvement on the present invention:The vehicle-mounted active safety system module is deviation alarm module, preceding
One or more into collision alarm module or pedestrian protection system.
As a further improvement on the present invention:Driver attention's detection module includes forward looking camera, eye tracker
And computer, the forward looking camera is installed in the windshield of vehicle, frontal scene and transmission for shooting vehicle
To computer;The eye tracker is arranged on the driving panel of vehicle, for gathering the eyeball concern information of driver and sending
To computer;The car body signal directly leads to computer from car body.
As a further improvement on the present invention:Driver attention's detection module is by two aspects to driver's essence
Refreshing state is judged:One is the driver attention point that actual attention sight and computation model by analyzing driver are predicted
Cloth, to judge the state of mind of driver;Two be by detecting whether driver carried out significant operation in a short time,
To judge the state of mind of driver.
As a further improvement on the present invention:Actual attention sight and the computation model prediction by analyzing driver
Driver attention be distributed to judge driver's state of mind the step of it is as follows:
(1):It is distributed according to the attention of the image prediction driver currently gathered.
(2):Count probability density of the blinkpunkt residing in prediction notice distribution map in l blinkpunkt sampling period
Situation, judge whether driver is invested in notice the traffic events that will cause danger according to these situations related
Region, if accident can be found in time.
As a further improvement on the present invention:The step(1)Detailed process be:
(1.1), gather present image I first, should while initialization notice desired distribution figure A (x)=0 of one complete zero
Figure is of the same size with original image;Then the scene in image is analyzed using traditional computer vision methods
(Enter driveway line, vehicle, the detection and tracking of pedestrian).Detect lane line to determine and be partitioned into road area lane, then
Determine deviation danger relevant range R_lane(The lane line region nearest apart from vehicle front-wheel line midpoint);Detection is simultaneously
Front vehicles region vehicle is partitioned into, it is then determined that going out front vehicles risk of collision relevant range R_vehicle(Apart from car
Vehicle region in nearest this track of front in front-wheel line midpoint);Detect and be partitioned into front pedestrian area pdstn, so
After determine front pedestrian impact danger relevant range R_pdstn(The pedestrian in the front nearest apart from vehicle front-wheel line midpoint
Region).
(1.2), calculated for the attention distribution probability at image x according to following formula:
A(x)=max(e-d(x,R_lane),e-d(x,R_vehicle),e-d(x,pdstn)) (1)
Wherein d (x, R) represents point x to region R Euclidean distance, if x ∈ R, d (x, R)=0, otherwise d (x, R)=min { d
(x,y∈R)};
(1.3), have prediction driver attention's distribution map, when active safety system judge may cause danger when, i.e.,
The current notice and response situation of driver can be judged according to the distribution situation of driver's blinkpunkt of collection;It is initial first
Change the empty array B that a length is l;Before new blinkpunkt collection, all elements in array move to left one, vacate most right
B (l);After current fixation point is collected, viewpoint position x is lived according to this, is last assignment of array, is made:
B(l)=A(x)=max(e-d(x,R_lane),e-d(x,R_vehicle),e-d(x,pdstn)) (2)
Then the index of the respond of current driver's is calculated:
If C is more than certain threshold value Tc, then can be determined that driver has had appreciated that current unsafe condition, it reacts energy
Power is good.This says that the appearance of present case is the conscious behavior of driver's active, is consequently belonging to safe condition without announcement
It is alert.
As a further improvement on the present invention:It is described by detecting it is significant whether driver was carried out in a short time
Operate to judge the state of mind system of driver, be specially:The last meaningful operation of detection is first had to away from current time
Time;Four timers Tt, Tb, Tr, Tw are initialized, steering indicating light of fighting each other successively, brake, wiper, action of blowing a whistle are detected;
If detecting the end signal acted accordingly, corresponding timer is reset, otherwise, the normal timing of timer;Every
In one detection cycle, the state of mind of driver is estimated using following formula:
D=max(e-Tt,e-Tb,e-Tr,e-Tw) (4)
If D is more than certain threshold value Td this moment, illustrate that driver has done significant action at no distant date, therefore this moment
It is clear-headed, Td is to treat training parameter.
As a further improvement on the present invention:Initial alignment further is carried out to driver attention's detection module, specifically
Step is as follows:
(1):Under dead ship condition, some objects are put in front of forward looking camera at certain position on vehicle, with these
Other intrinsic distant objects are used as reference-calibrating in object and scene;
(2):Under dead ship condition, support staff passes through certain specified being presented on the screen of forward looking camera content of shooting
One object, and guide driver to be watched attentively, the direction of visual lines of the driver that now record eye tracker is recorded this moment;
(3):The position of object is noted by changing or object is noted in itself, repeated several times step(2);
(4):Calibration result is verified;Now guiding driver notes objects different in scene, and eye tracker is estimated
The pilot's line of vision counted out is shown on the realtime graphic of forward looking camera shooting, is observed and is recorded these objects and detect
Whether eye position overlaps or approaches;Required if error is met, terminate initial alignment process and terminate;Otherwise, repeat to walk
Suddenly(2)Extremely(3), start to re-scale.
Compared with prior art, the advantage of the invention is that:
1st, the present invention can significantly reduce the false alarms rate of active safety warning system, improve driver to active safety
The trusting degree of system.Using the present apparatus and traditional active safety warning system as an entirety security alarm system, with biography
When system active safety system compares, the present apparatus introduces driver's state of mind factor, and the state of mind of driver(Referring to
The application submitted simultaneously)Can accurately it measure.By the monitoring to driver attention, traditional active can be accurately judged
The situation that security system is alerted belongs to the conscious behavior or automatism of driver.Under driver's automatism
Unsafe condition is only the root that results in an automobile accident, therefore must alert;And driver is consciously, produced by under the behavior of active
False unsafe condition, should not belong to the category of alarm.And traditional active safety system inaccurately distinguishes having for driver
, as long as meeting limited compromising feature is at work and alarm condition, therefore there is suitable number of times in conscious & subconscious behavior
False alarms, so as to have impact on trust of the driver to such system.The present apparatus can be by sentencing to driver's state of mind
It is disconnected, so as to solve this problem well.
2nd, the present invention is more accurate to driver's state of mind testing result.Some active safety warning systems before
In, also the state of mind to driver has carried out simple estimation.The deviation such as proposed in some practitioners alerts system
In system, system judges that driver's state of mind is good and shielding harness alarm signal if it find that driver opens steering indicating light, then
Number.But the method judges it is quite inaccurate to the state of mind of driver.If such as driver does not play steering indicating light simultaneously
Do not mean that driver handles the bad state of mind.Due to the present invention without dependence fatigue state or driver to particular event
React to judge the driving condition of driver indirectly, testing result can more accurately react driver and hand over present road
The judgement of interpreter's part and disposal ability.Under normal driving state, driver will necessarily be to the marking area in present road scene
Automobile driving and dispensing are carried out, therefore by actual notice situation and expects the comparison between notice situation, Ke Yiyou
The driving condition of effect ground estimation driver.The notice to driver is surveyed and the calculating of scene notice is respectively provided with very high simultaneously
Precision, therefore the state of mind of driver can be accurately judged that.
3rd, detection process of the invention is contactless.The detection method and detection process of the present invention need not be to driving
Member provides extra stimulation, without interference with the normal driving behavior of driver.On the other hand, detection process of the invention need not
By measuring the physiological parameter of driver with the direct contact of driver, measurement process is contactless.
Brief description of the drawings
Fig. 1 is the circuit theory schematic diagram that the present invention is constituted.
Fig. 2 is system scheme of installation of the present invention in concrete application example.
Fig. 3 is the frame structure principle schematic of driver attention's detection module in the present invention.
Fig. 4 is the overall procedure schematic diagram that the present invention works.
Fig. 5 is progress driver attention's prediction in the present invention(The calculating of notice desired distribution figure)Sub-process signal
Figure.
Fig. 6 is the sub-process schematic diagram of the current respond analysis of progress driver in the present invention.
Fig. 7 is progress driver actions' analysis in the present invention(Judge whether driver has carried out significant behaviour in a short time
Make)Sub-process schematic diagram.
Fig. 8 is the curve synoptic diagram of institute's gathered data in concrete application embodiment.
Embodiment
The present invention is described in further details below with reference to Figure of description and specific embodiment.
As shown in figure 1, the active safety systems of vehicles alarm shield device of the present invention, including:
Driver attention's detection module, predicts the outcome, and combine car body for comparing human eye sight data with notice
Signal carrys out the state of mind of comprehensive descision driver;
Vehicle-mounted active safety system module, for active analysis, judges the operation safe condition of vehicle, and points out to drive
Member;
Comprehensive warning control module, for being judged to open or shield car according to the testing result of the current driver's state of mind
Active safety system module is carried, or is actively alerted to driver.Alarm mode can be the alarm of acousto-optic or other modes.
The present invention combines existing vehicle-mounted active safety system module to form the higher active peace of new intellectuality
Total system.Existing above-mentioned vehicle-mounted active safety system module can include deviation alarm module, forward direction collision alarm mould
One or more in block, pedestrian protection system or other active safety systems.
From the foregoing, it will be observed that the present invention is the alarm shield device with reference to driver's state of mind detection technique, first to driving
The current state of mind of member is judged, if working as compromising feature(Deviation, Ben Che and front pedestrian's vehicle distances are too small
Or close to too fast)Go out current systems think current driver's it has been noted that or appreciate these compromising features, then system
Vehicle-mounted active safety system module can be masked(Vehicle-mounted deviation warning system, forward direction collision alarm system, pedestrian actively protect
The alarm signal of the active safety such as protecting system equipment or system);If system judges that current driver's do not recognize that these are endangered
Dangerous feature, then system can open all alarm signals of above-mentioned vehicle-mounted active safety system module.
As shown in Fig. 2 in concrete application example, driver attention's detection module includes forward looking camera, eye tracker
And computer(Do not show in figure), forward looking camera is installed in the windshield of vehicle, the frontal scene for shooting vehicle.
The installation of forward looking camera, which should be guaranteed that, can obtain the visual field more consistent with driver.Eye tracker is arranged on the driving of vehicle
On panel, it shall be guaranteed that eye tracker can rightly catch the sight of driver.Car body signal is directly drawn from car body(For example it is straight
Connect and extracted from car body CAN).Comprehensive warning control module is installed according to existing Installation Modes.
The present invention is main in the specific implementation to include two main processes:Initial alignment process and real-time calculating process.
Initial alignment process.As shown in Figure 2, vehicle-mounted forward looking camera is arranged on vehicle front windshield plate medial roof
End, primary optical axis is parallel with car body, and forward looking camera shoots vehicle front road information towards preceding installation.The present invention need to by
Eye tracker obtains driver of driver's actual measurement notice distribution with being obtained according to vehicle-mounted vidicon and notice computation model and noted
It is compared between meaning power distribution, the driver that road scene and eye tracker that answering this needs to obtain forward looking camera are captured
Need to be demarcated between sight, so that catching scene objects and mesh that obtained pilot's line of vision is watched attentively by eye tracker
It is marked on the position correspondence presented in video camera consistent.
In the present invention, initial alignment process is comprised the following steps that:
Step one:Under dead ship condition, in vehicle(Forward looking camera)20 meters of front(Or it is farther)Manually put one in place
A little objects, these objects should have significantly and be easy to the spherical object of feature, in this way red described.With these objects and field
Other intrinsic distant objects are used as reference-calibrating in scape;
Step 2:Under dead ship condition, support staff is presented on the screen of forward looking camera content of shooting by specifying
Some object(Point), and guide driver to be watched attentively.The sight of the driver that now record eye tracker is recorded this moment
Direction;
Step 3:The position of object is noted by changing or object is noted in itself, repeat step two, it usually needs change
Become 5 to 7 different objects(Point);
Step 4:Calibration result is verified.Now guiding driver notes objects different in scene(Point), by eye
The pilot's line of vision that dynamic instrument is estimated is shown on the realtime graphic of forward looking camera shooting, is observed and is recorded these objects(Point)
Whether overlap or approach with the eye position detected.Required if error is met(Such as less than 1.5 degree), then terminate initial
Calibration process terminates;Otherwise, repeat step two to three, starts to re-scale.
Due to driver's eyes height and forward looking camera setting height(from bottom) it is inconsistent, can cause front camera coverage with
There is certain error in the actual visual field that driver's eyes are seen.Specifically, for driver under the same visual angle of vehicle front
Different objects(Point)Cause to produce different visual angles in camera coverage with the difference of distance.By calculating, if peace
Difference in height control is filled in 20cm, then visual angle error can be controlled in 0.6 degree between the different objects outside vehicle front 20m,
This is already below the precision that eye tracker can be provided(About 1 degree)., can be by driver per for the moment by initially determining process with subscript
The position that watched attentively object is presented in front view picture is carved, the sight for observing driver this moment with eye tracker is corresponded
Get up.Therefore, the pilot's line of vision direction that can be observed at any time according to eye tracker is deduced before driver watches attentively
The particular location that square object body is presented in front view picture, it is possible thereby to the position in knowing the actual scene that driver notes.Need
It is noted that in order to ensure the convenience and accuracy of demarcation, it is necessary to control the distortion degree of forward looking camera, or pass through
The image that algorithm is shot to forward looking camera carries out going distortion to handle.
Real-time calculating process.As shown in figures 1 and 3, general principle of the invention and the course of work are as follows:
When currently there is compromising feature in active safety system judgement and will alert, spirit of the system first to driver
State is judged.System is judged driver's state of mind by two aspects:One is the reality by analyzing driver
Border notes driver attention's distribution of sight and computation model prediction, to judge the state of mind of driver;Two be by inspection
Survey whether driver carried out significant operation in a short time, to judge the state of mind of driver.System integrate this two
Plant the judgement that testing result provides driver's state of mind.When system assert that driver's state of mind is good, the synthesis in Fig. 1
ACM alarm control module will shield the alarm signal of all vehicle-mounted active safety systems;When system assert driver's state of mind not
When good, comprehensive warning control module will these open vehicle-mounted active safety systems alarm signal, make its be operated in it is traditional just
Normal alarm status.
When wherein system is judged driver attention, the actual attention sight and computation model of used driver is pre-
The basic step of the method for driver attention's distribution of survey is as follows:
Step one:It is distributed according to the attention of the image prediction driver currently gathered.Its calculating process is described in detail later;
Step 2:Blinkpunkt in (l blinkpunkt sampling period) was counted in nearest a period of time in prediction notice point
The situation of residing probability density in Butut.Judge whether driver is invested in notice and will occur according to these situations
The related region of dangerous traffic events(Track, close front vehicles or the pedestrian that such as will deviate from).
As shown in figure 5, the computational methods of prediction driver attention's distribution map are as follows.
Step 1: present image I is gathered first, while notice desired distribution figure A (x)=0 of initialization one complete zero,
The figure is of the same size with original image;Then the scene in image is analyzed using traditional computer vision methods
(Enter driveway line, vehicle, the detection and tracking of pedestrian).Detect lane line to determine and be partitioned into road area lane, then
Determine deviation danger relevant range R_lane(The lane line region nearest apart from vehicle front-wheel line midpoint);Detection is simultaneously
Front vehicles region vehicle is partitioned into, it is then determined that going out front vehicles risk of collision relevant range R_vehicle(Apart from car
Vehicle region in nearest this track of front in front-wheel line midpoint);Detect and be partitioned into front pedestrian area pdstn, so
After determine front pedestrian impact danger relevant range R_pdstn(The pedestrian in the front nearest apart from vehicle front-wheel line midpoint
Region).
Step 2: there are these regions, prediction driver attention's distribution map can be further calculated.As shown in fig. 6, right
Attention distribution probability at image x is calculated according to following formula:
A(x)=max(e-d(x,R_lane),e-d(x,R_vehicle),e-d(x,pdstn)) (1)
Wherein d (x, R) represents point x to region R Euclidean distance, if x ∈ R, d (x, R)=0, otherwise d (x, R)=min { d
(x,y∈R)};
Step 3: there is prediction driver attention's distribution map, you can according to the distribution feelings of driver's blinkpunkt of collection
Condition judges the current notice and response situation of driver.The collection of driver's viewpoint is 10 viewpoints of record per second.Specifically
Process is as shown in Figure 6:It is l to initialize a length first(L=50 in embodiment)Empty array B;Adopted in new blinkpunkt
Before collection, all elements in array move to left one, vacate most right B (l);After current fixation point is collected, lived according to this
Viewpoint position x, is last assignment of array, order:
B(l)=A(x)=max(e-d(x,R_lane),e-d(x,R_vehicle),e-d(x,pdstn)) (2)
Then the index of the respond of current driver's is calculated:
If C is more than certain threshold value Tc, then can be determined that the respond of driver is good.Threshold value TcDetermination process it is as follows:
Step 1: the set-up mode of driver operational data collection.20 drivers are invited to be tested as data acquisition.It is right
In each driver, the logger normal driving data of 10 hours.Here the data recorded are by formula(3)The data calculated(C corresponding to n-th of view collection moment of m-th of driver).
Step 2: threshold value Tc。
Training and alarm example.In parameter TcTraining process in, acquire 7200000 viewpoint institutes of 20 drivers
Corresponding TcData.Wherein minimum value TcFor 2.5079.
The state of mind of driver is judged except the blinkpunkt according to driver, system is always according to driver in a timing
It is interior whether to have carried out significant operation to judge its state of mind.As shown in fig. 7, system first has to the detection last time intentionally
Justice time of the operation away from current time.To realize this purpose, system initialization four timers Tt, Tb, Tr, Tw.Then according to
Secondary air exercise steering indicating light, brake, wiper, the action such as blow a whistle are detected.If detecting the end signal acted accordingly,
Corresponding timer is reset, otherwise, the normal timing of timer.In each detection cycle, system is using following formula to driving
The state of mind of member is estimated:
D=max(e-Tt,e-Tb,e-Tr,e-Tw) (4)
If D is more than certain threshold value Td this moment, illustrate that driver has done significant action at no distant date, therefore this moment
It is clear-headed.Here Td is to treat training parameter.Td training process is as follows:
Step one:20 drivers are invited to be tested as data acquisition.20 drivers drive one and are provided with respectively
The common vehicle of traditional deviation warning system, active safety system and pedestrian detecting system.Installed simultaneously on vehicle
There are steering indicating light, brake, wiper, operation signal harvester and the corresponding timer such as blow a whistle.If detecting corresponding actions
End signal, then corresponding timer is reset, otherwise, the normal timing of timer;
Step 2:When any one in deviation warning system, active safety system and pedestrian detection warning system
During individual ALM, the validity of this time alarm is judged by driver.If driver judges that current alarm effectively, sends collection
Data command(One button is set), system will record four values Tt, Tb, Tr, Tw of current timer, and by this four numerical value
Minimum value record it is stand-by:min(Tt,Tb,Tr,Tw), then four timers are reset simultaneously.
Step 3:Threshold value Td.The gathered data of all subjects of record is compared, their minimum value is taken out
T.Then T is maded=e-T。
Training and alarm example.In parameter Td training process, acquire 20 drivers 423 times effectively alert institute
Corresponding numerical value.Wherein minimum value T is 6.53s.Carried out this means driver in 6.53 seconds after significant operation
Focus on, thus will not make mistakes substantially.Thus Td=e-T=0.0015。
The overall course of work combination Fig. 4 of system is further described.As shown in figure 4, a certain moment, system is read first
Pick up the car and carry active safety warning system(Deviation warning system, forward direction vehicle collision warning system, pedestrian protection system etc.)
Alarm signal.
If the vehicle-mounted active safety warning system of system discovery have identified compromising feature(Deviation, with front vehicles row
People is too fast apart from too small or closing speed), then system follow the steps below judgement:
Step one:The process according to Fig. 7 judges whether driver has carried out significant operation.When driver is had
During the operation of meaning, show that driver is good during its state of mind within a period of time before and after operation.If system detectio
Significant operation has been carried out within certain moment before to driver, it can be assumed that the state of mind is good this moment by driver,
System directly shields the alarm signal of existing vehicle-mounted active safety system, in case false alarms bother driver.
Step 2:If driver is not detected in step one has carried out significant operation.System not can determine that and drive
The person's of sailing state of mind is bad, it is therefore desirable to be further analyzed.System carries out road extraction to image simultaneously and driver is regarded
The collection of line;According to the road image in front of the current car body of collection, scene analysis calculating is carried out and to the note to driver
Meaning power is predicted(Detailed process is as shown in Figure 5);The current sight data of driver to record and the attention of prediction simultaneously
Power is distributed, and the process according to Fig. 6 is analyzed, so as to complete the judgement again to the state of mind of driver.Such as system
Judge that driver's state of mind is good, then system directly shields the alarm signal of existing vehicle-mounted active safety system, Yi Mianxu
Driver is bothered in vacation alarm;
Step 3:If the state of mind that driver is detected in step 2 is bad, system will open all vehicle-mounted actives
Security system is alerted.
Concrete application example:Two are invited to be tested.For the sake of security, everyone tests 10 seconds.First
Subject carries out normal driving within this time.And second place is tested in preceding 5 seconds normal drivings, the subject is closed in latter 5 seconds
Eyes are to simulate fatigue driving(Note:For the sake of security, this time experiment copilot can be braked at any time).First place
Subject does not carry out steering indicating light, brake, wiper, the operation such as blow a whistle all the time in driving procedure, and second place subject was at 2.2 seconds
Moment big steering indicating light is once.During this period of time, two driver Jun Shiyue roads line balls(Generation deviation)Traveling.
Due to being that more road line ball is travelled, therefore vehicle-mounted active safety system detects compromising feature and will alerted.This
When system will according to shown in Fig. 4 flow carry out it is a series of judge.As shown in figure 8, being according to formula(3)Two calculated drive
The state of mind situation for the person of sailing.The situation of first place driver is analyzed first.Because no driver within the time period does not have
There is the significant operation of progress, it is therefore desirable to further pass through its state of mind of sight data analysis.As shown in Figure 8, the driving
The notice status index C of member is always to be higher than threshold value Tc, therefore judge that the driver notices compromising feature region, because
This judges that its is in good condition, and category operates or had appreciated that danger consciously, so that system will shielding within the period
Alarm device.For second place driver, because it 2.2 second moment took significant operation, therefore according to formula(4)
Can the driver be that the state of mind is good within the period between 2.2 seconds to 8.73 seconds.Therefore within the period institute
It is shielding to have vehicle-mounted active safety system.Then other times section is investigated, it can be found that within 8.73 seconds to 10 second periods
The mental status C of driver is always to be less than threshold value Tc, therefore illustrate the driver within this period, both nonsensical behaviour
Make, not the good state of mind(The presence of current dangerous is not recognized).Therefore, now need vehicle-mounted in device for opening
Active safety system is simultaneously alerted.
In a particular application, in the present invention notice calculating section and comprehensive warning control section can be in embedded meter
Realize, realized on such as system-on-a-chip based on FPGA in calculation machine system.In the present embodiment, it be able to can be divided from Tobbi companies
From formula eye tracker, wherein will be arranged on for the Sensor section for catching pilot's line of vision in vehicle drive room(As shown in Figure 2).
The screen portions of eye tracker are only used in calibration process, for showing that front shot by camera obtains image and display system
Driver's blinkpunkt that system is estimated.
It the above is only the preferred embodiment of the present invention, protection scope of the present invention is not limited merely to above-described embodiment,
All technical schemes belonged under thinking of the present invention belong to protection scope of the present invention.It should be pointed out that for the art
For those of ordinary skill, some improvements and modifications without departing from the principles of the present invention should be regarded as the protection of the present invention
Scope.
Claims (6)
1. a kind of active safety systems of vehicles alarm shield device, including:
Driver attention's detection module, predicts the outcome for comparing human eye sight data with notice, and combines car body signal
Carry out the state of mind of comprehensive descision driver;
Vehicle-mounted active safety system module, for active analysis, judges the operation safe condition of vehicle, and point out driver;
Comprehensive warning control module, for being judged to open or shield vehicle-mounted master according to the testing result of the current driver's state of mind
Dynamic security system modules, or actively alerted to driver;Characterized in that,
Driver attention's detection module includes forward looking camera, eye tracker and computer, and the forward looking camera is installed
In the windshield of vehicle, for shooting the frontal scene of vehicle and sending to computer;The eye tracker is arranged on car
Driving panel on, pay close attention to information for gathering the eyeball of driver and send to computer;The car body signal directly from
Car body leads to computer;
Driver attention's detection module is judged driver's state of mind by two aspects:One is by analysis
Driver attention's distribution of actual attention sight and the computation model prediction of driver, to judge the state of mind of driver;
Two be by detecting whether driver carried out significant operation in a short time, to judge the state of mind of driver.
2. active safety systems of vehicles alarm shield device according to claim 1, it is characterised in that the vehicle-mounted active
Security system modules are the one or more in deviation alarm module, forward direction collision alarm module or pedestrian protection system.
3. active safety systems of vehicles alarm shield device according to claim 1, it is characterised in that described to pass through analysis
The driver attention of actual attention sight and the computation model prediction of driver is distributed to judge the step of driver's state of mind
It is rapid as follows:
(1):It is distributed according to the attention of the image prediction driver currently gathered;
(2):Count the feelings of probability density of the blinkpunkt residing in prediction notice distribution map in l blinkpunkt sampling period
Condition, judges whether driver is invested in the related region of traffic events by notice according to these situations, if can be timely
It was found that accident.
4. active safety systems of vehicles alarm shield device according to claim 3, it is characterised in that the step (1)
Detailed process be:
(1.1) present image I, is gathered first, while notice desired distribution figure A (x)=0 of initialization one complete zero, described
Complete zero notice desired distribution figure is of the same size with original image;Then using traditional computer vision methods to figure
Scene as in is analyzed, and enters driveway line, vehicle, the detection and tracking of pedestrian;Lane line is detected to determine and be partitioned into
Road area lane, it is then determined that go out deviation danger relevant range R_lane, nearest apart from vehicle front-wheel line midpoint
Lane line region;Detect and be partitioned into front vehicles region vehicle, it is then determined that going out front vehicles risk of collision relevant range
Vehicle region in R_vehicle, this track of front nearest apart from vehicle front-wheel line midpoint;Detect and be partitioned into front row
People region pdstn, it is then determined that going out front pedestrian impact danger relevant range R_pdstn, apart from vehicle front-wheel line midpoint most
The pedestrian area near front;
(1.2), calculated for the attention distribution probability at image x according to following formula:
A (x)=max (e-d(x,R_lane),e-d(x,R_vehicle),e-d(x,pdstn)) (1)
Wherein d (x, R) represents point x to region R Euclidean distance, if x ∈ R, d (x, R)=0, otherwise d (x, R)=min { d (x
∈R)};
(1.3), there is prediction driver attention's distribution map, you can sentence according to the distribution situation of driver's blinkpunkt of collection
The current notice and response situation of disconnected driver;The empty array B that a length is l is initialized first;Adopted in new blinkpunkt
Before collection, all elements in array move to left one, vacate most right B (l);After current fixation point is collected, according to live regard
Point position x, is last assignment of array, order:
B (l)=A (x)=max (e-d(x,R_lane),e-d(x,R_vehicle),e-d(x,pdstn)) (2)
Then the index of the respond of current driver's is calculated:
<mrow>
<mi>C</mi>
<mo>=</mo>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>l</mi>
</munderover>
<msup>
<mi>e</mi>
<mrow>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>-</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
<mo>*</mo>
<mn>0.1</mn>
</mrow>
</msup>
<mi>B</mi>
<mrow>
<mo>(</mo>
<mi>l</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
If C is more than certain threshold value Tc, then can be determined that the respond of driver is good.
5. active safety systems of vehicles alarm shield device according to claim 1, it is characterised in that described to pass through detection
Whether driver carried out significant operation to judge the state of mind system of driver in a short time, was specially:First
Detect last meaningful time of the operation away from current time;Four timers Tt, Tb, Tr, Tw are initialized, successively to spinning
Detected to lamp, brake, wiper, action of blowing a whistle;If detecting the end signal acted accordingly, it will count accordingly
When device reset, otherwise, the normal timing of timer;In each detection cycle, the state of mind of driver is entered using following formula
Row is assessed:
D=max (e-Tt,e-Tb,e-Tr,e-Tw) (4)
If D is more than certain threshold value Td this moment, illustrate that driver has done significant action at no distant date, therefore be clear this moment
Awake, Td is to treat training parameter.
6. active safety systems of vehicles alarm shield device according to claim 2, it is characterised in that further to driving
Member's notice detection module carries out initial alignment, comprises the following steps that:
(1):Under dead ship condition, some objects are put in front of forward looking camera at certain position on vehicle, with these objects
Reference-calibrating is used as with other distant objects intrinsic in scene;
(2):Under dead ship condition, support staff passes through some specified being presented on the screen of forward looking camera content of shooting
Object, and guide driver to be watched attentively, the direction of visual lines of the driver that now record eye tracker is recorded this moment;
(3):The position of object is noted by changing or object is noted in itself, repeated several times step (2);
(4):Calibration result is verified;Now guiding driver notes objects different in scene, and eye tracker is estimated
Pilot's line of vision be shown in forward looking camera shooting realtime graphic on, observe and record these objects and the sight detected
Whether position overlaps or approaches;Required if error is met, terminate initial alignment process and terminate;Otherwise, repeat step (2)
To (3), start to re-scale.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410017718.1A CN103770780B (en) | 2014-01-15 | 2014-01-15 | A kind of active safety systems of vehicles alarm shield device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410017718.1A CN103770780B (en) | 2014-01-15 | 2014-01-15 | A kind of active safety systems of vehicles alarm shield device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103770780A CN103770780A (en) | 2014-05-07 |
CN103770780B true CN103770780B (en) | 2017-10-03 |
Family
ID=50563676
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410017718.1A Active CN103770780B (en) | 2014-01-15 | 2014-01-15 | A kind of active safety systems of vehicles alarm shield device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103770780B (en) |
Families Citing this family (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105270258A (en) * | 2014-06-12 | 2016-01-27 | 陕西重型汽车有限公司 | Automotive steering safety assistance system |
CN104217615B (en) * | 2014-09-16 | 2016-08-24 | 武汉理工大学 | A kind of pedestrian anti-collision system and method collaborative based on bus or train route |
CN105788364B (en) * | 2014-12-25 | 2019-05-14 | 中国移动通信集团公司 | A kind of dissemination method and device of warning information |
CN104636723B (en) * | 2015-01-29 | 2018-05-25 | 何灌昌 | Driving method for early warning based on certain objects image analysing computer in video image |
JP6563798B2 (en) * | 2015-12-17 | 2019-08-21 | 大学共同利用機関法人自然科学研究機構 | Visual recognition support system and visual object detection system |
JP6477529B2 (en) * | 2016-02-02 | 2019-03-06 | 株式会社デンソー | Driving support device |
CN108021849B (en) * | 2016-11-03 | 2022-04-05 | 杭州海康威视数字技术股份有限公司 | Pedestrian early warning method and device |
CN106781822B (en) * | 2016-12-03 | 2019-07-05 | 西安科锐盛创新科技有限公司 | The system and method for improving drive safety |
KR20180070401A (en) * | 2016-12-16 | 2018-06-26 | 현대자동차주식회사 | System and Method for determining override of self-driving vehicle |
CN108725454B (en) * | 2017-04-25 | 2021-08-24 | 奥迪股份公司 | Safe driving assistance system and control method thereof |
CN107169465A (en) * | 2017-05-24 | 2017-09-15 | 成都志博科技有限公司 | The decision method of driver fatigue |
TWI653170B (en) | 2017-11-01 | 2019-03-11 | 宏碁股份有限公司 | Driving notification method and driving notification system |
CN107928685A (en) * | 2017-12-13 | 2018-04-20 | 吉林大学 | A kind of response analytical equipment and method based on driving human eye dynamic characteristic |
CN108099901B (en) * | 2017-12-18 | 2019-07-09 | 长春工业大学 | A kind of method of active front wheel steering control precision under raising limiting condition |
CN109941278B (en) * | 2017-12-20 | 2021-06-01 | 广州汽车集团股份有限公司 | Vehicle control method, device and system based on driver state monitoring |
CN110139205B (en) * | 2018-02-09 | 2021-11-02 | 驭势(上海)汽车科技有限公司 | Method and device for auxiliary information presentation |
CN108609013B (en) * | 2018-05-01 | 2020-09-01 | 南京溧水高新创业投资管理有限公司 | Active safety system for vehicle |
EP3575175B1 (en) * | 2018-06-01 | 2024-03-13 | Zenuity AB | Method and system for assisting drivers to drive with precaution |
TWI719330B (en) * | 2018-07-27 | 2021-02-21 | 群邁通訊股份有限公司 | Electronic device and driving safety reminding method |
CN109367540A (en) * | 2018-12-03 | 2019-02-22 | 南京云计趟信息技术有限公司 | A kind of driver's response identifying system based on Risk-warning |
CN111292548B (en) * | 2020-02-06 | 2021-02-05 | 温州大学 | Safe driving method based on visual attention |
CN113044048B (en) * | 2021-03-31 | 2023-03-28 | 东风商用车有限公司 | Method, device and equipment for identifying vehicle deviation and readable storage medium |
CN117382644B (en) * | 2023-12-11 | 2024-02-27 | 安徽蔚来智驾科技有限公司 | Distraction driving detection method, computer device, storage medium and intelligent device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101523455A (en) * | 2006-10-13 | 2009-09-02 | 丰田自动车株式会社 | On-board warning apparatus and warning method |
CN103442925A (en) * | 2011-03-25 | 2013-12-11 | Tk控股公司 | System and method for determining driver alertness |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004108466A1 (en) * | 2003-06-06 | 2004-12-16 | Volvo Technology Corporation | Method and arrangement for controlling vehicular subsystems based on interpreted driver activity |
US8384534B2 (en) * | 2010-01-14 | 2013-02-26 | Toyota Motor Engineering & Manufacturing North America, Inc. | Combining driver and environment sensing for vehicular safety systems |
-
2014
- 2014-01-15 CN CN201410017718.1A patent/CN103770780B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101523455A (en) * | 2006-10-13 | 2009-09-02 | 丰田自动车株式会社 | On-board warning apparatus and warning method |
CN103442925A (en) * | 2011-03-25 | 2013-12-11 | Tk控股公司 | System and method for determining driver alertness |
Also Published As
Publication number | Publication date |
---|---|
CN103770780A (en) | 2014-05-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103770780B (en) | A kind of active safety systems of vehicles alarm shield device | |
CN103770733B (en) | Method and device for detecting safety driving states of driver | |
CN103767715B (en) | A kind of driver safety driving state detection device | |
CN110264783B (en) | Vehicle anti-collision early warning system and method based on vehicle-road cooperation | |
CN106251701B (en) | Vehicle rearview monitor and alarm system and method based on rotation zoom multi-cam | |
CN105015548B (en) | A kind of longitudinal collision avoidance is reminded and follows integrated system and method automatically | |
CN105459907B (en) | Vehicle security system, blind area monitoring method and device | |
CN112180605B (en) | Auxiliary driving system based on augmented reality | |
CN112224211A (en) | Driving simulation system based on multi-autonomous-body traffic flow | |
CN110316198A (en) | A kind of safe-guard system and operation method for highway speed-raising | |
CN109466488A (en) | A kind of vehicle collision automatic help alarm system | |
CN101540090A (en) | Driver fatigue monitoring device based on multivariate information fusion and monitoring method thereof | |
CN106373461A (en) | Driver training management service system and method | |
CN115877343B (en) | Man-car matching method and device based on radar target tracking and electronic equipment | |
CN107303840A (en) | Tank truck with active safety pre-alarming control system | |
CN109747537A (en) | A kind of unmanned early warning system of automobile | |
CN207345852U (en) | A kind of train Safety of Shunting Operation prevention and control system | |
CN206961331U (en) | A kind of high threat vehicle monitoring early warning system | |
Wu et al. | A novel skateboarder-related near-crash identification method with roadside LiDAR data | |
CN112466159B (en) | Right-turning safety early warning system for large vehicle | |
CN112906664B (en) | Neural network-based method and system for detecting abnormal events in tunnel | |
CN113435404A (en) | Safe auxiliary driving method and system for shunting electric bus based on radar and image detection | |
CN113562005A (en) | Intelligent monitoring and obstacle intrusion limit detecting system for driver behaviors of rail train | |
CN116543542B (en) | Tunnel entrance operation risk directional early warning equipment layout method | |
CN205608963U (en) | Vehicle buffer stop based on machine vision |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |