CN103448722B - Auxiliary control system method and apparatus are kept using the method and apparatus and track in the identification of the weighted value such as not track - Google Patents
Auxiliary control system method and apparatus are kept using the method and apparatus and track in the identification of the weighted value such as not track Download PDFInfo
- Publication number
- CN103448722B CN103448722B CN201210524390.3A CN201210524390A CN103448722B CN 103448722 B CN103448722 B CN 103448722B CN 201210524390 A CN201210524390 A CN 201210524390A CN 103448722 B CN103448722 B CN 103448722B
- Authority
- CN
- China
- Prior art keywords
- track
- weighted value
- identification
- lane
- filter
- 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
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000008859 change Effects 0.000 claims abstract description 18
- 230000007274 generation of a signal involved in cell-cell signaling Effects 0.000 claims abstract description 3
- 238000012937 correction Methods 0.000 claims description 22
- 101100381996 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) BRO1 gene Proteins 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 2
- 230000006641 stabilisation Effects 0.000 description 2
- 238000011105 stabilization Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 239000013589 supplement Substances 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
- B60W40/02—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 ambient conditions
- B60W40/06—Road conditions
-
- 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
- B60W30/10—Path keeping
- B60W30/12—Lane keeping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- 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
- B60W2050/0001—Details of the control system
- B60W2050/0019—Control system elements or transfer functions
- B60W2050/0022—Gains, weighting coefficients or weighting functions
-
- 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
- B60W2050/0001—Details of the control system
- B60W2050/0043—Signal treatments, identification of variables or parameters, parameter estimation or state estimation
- B60W2050/0052—Filtering, filters
- B60W2050/0054—Cut-off filters, retarders, delaying means, dead zones, threshold values or cut-off frequency
- B60W2050/0056—Low-pass filters
-
- 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
- B60W2552/30—Road curve radius
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Y—INDEXING SCHEME RELATING TO ASPECTS CROSS-CUTTING VEHICLE TECHNOLOGY
- B60Y2300/00—Purposes or special features of road vehicle drive control systems
- B60Y2300/10—Path keeping
- B60Y2300/12—Lane keeping
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
Not etc. the method and apparatus in weighted value identification track and track using the method and device do not keep auxiliary control system method and apparatus using the weighted value such as not to improve the application of lane identification rate in the method that the present invention proposes using multiple filter identification track.Application of the invention not etc. weighted value identification track device be using the weighted value identification track device such as not from the lane recognition method using multiple filter, including：Image recognizing section, identification left side and right-hand lane calculate the side-play amount and confidence level of the left side and right-hand lane；Identification judging part, judge the identification left side and right-hand lane information in whether there is it is unidentified or misidentify information；Senior filter, unidentified or misrecognition judgement is whether based on described, changes several low pass filters（LPF, Low Pass Fillter）Weighted value, using by application it is described change weighted value the low pass filter signal generation correct track.
Description
Technical field
The present invention relates to the application method and apparatus in weighted value identification track and using the method and the track of device such as not
Keep auxiliary control system method and apparatus, specifically using in the lane recognition method of multiple filter using the weighted value such as not
Not etc. the lane recognition method and device of weighted value and the track using the method and device are not protected to improve the application of lane identification rate
Hold auxiliary control system method and apparatus.
Background technology
Track keeps accessory system（LKAS:Lane Keeping Assistance System）It is to be recognized using camera
The technology of auto-steering is implemented in track, the image procossing measurement lane width based on camera, the lateral direction of car position on track,
With a distance from the track of both sides and track form, road curvature radius, car is controlled using the vehicle location and road information of the acquisition
.
The control performance of the LKAS performances is heavily dependent on the lane information confidence level obtained from camera.But
It is that track major part on Ordinary Rd is all dotted line rather than solid line, because of railing or central shunting area, railing shadow etc. easily
Cause unidentified and misrecognition phenomenon.
Straight line is not only on actual highway, also many curved sections, even straightway, because the track on road surface is tinted shape
The pavement state such as state and rainy day, it is impossible to which the situation of stabilization received image signal also happens occasionally.
KR published patent application No. 10-2009-53412 and No. 10-2010-34409, United States Patent (USP) 7,532,981, U.S.
The patents such as state's Patent Application Publication 2010/0076684 are the conventional arts for being related to recognize track method.According to conventional art
Using the filtering technique in lane recognition method, Performance comparision stabilization is played at unidentified track, but it is relative during misrecognition
It is weak.For the conventional art for lane identification using image processing techniques, need to by the offering question of area-of-interest and
Multiple stages of image procossing and its operand is relative increases.And, to constitute multiple filter in the way of strengthen to track
Unidentified or misrecognition, even if then identification side track can also improve the filtering level in both sides track, thus causes track to be known
The problem that rate does not decline.
The content of the invention
To solve described problem, in the method it is an object of the invention to provide a kind of utilization multiple filter identification track
Using the application method that not etc. weighted value does not recognize track of the lifting lane identification rate of the weighted value such as not.
It is to solve described problem, knows using the weighted value such as not in the lane recognition method of utilization multiple filter of the invention
Other track device, including：Image recognizing section, identification left side and right-hand lane calculate the left side and right-hand lane side-play amount and
Confidence level；Judge identification part, judge the identification left side and right-hand lane information in whether there is it is unidentified or misrecognition information；
Senior filter, unidentified or misrecognition judgement is whether based on described, changes several low pass filters（LPF, Low Pass
Fillter）Weighted value, using by application it is described change weighted value the low pass filter signal generation correct car
Road.
Preferably, the senior filter is the width in track described in the belief update using the calculating, using described
The lane width of renewal generates the track of the correction.
Preferably, the low pass filter has four, and two low pass filters are described corrections in four low pass filters
The left-hand lane of identification, remaining two low pass filter are the right-hand lanes for correcting the identification.
Preferably, the weighted value becomes and is even more, judge the identification left side and right-hand lane at least some be not
During identification, the wherein weighted value of some is increased from two low pass filters for being connected to the unidentified track.
Preferably, the weighted value becomes and is even more, judge the identification left side and right-hand lane at least some be mistake
During identification, the wherein weighted value of some is increased from two low pass filters for being connected to the misrecognition track.
Preferably, the senior filter is to generate the correction track using following [mathematical expression 1] and [mathematical expression 2].
[mathematical expression 1]
[mathematical expression 2]
Wherein, W is minimum lane width, and D is maximum unidentified command range, and V is speed, and t is unidentified or misrecognition
Elapsed time, LPF1 is of the wave filter related to right-hand lane identification, and LPF2 is the filter related to right-hand lane identification
One in ripple device, LPF3 is in the wave filter related to left-hand lane identification, and LPF4 is to recognize phase with left-hand lane
One in the wave filter of pass, α is the filter weights related to right-hand lane identification, and β is related to left-hand lane identification
Filter weights.
It is to solve described problem, the present invention is proposed described in applications exploiting using the identification track device identification of the weighted value such as not
Lane information control track keeps accessory system（LKAS,Lane Keeping Assist System）.
It is to solve described problem, the present invention is that the weighted value such as not is applied from the lane recognition method using multiple filter
The method for recognizing track, the implementation steps include：Identification image, identification left side and right-hand lane calculate the left side and right side car
The side-play amount and confidence level in road；Judge identification, from the left side of the identification and right-hand lane information, judge whether not knowing
Other or misrecognition information；Senior filter, based on it is described whether unidentified or misrecognition judgement, change several low pass filters
（LPF, Low Pass Filter）Weighted value, using by application it is described change weighted value the low pass filter letter
The track of breath generation correction.
Preferably, the senior filter step is, using the calculating belief update described in lane width, using institute
The lane width for stating renewal generates the track of the correction
Preferably, the low pass filter has four, and two low pass filters are described corrections in four low pass filters
The left-hand lane of identification, remaining two low pass filter are the right-hand lanes for correcting the identification.
Preferably, the weighted value becomes and is even more, judge in the unidentified left side and right-hand lane at least some
For it is unidentified when, increase the wherein weighted value of some from two low filters for being connected to the unidentified track.
Preferably, the weighted value becomes and is even more, judge the identification left side and right-hand lane at least some be mistake
During identification, the wherein weighted value of some is increased from two low pass filters for being connected to the misrecognition track.
Preferably, the senior filter is the track that the correction is generated using following [mathematical expression 1] and [mathematical expression 2].
[mathematical expression 1]
[mathematical expression 2]
Wherein, W is minimum lane width, and D is maximum unidentified command range, and V is speed, and t is unidentified or misrecognition
Elapsed time, LPF1 is in the wave filter related to right-hand lane identification, and LPF2 is related to right-hand lane identification
One in wave filter, LPF3 is in the wave filter related to left-hand lane identification, and LPF4 is recognized with left-hand lane
One in related wave filter, α is the filter weights related to right-hand lane identification, and β is to recognize phase with left-hand lane
The filter weights of pass.
It is to solve described problem, the present invention is, using the weighted value identification track method identification such as not described in applications exploiting
Lane information control track keeps accessory system（LKAS,Lane Keeping Assist System）.
Present invention has the advantage that：
The present invention can strongly tackle the situation in the misrecognition track that may occur after the change of track.
The present invention is to reduce lane width using the function of time, even if unidentified for a long time or misrecognition track can also be high
Degree recognizes track exactly.
The present invention can utilize the reliability information in track to improve the track filtering level of misrecognition, so as to improve track knowledge
Not other rate.
Even if the present invention is the change according to lane width changeably using lane width critical value in vehicle traveling
Lane width is changed still can tackle rapidly.
Brief description of the drawings
Fig. 1 is to illustrate the structure chart using the vehicle lane identifying device of multiple filter；
Fig. 2 is that not etc. weighted value does not recognize the structure chart of track device for the application of the preferred embodiment of the present invention；
Fig. 3 is that the device lane identification judging part in the application of the invention weighted value identification track such as not judges unidentified or mistake
The explanatory diagram of the method for the lane information of identification；
Fig. 4 is the flow chart of the application method in weighted value identification track such as or not the preferred embodiment of the present invention.
In figure：
100：Using the Lane recognition device of the vehicle of multiple filter;
110：Camera left-side signal; 112：Local filter L1;
114：Local filter L2;
120：Camera right-side signal; 122：Local filter R1;
124：Local filter R2;
200：The device in track is recognized using the weighted value such as not;
210：Image recognizing section; 212：Right-hand lane confidence calculation section;
216：Right-hand lane side-play amount calculating part; 214：Left-hand lane confidence calculation section;
218：Left-hand lane side-play amount calculating part;
220：Track correction unit; 222：Lane identification judging part;
222a：The unidentified judging part in track; 222c：Track misidentifies judging part;
224：Low pass filter; 224a:1LPF;
224c：2LPF; 224e：3LPF;
224g：4LPF;
226：Senior filter; 226a：First senior filter portion;
226c：Second senior filter.
Specific embodiment
With reference to the accompanying drawings, invention preferred embodiment is described in detail.Reality in description and accompanying drawing is same
Individual component is represented with same symbol, and explanation is not repeated.And during the explanation present invention, if to relevant known function or
Illustrating for structure makes present inventive concept thicken, and omits the explanation.
Some component " being connected to " or " being connected in " other components are described, it is possible to be directly connected in or be connected in this its
Its component, but be understood to centre and there may exist other components.Conversely, describe some component " being directly connected in " or
" being coupled directly to " other components, are understood to middle in the absence of other components.
Singulative in this specification, on the premise of not having special suggestion in sentence, also comprising plural form.Explanation
Used in book " including（comprises）" or " including (comprising) " be not excluded for involved component, step, dynamic
More than one other components, step, action and/or the presence of element or supplement beyond work and/or element.
Fig. 1 is to illustrate the structure chart using the vehicle lane identifying device of multiple filter.As shown in figure 1, using multiple
The Lane recognition device 100 of the vehicle of wave filter is distinguished for the left-side signal 110 of camera and the right-side signal 120 of camera
Use two local filters.That is, local filter L1112 and local filter are used to camera right-side signal 110
L2114, local filter R1122 and local filter R2124 is used to camera right-side signal 120.But use wave filter
Quantity is not limited to two, can adjust quantity according to controlled level.Embodiments of the invention are to illustrate that facility is defined to 2
It is individual.
Each local filter is made up of the Kalman filter of following mathematical expression 1.
【Mathematical expression 1】
In mathematical expression 1, P is system covariance, and Q and R represents process noise covariance and measurement noise covariance, K respectively
It is the kalman gain calculated using covariance.Signal is sent to left side senior filter 116 and right side by each local filter
Senior filter 126.Obtained according to the signal difference change entered by camera and the weighted value of time in penalty factor block 130
After send each senior filter 116,126 to.
In penalty factor block weighted value is obtained according to following [mathematical expression 2].
【Mathematical expression 2】
W refers to minimum lane width in mathematical expression 2, and D refers to maximum unidentified command range, and V refers to speed, and t refers to not
In the elapsed time of identification/misrecognition, α refers to filtration combined weighted value.
Then utilized from the value of local filter acquisition and from the acquisition of penalty factor block with letter in senior filter 116,126
Number the change of divergence and the weighted value of time, obtain final track estimated value.
Fig. 2 is that not etc. weighted value does not recognize the structure drawing of device in track for the application of the preferred embodiment of the present invention.Add using not waiting
The device 200 in weights identification track includes Image recognizing section 210 and track correction unit 220.Image recognizing section 210 is identification vehicle
Left-hand lane and right then track, calculate the lane shift amount and confidence level of identification.
Image recognizing section 210 includes the shoot part such as camera, shoots the front of vehicle in shoot part and recognizes track, and
Receive the image recognition track of the capture apparatus photographs by imaging first-class vehicle.
Image recognizing section 210 includes right-hand lane confidence calculation section 212, right-hand lane side-play amount calculating part 216, left side
Track confidence calculation section 214 and left-hand lane side-play amount calculating part 218.Side-play amount is track distance.
Right-hand lane side-play amount calculating part 216 is the side-play amount for calculating vehicle right side.
Left-hand lane side-play amount calculating part 218 is the side-play amount for calculating vehicle left side.
Right-hand lane confidence calculation section 212 is the confidence level degree of the lane information for calculating vehicle right side.
Left-hand lane confidence calculation section 214 is the information credibility information for calculating vehicle left side track.Because vehicle is got dirty
Contaminate, thicken and unidentified or misrecognition track, then Image recognizing section 210 possibly cannot accurately calculate side-play amount and credible
Degree.
For example, when track is unrecognized, the unidentified lane shift amount information and confidence level for calculating may be inaccurate, by mistake
During identification track, the track confidence level of the misrecognition for calculating is higher, and offset information is then inaccurate.
Lane information portion 220 is to judge whether there is unidentified or misrecognition in the lane information that Image recognizing section 210 is recognized
Information, according to the judgement several low pass filters of change（LPF,Low Pass Filter）Weighted value.Track correction unit 220 should
The track of correction is generated with the weighted value of change.Specifically, track correction unit 220 includes lane identification judging part 222, low pass filtered
Ripple device 224, senior filter 226.Lane identification judging part 222 is to judge whether there is in the lane information that Image recognizing section 210 is recognized
Unidentified or misrecognition lane information.
Fig. 3 be the application of the invention weighted value identification track device such as not lane identification judging part judge it is unidentified or
The explanatory diagram of the method for the lane information of misrecognition.Specifically, lane identification judging part 222 includes the unidentified judging part in track
222a and track misidentify judging part 222c.With reference to Fig. 3, judging part 222a unidentified to track and track misrecognition judge
Portion 222c is illustrated.
If the unidentified judging part 222a in track is the with a low credibility in restriction level of the calculating of Image recognizing section 210, will
The lane information of the identification is judged as that track is misidentified（S305 steps and S310 steps）.
Track misrecognition judging part is the part that 222c is misidentified in the track that Image recognizing section 210 is recognized.Specifically
It is that track misrecognition judging part 222c knows in real time when being and being reached more than certain level by 210 pairs of confidence levels of calculating of Image recognizing section
Other lane width and before lane width are implemented to compare（S340 steps）.Lane width is left-hand lane and right-hand lane before
Confidence level keep within a certain period of time when more than certain level, extract and take out from the lane information of the identification of Image recognizing section 210
Sample data calculate average value（S315 steps and S320 steps）.And update lane width information using the average value for calculating（S325
Step）.If the confidence level of left side and right-hand lane does not keep more than certain level within a certain period of time, former track is kept
Width information.That is, according in S315 steps distinguish situation, before lane width can turn into update lane width or
Person's original lane width information.
The lane width difference that the lane width of Real time identification compares with lane width before is less than critical value, then judge car
Road is normally recognized（S355 steps）.
The lane width difference that the lane width of Real time identification compares with lane width before is more than critical value, then judging should
Track is misidentified（S350 steps）.
That is, track misrecognition judging part 222c is the left-hand lane and right-hand lane letter that Image recognizing section 210 is calculated
The confidence level of breath is reached when more than certain level, using the left-hand lane and the side-play amount of right-hand lane of right-hand lane, periodically
Storage lane width information, left-hand lane width that each moment will be recognized by Image recognizing section 210 and right-hand lane width with
Periodically the lane width of storage is compared.The difference of the lane width for comparing is considered as both sides car below certain critical value, then
Road is normal, when the difference of the lane width for comparing is more than certain critical value, judges that this is identified by track misrecognition judging part
Information be misrecognition.
Low pass filter portion 224 is made up of several low pass filters.For example, low pass filter portion 224 can be by 4 low passes
Wave filter group into.The low pass filter portion 224 being made up of 4 low pass filters is illustrated in Fig. 2, but is not limited to 4.
Low pass filter portion 224 is the lane identification information for receiving the identification of Image recognizing section 210, is passed to LPF
Device and remove noise etc..
Low pass filter portion 224 utilizes when LPF is passed through from the lane information of the identification of Image recognizing section 210
Unidentified or misrecognition the lane identification information that lane identification judging part 222 judges, changes unidentified or misrecognition part
The weighted value of low pass filter that passes through of track after be run through.
For example, low pass filter portion 224 judges that right-hand lane is unrecognized in lane identification judging part 222, then by signal
The weighted value α for postponing big but stronger to noise 1LPF224a increases.Conversely, the side-play amount of left-hand lane and confidence level keep
Higher level and be information directly using identification, the weighted value β of 3LPF224e is to maintain basic value.
Another example is that low pass filter portion 224 is to judge that right-hand lane is misidentified in lane identification judging part 222
When, the weighted value α of larger but stronger to the noise 1LPF224a of signal delay is increased.Lane identification judging part 222 judges
When left-hand lane and right-hand lane are misidentified, α and β is all increased.
Senior filter portion 226 generates track using the information in the low pass filter portion by application weighting value α and β.
If the track of the side-play amount that calculates of Image recognizing section 210 and normal identification with a high credibility, then senior filter
Portion 226 generates the track identical track with identification, when lane identification judging part 222 misidentifies or be unidentified, using application
The information of the low pass filter of weighted value α and β generates virtual track.
Senior filter portion 226 includes the second of the first senior filter portion 226a of generation right-hand lane and generation left-hand lane
Senior filter 226c.
Specifically, the first senior filter 226a is to generate track using following mathematical expression 3.
【Mathematical expression 3】
Specifically, the first senior filter 226c is to generate track using following mathematical expression 4.
【Mathematical expression 4】
The weighted value α and β that mathematical expression 3 is applied to mathematical expression 4 are identical basic values, but as it was previously stated, according to track
Unidentified or misrecognition situation, weighted value α and β may be different, and the ground application such as not.
W is minimum lane width in mathematical expression 3 and mathematics 4, and D is maximum unidentified command range, and V is speed, and t is not know
In other or misrecognition elapsed time, α refers to filter weights.
From the point of view of mathematical expression 3 and mathematical expression 4, senior filter portion 226 is by using left side and right side difference application weighting value
Information and Image recognizing section 210 calculate reliability information update lane width, it is unidentified or misrecognition the elapsed time
Utilize together and the track of generation more precise calibration.
LPF1 is to represent 1LPF224a, and LPF2 represents that 2LPF224c, LPF3 represent that 3LPF224e, LPF4 represent
4LPF224g, LPF1 and LPF2 are same low pass filters, but concrete property has difference.Equally, LPF3 and LPF4 are also
Same low pass filter, but concrete property can be different.
Not etc. weighted value identification track device 200 generation lane information can be more not critically for applications exploiting application of the invention
Control track keeps accessory system（LKAS,Lane Keeping Assist System）.The control can be kept by track
Auxiliary control system portion（It is not shown）Perform.
Not etc. weighted value identification track device 200 is not the misrecognition car that may occur after track is changed for application of the invention
Still its performance is played consistently under road state.
Not etc. weighted value identification track device 200 is not to reduce lane width by the function of time for application of the invention, therefore
Even if unidentified for a long time or misrecognition, still recognizes track exactly.
Not etc. weighted value identification track device 200 is not only to lift misrecognition using track reliability information for application of the invention
Track wave filter level and lift lane identification rate.
Not etc. weighted value identification track device 200 is not changeably to apply car according to the change of lane width for application of the invention
Road width critical value, even if lane width change in vehicle traveling, still rapid reply.
Fig. 4 is the flow chart of the application method in weighted value identification track such as or not the preferred embodiment of the present invention.
According to Fig. 4, Image recognizing section 210 identification left-hand lane and right-hand lane（S410 steps）, calculate the track of identification
Information offset amount and confidence level（S420 steps）.
Lane identification judging part 222 judges track by unidentified or misrecognition using the side-play amount and reliability information that calculate
Part（S430）.
Low pass filter portion 224 is, according to unidentified or misrecognition situation, to change the weighted value of low pass filter（S440
Step）.
Senior filter portion 226 is the track that correction is generated using the information of the low pass filter by application change weighted value
Information（S450 steps）.
Using application of the invention, not etc. the method generation lane information in weighted value identification track not can be used for track and keep
The control of accessory system.
Track keeps auxiliary control system portion to keep accessory system using the lane information control track of generation（S460 is walked
Suddenly）.
Not etc. the structure chart of the vehicle identifier of weighted value is not understood to represent for the application of the preferred embodiment of the present invention
Inventive principle is set to become the notional viewpoint of illustration definitely.Similar, all flow charts are understood to represent real
Shown on the medium that border can be read in the computer, and whether there is the clearly diagram of computer or processor but be all by electricity
The various treatment progress that brain or processor are implemented.
Including the work(with the various devices illustrated on the figure including processor or similar representation of concept functional block
The hardware that specialized hardware can be available with or even Suitable software ability can be run is provided.It is described when being provided using processor
Function can be provided by single application specific processor, single shared processor or multiple Respective processors, and wherein one
Dividing to share.
And can not be construed as clearly using processor, control or with the term of similar representation of concept exclusive
Property ground quote with runs software ability hardware, be understood to infer unrestrictedly comprising digital signal processor（DSP）
ROM, RAM and non-voltile memory of hardware, storage software.Known other usual hardware can also be included.
Above example and accompanying drawing are merely illustrative of the technical solution of the present invention, rather than its limitations；Although with reference to foregoing
Embodiment has been described in detail to the present invention, it will be understood by those within the art that：It still can be to foregoing
Technical scheme described in each embodiment is modified, or carries out equivalent to which part technical characteristic；And these are changed
Or replace, do not make the scope of technical scheme described in the essence disengaging various embodiments of the present invention of appropriate technical solution.The present invention
Protection domain should be explained according to following rights, and all technical schemes in its equivalents should
Belong to scope of the presently claimed invention.
Claims (14)
1. a kind of device for applying the weighted value such as not to recognize track, it is characterised in that know as the track using multiple filter
The device in track is recognized in other device using the weighted value such as not, including：
Image recognizing section, identification left side and right-hand lane, calculate the side-play amount and confidence level of the left side and right-hand lane；
Identification judging part, judges whether there is unidentified or misrecognition information in recognized left side and right-hand lane information；
Senior filter, based on unidentified or misrecognition the judgement is whether there is, changes the weighted value of several low pass filters, profit
The track corrected with the signal generation of the low pass filter by the changed weighted value of application.
2. not etc. weighted value does not recognize the device in track for application according to claim 1, it is characterised in that
The senior filter is the width for utilizing track described in calculated belief update, utilizes updated lane width
Generate the track of the correction.
3. not etc. weighted value does not recognize the device in track for application according to claim 1, it is characterised in that
The low pass filter has four, and two low pass filters are the recognized left side cars of correction in four low pass filters
Road, remaining two low pass filter are the recognized right-hand lanes of correction.
4. not etc. weighted value does not recognize the device in track for application according to claim 3, it is characterised in that
The change of the weighted value is even more, and judges that at least some is unidentified in the left-hand lane and right-hand lane of the identification
When, increase the wherein weighted value of some from two low pass filters for being connected to unidentified track.
5. not etc. weighted value does not recognize the device in track for application according to claim 3, it is characterised in that
The change of the weighted value is even more, and judges that at least some is misrecognition in the left-hand lane and right-hand lane of the identification
When, increase the wherein weighted value of some from two low pass filters for being connected to the track of misrecognition.
6. not etc. weighted value does not recognize the device in track for application according to claim 3, it is characterised in that
The senior filter is the track that the correction is generated using following [mathematical expression 1] and [mathematical expression 2]；
[mathematical expression 1]
[mathematical expression 2]
Wherein, W is minimum lane width, and D is maximum unidentified command range, and V is speed, and t is unidentified or misrecognition is passed through
Time, LPF1 is of the wave filter related to right-hand lane identification, and LPF2 is the wave filter related to right-hand lane identification
In one, LPF3 is in the wave filter related to left-hand lane identification, and LPF4 is related to left-hand lane identification
One in wave filter, α is the filter weights related to right-hand lane identification, and β is the filter related to left-hand lane identification
Ripple device weighted value.
7. a kind of track keeps auxiliary control system device, it is characterised in that
Accessory system is kept using lane information control track, the lane information is utilized according to any one of claim 1 to 6
The device in the described application weighted value identification track such as not is recognized.
8. it is a kind of to apply the method that not etc. weighted value does not recognize track, it is characterised in that to know as the track using multiple filter
The method for recognizing track in other method using the weighted value such as not, implementation step includes：
Identification image, identification left side and right-hand lane, calculate the side-play amount and confidence level of the left side and right-hand lane；
Judge identification, from the left side and right-hand lane information that are recognized, judge whether unidentified or misrecognition information；
Senior filter, based on whether there is unidentified or misrecognition the judgement, changes the weighted value of several low pass filters,
The track of correction is generated using the information of the low pass filter by the changed weighted value of application.
9. the application method that not etc. weighted value does not recognize track according to claim 8, it is characterised in that
The senior filter step be utilize calculated belief update described in lane width, utilize updated track wide
The track of the degree generation correction.
10. the application method that not etc. weighted value does not recognize track according to claim 8, it is characterised in that
The low pass filter has four, and two low pass filters are the recognized left side cars of correction in four low pass filters
Road, remaining two low pass filter are the recognized right-hand lanes of correction.
The 11. application methods that not etc. weighted value does not recognize track according to claim 10, it is characterised in that
The change of the weighted value is even more, judge in unidentified left-hand lane and right-hand lane at least some for unidentified
When, increase the wherein weighted value of some from two low pass filters for being connected to unidentified track.
The 12. application methods that not etc. weighted value does not recognize track according to claim 10, it is characterised in that
The change of the weighted value is even more, and judges that at least some is misrecognition in the left-hand lane and right-hand lane of the identification
When, increase the wherein weighted value of some from two low pass filters for being connected to the track of misrecognition.
The 13. application methods that not etc. weighted value does not recognize track according to claim 10, it is characterised in that
The senior filter is the track that the correction is generated using following [mathematical expression 1] and [mathematical expression 2]；
[mathematical expression 1]
[mathematical expression 2]
Wherein, W is minimum lane width, and D is maximum unidentified command range, and V is speed, and t is unidentified or misrecognition is passed through
Time, LPF1 is in the wave filter related to right-hand lane identification, and LPF2 is the filtering related to right-hand lane identification
One in device, LPF3 is in the wave filter related to left-hand lane identification, and LPF4 is related to left-hand lane identification
Wave filter in one, α is the filter weights related to right-hand lane identification, and β is related to left-hand lane identification
Filter weights.
A kind of 14. tracks keep auxiliary control system method, it is characterised in that
Accessory system is kept using lane information control track, the lane information is utilized according to any in claim 8 to 13
The method in application described in the weighted value identification track such as not is recognized.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR10-2012-0059627 | 2012-06-04 | ||
| KR1020120059627A KR101882249B1 (en) | 2012-06-04 | 2012-06-04 | Method and Appartus for Recogniting Lane Using Difference Weight, Method and Appartus for Controlling Lane Keeping Assist System Using the Same |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103448722A CN103448722A (en) | 2013-12-18 |
| CN103448722B true CN103448722B (en) | 2017-06-23 |
Family
ID=49731685
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210524390.3A Active CN103448722B (en) | 2012-06-04 | 2012-12-07 | Auxiliary control system method and apparatus are kept using the method and apparatus and track in the identification of the weighted value such as not track |
Country Status (2)
| Country | Link |
|---|---|
| KR (1) | KR101882249B1 (en) |
| CN (1) | CN103448722B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9286679B2 (en) * | 2014-07-22 | 2016-03-15 | GM Global Technology Operations LLC | Misalignment correction and state of health estimation for lane management fusion function |
| KR20210070788A (en) * | 2019-12-05 | 2021-06-15 | 현대모비스 주식회사 | Apparatus for recogniting vehicle width and method |
| CN113386773A (en) * | 2021-07-30 | 2021-09-14 | 蔚来汽车科技(安徽)有限公司 | Method and device for judging reliability of visual identification |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3732108B2 (en) * | 2001-04-12 | 2006-01-05 | 三菱電機株式会社 | Image processing apparatus and image processing method for lane recognition |
| JP4539442B2 (en) * | 2005-06-02 | 2010-09-08 | 日産自動車株式会社 | Leading vehicle detection device |
| JP4434234B2 (en) * | 2007-05-30 | 2010-03-17 | トヨタ自動車株式会社 | VEHICLE IMAGING SYSTEM AND VEHICLE CONTROL DEVICE |
| KR100963795B1 (en) * | 2008-02-28 | 2010-06-24 | 연세대학교 산학협력단 | Apparatus and method for tracking position of object |
| CN101596903B (en) * | 2009-07-07 | 2012-02-15 | 清华大学 | Assisting method for transverse driving of multipurpose automobile and assisting system therefor |
| KR101236234B1 (en) * | 2010-07-29 | 2013-02-22 | 한국과학기술연구원 | Detection system of road line using both laser sensor and camera |
| CN102117546B (en) * | 2011-03-10 | 2013-05-01 | 上海交通大学 | On-vehicle traffic light assisting device |
| KR101805717B1 (en) * | 2011-08-24 | 2017-12-06 | 현대모비스 주식회사 | Method and apparatus for lane detection of vehicle |
-
2012
- 2012-06-04 KR KR1020120059627A patent/KR101882249B1/en active IP Right Grant
- 2012-12-07 CN CN201210524390.3A patent/CN103448722B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| KR101882249B1 (en) | 2018-08-24 |
| CN103448722A (en) | 2013-12-18 |
| KR20130136077A (en) | 2013-12-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kim et al. | Robust lane detection based on convolutional neural network and random sample consensus | |
| Apostoloff et al. | Robust vision based lane tracking using multiple cues and particle filtering | |
| CN110203202B (en) | Lane changing auxiliary early warning method and device based on driver intention recognition | |
| WO2019218861A1 (en) | Method for estimating driving road and driving road estimation system | |
| Low et al. | Simple robust road lane detection algorithm | |
| CN105313897B (en) | The traveling lane identifying system and method for vehicle | |
| CN108133484B (en) | Automatic driving processing method and device based on scene segmentation and computing equipment | |
| CN105263776B (en) | Vehicle control system | |
| EP2741270A1 (en) | Driving assistance apparatus and driving assistance method | |
| CN102951151B (en) | The track of vehicle keeps aid system and method thereof | |
| CN106335509A (en) | Driving assistant for vehicles | |
| CN103448722B (en) | Auxiliary control system method and apparatus are kept using the method and apparatus and track in the identification of the weighted value such as not track | |
| CN109835337B (en) | Turning control method and device and automatic driving vehicle | |
| CN112793576B (en) | Lane change decision method and system based on rule and machine learning fusion | |
| CN105286802A (en) | Driver fatigue detection method based on video information | |
| KR20170085752A (en) | Apparatus and method for traffic lane recognition in automatic steering control of vehilcles | |
| JP2008117073A (en) | Interruption vehicle detection device | |
| CN116118780A (en) | Vehicle obstacle avoidance track planning method, system, vehicle and storage medium | |
| CN102951157B (en) | The radius of curvature evaluation method and its device of vehicle | |
| CN103101535B (en) | LKAS track compensation device and method | |
| CN103077371B (en) | The lane recognition method and its device of vehicle | |
| Yang et al. | Autonomous lane keeping control system based on road lane model using deep convolutional neural networks | |
| Mineta et al. | Development of a lane mark recognition system for a lane keeping assist system | |
| Lee et al. | A cumulative distribution function of edge direction for road-lane detection | |
| CN115571136A (en) | Data processing method and device of lane line, processor and automatic driving system |
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 |