CN108569289A - A kind of trailer-mounted radar and its approach detection and target-recognition method - Google Patents
A kind of trailer-mounted radar and its approach detection and target-recognition method Download PDFInfo
- Publication number
- CN108569289A CN108569289A CN201710358174.9A CN201710358174A CN108569289A CN 108569289 A CN108569289 A CN 108569289A CN 201710358174 A CN201710358174 A CN 201710358174A CN 108569289 A CN108569289 A CN 108569289A
- Authority
- CN
- China
- Prior art keywords
- target
- vehicle
- road
- estimation
- track
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000001514 detection method Methods 0.000 title claims abstract description 27
- 238000013459 approach Methods 0.000 title claims abstract description 25
- 230000033001 locomotion Effects 0.000 claims abstract description 16
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 238000005259 measurement Methods 0.000 claims description 8
- 230000005484 gravity Effects 0.000 claims description 7
- 230000004069 differentiation Effects 0.000 abstract description 9
- 238000009434 installation Methods 0.000 abstract description 5
- 238000004088 simulation Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 7
- 238000011160 research Methods 0.000 description 7
- 238000001914 filtration Methods 0.000 description 6
- 241000272525 Anas platyrhynchos Species 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000007480 spreading Effects 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
- 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
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
-
- 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
-
- 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
- B60W40/072—Curvature of 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
- B60W50/0097—Predicting future 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
- 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
- B60W2050/0028—Mathematical models, e.g. for simulation
- B60W2050/0031—Mathematical model of the vehicle
- B60W2050/0034—Multiple-track, 2D vehicle model, e.g. four-wheel model
-
- 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
-
- 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
- B60W2554/00—Input parameters relating to objects
-
- 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
- B60W2554/00—Input parameters relating to objects
- B60W2554/80—Spatial relation or speed relative to objects
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Radar Systems Or Details Thereof (AREA)
- Traffic Control Systems (AREA)
Abstract
A kind of trailer-mounted radar and its approach detection and target-recognition method, this method and its equipment greatly improve the differentiation rate of vehicle front common-denominator target, when the variation of motion state occurs suddenly for close-in target, the phenomenon that overcoming existing algorithm there are endless loops and its cruise vehicle be easy to causeing are in the hole;General Promotion execution efficiency, real-time is good, when cruise front side vehicle rapidly increases, the differentiation for the target vehicle that can be finished at the appointed time, and then identify the motion state of target vehicle;Equipment volume is small, occupies that interior space is few, and the comfortableness and security of driving are higher;Relevant device is cheap, is suitble to economical vehicle installation.
Description
Technical field
The present invention relates to a kind of trailer-mounted radar and its detection and target-recognition method are approached, belongs to trailer-mounted radar or vehicle mounted electric
Subsystem technology field.
Background technology
The common-denominator target of cruise front side is detected for using in-vehicle electronic system (for example, radar) and is differentiated
The external starting of research is more early, and there are many different research methods.The country detects trailer-mounted radar the key of cruise front side
Than later, but with the continuous development of vehicle electronics technology, especially Radar Technology, the country is right for the differentiation research starting of target
The differentiation research that front common-denominator target is detected in trailer-mounted radar also gradually increases.
Mainly there is the differentiation research of the common-denominator target for front side of cruising in the country:(1) in adaptive cruise control system
Front effective target recognizer research, basic ideas are the prediction rails using track fitting formula difference each target of evaluation fitting
Mark, each target are fitted quality according to it and assign corresponding weights, and the future of this vehicle is then determined by the method for medium filtering
It is expected that driving trace.And then distance of each target vehicle in front apart from this lane center is calculated, whether to differentiate target carriage
In this track, the common-denominator target in front is determined.(2) the self-adaption cruise system target identification method based on trajectory analysis is ground
Study carefully, basic ideas are one section of tracks according to objects ahead vehicle and main vehicle, judge front vehicles and the respective travelings of Ben Che
State, including straight-line travelling, travel in negotiation of bends and lane-change.Then, by Ben Che and the movable information of target carriage to road
Curvature is estimated, and is optimized to road curvature by curvature weighting calculating.And then predict this bus or train route diameter, calculate main vehicle and
Lateral distance between front truck determines the common-denominator target in front.(3) automotive self-adaptive cruise control system effective target recognizes
The research of algorithm, basic ideas are to utilize Kalman filtering and Three Degree Of Freedom car model, are realized to the online of road curvature
Then estimation is identified the common-denominator target for front side of cruising using the front common-denominator target method of discrimination for determining curvature.
Common defects are existing for above-mentioned detection and method of discrimination:(1) the differentiation rate of vehicle front common-denominator target is low, close
When the variation of motion state occurs suddenly for distance objective, algorithm the phenomenon that there are endless loops, it be easy to cause cruise vehicle and is in danger
State;(2) execution efficiency is low, and real-time is poor, when cruise front side vehicle rapidly increases, cannot be finished at the appointed time
The differentiation of target vehicle, and then identify the motion state of target vehicle;(3) relevant device volume is big, and it is big to occupy interior space, drop
The comfortableness and security of low driving;(4) relevant device is expensive, is not suitable for economical vehicle installation.
Invention content
In order to overcome defect existing in the prior art, the present invention provides one kind being based on fuzzy logic and spreading kalman
Filtering approaches the trailer-mounted radar electronic system of detection and target-recognition and its approaches detection and target-recognition method.
A kind of trailer-mounted radar Department of Electronics for approaching detection and target-recognition based on fuzzy logic and Extended Kalman filter
System approaches detection and target-recognition method, it is characterised in that including with step:
S1:The estimation of this car state is carried out based on Extended Kalman filter, to the yaw velocity and side slip angle of this vehicle
Estimated, and then realizes the real-time online estimation of road curvature;
S2:Judge that travelled road is by obtaining the curvature for being presently in road to the estimation of road curvature
Straight way or bend;
S3:For under the different situations of straight way and bend, establishing corresponding common-denominator target discrimination model respectively, to front
Common-denominator target differentiated;
S4:Early warning is carried out according to the approximation ratio of Ben Che and target object, to ensure vehicle driving safety.
Further, the model of the vehicle used when estimating this car state in step S1 is two degrees of freedom car model.
Further, two degrees of freedom car model does not consider the influence of steering, by front wheel angle (steering wheel angle)
As input, vehicle body only does plane motion, and forward speed is constant, and tire cornering characteristics are in the range of linearity, does not consider that air hinders
The influence of power.
Further, common-denominator target is differentiated using fuzzy logic algorithm in step S3, the fuzzy logic algorithm
Need the two-dimensional discrete amount using target object.
A kind of trailer-mounted radar Department of Electronics for approaching detection and target-recognition based on fuzzy logic and Extended Kalman filter
System, it is characterised in that including:
This car state estimation module carries out the estimation of this car state, to the sideway of this vehicle for being based on Extended Kalman filter
Angular speed and side slip angle are estimated, and then realize the real-time online estimation of road curvature;
Road curvature estimation module is judged by obtaining the curvature for being presently in road to the estimation of road curvature
The road travelled is straight way or bend;
Target-recognition module differentiates under the different situations of straight way and bend, establishing corresponding common-denominator target respectively
Model differentiates the common-denominator target in front;
Warning module carries out early warning, to ensure vehicle driving safety according to the approximation ratio of Ben Che and target object.
Further, in this car state estimation module further include two degrees of freedom car model module.
Further, two degrees of freedom car model module does not consider the influence of steering, and by front wheel angle, (steering wheel turns
Angle) as input, vehicle body only does plane motion, and forward speed is constant, and tire cornering characteristics are in the range of linearity, do not consider air
The influence of resistance.
Beneficial technique effect:(1) the differentiation rate for greatly improving vehicle front common-denominator target is sent out suddenly in close-in target
When the variation of raw motion state, the phenomenon that overcoming existing algorithm there are endless loops and its cruise vehicle be easy to causeing are in danger
State;(2) General Promotion execution efficiency, real-time is good, when cruise front side vehicle rapidly increases, can execute at the appointed time
The differentiation of target vehicle is finished, and then identifies the motion state of target vehicle;(3) equipment volume is small, and it is few to occupy interior space, drives
The comfortableness and security sailed are higher;(4) relevant device is cheap, is suitble to economical vehicle installation.
Specific implementation mode
The present invention provides a kind of vehicle for approaching detection and target-recognition based on fuzzy logic and Extended Kalman filter
It carries radar electric system and its approaches detection and target-recognition method.
A kind of trailer-mounted radar Department of Electronics for approaching detection and target-recognition based on fuzzy logic and Extended Kalman filter
System approaches detection and target-recognition method, it is characterised in that including with step:
S1:The estimation of this car state is carried out based on Extended Kalman filter, to the yaw velocity and side slip angle of this vehicle
Estimated, and then realizes the real-time online estimation of road curvature;
S1-1:In conjunction with automobile two-freedom model, application extension Kalman filtering algorithm establishes vehicle state estimation mistake
Journey;The filter state equation and observational equation of vehicle two-freedom model are as follows:
State equation:
Observational equation:
Wherein, ω is yaw velocity, and β is side slip angle, and δ is front wheel angle, w1、w2For system noise, v1、v2For
Sensor measurement noise, w1、w2、v1、v2For mutually independent white noise;
S1-2:The foundation of the state equation and measurement equation of Extended Kalman filter;
State equation:xk+1=f (xk,uk,wk)·················(7)
Observational equation:yk+1=h (xk,vk)··············(8)
State variable:[ω, β]
Control input quantity:[δ]
Both sides export:
S2:Judge that travelled road is by obtaining the curvature for being presently in road to the estimation of road curvature
Straight way or bend;
S2-1:Pass through one or more in steering wheel angle, yaw velocity, side acceleration and each wheel difference in wheel
It is a to calculate road curvature;
S2-2:Judge that road is straight way or bend according to curvature, when curvature value is more than predetermined value, it is believed that current
Road is straight way, when curvature value is less than predetermined value, it is believed that present road is bend;
It is described that road curvature is calculated by steering wheel angle, curvature is calculated using formula (9),Its
Middle δsSteering wheel, i are steering mechanism's speed ratio, dxIt is the wheelspan of front and back wheel;
It is described that road curvature is calculated by side velocity, curvature is calculated using formula (10),Wherein
ayIt is side acceleration, v is car speed;
It is described that curvature is calculated using formula (11) by wheel speed calculation road curvature,Wherein
vrIt is wheel velocity, d is lane width, to avoid the influence of transmission, v and v as possiblerIt is obtained from nonpowered axle, the method need to be estimated
Meter or with sensor measurement lane width;
It is described that road curvature is calculated by yaw velocity, curvature is calculated using formula (12),Its
Middle ω is yaw velocity, and v is car speed;
S3:For under the different situations of straight way and bend, establishing corresponding common-denominator target discrimination model respectively, to front
Common-denominator target differentiated;The case where being bend for present road, step S3-1 is executed, is straight way for present road
Situation executes step S3-2;
S3-1:When judge present road for bend when, using radar installation site as origin o, x-axis be parallel to ground direction before
Side, y-axis are directed toward on the left of driver,
As shown in Figure 1, when Ben Che and target carriage are all when negotiation of bends is turned to the left, target carriage can be obtained by radar
Relative to the distance ds of this vehicle, target carriage relative to this vehicle azimuth (arrive the azimuth of target closest approach line, relative to
Sensor is directed toward) θ obtains diametrically distance of the target carriage with respect to this vehicle;
It can be obtained by geometrical relationship in upper figure:
dsy=dssin θ
(13);
dsx=dscos θ
(14);
Symbol definition is as follows in figure:
ds:Target detection range;θ:Target relative sensors coordinate system azimuth;dsx:Target detection range is in sensor
The component in the directions x in coordinate system;dsy:The component in target detection range directions y in sensor coordinate system;ρ:This vehicle travels rail
The radius of curvature of mark;ρ1:The radius of curvature of target carriage driving trace;ρoffset:Radial distance of the target carriage to this track;
By formula (15) to this track of objects ahead spacing diametrically distance ρoffsetIt is calculated, before can obtaining
Each diametrically distance of the target vehicle away from this vehicle driving trace center line in side;Each target can be judged with respect to this track accordingly
Position relationship, i.e. target are in this track or in adjacent lane;
When target carriage is wide less than half track with respect to the radial distance in this track i.e.:
Wherein, l is that track is wide;When meeting formula (15), then it is assumed that target and this parking stall are latent in identical track
Common-denominator target;
S3-2:Take adjacent lane objects ahead vehicle laterally opposed speed and adjacent lane objects ahead vehicle it is laterally opposed
Input variable of the distance as fuzzy controller, the possibility that target vehicle cuts this track become as the output of fuzzy controller
Amount;
S3-2-1:Blurring;Laterally opposed speed, laterally opposed distance and target carriage cut the language of the probability in this track
Value is respectively:
Relative velocity:It is slow, and in it is slow, it is medium, in it is fast, it is fast };
v:{ SL, MS, MD, MQ, QU };
Relative distance:It is close, and in it is close, it is medium, in it is remote, it is remote };
d:{ NE, MN, MD, MF, FA };
Cut probability:It is low, and in low, medium, middle height, it is high };
P:{ LO, ML, MD, MH, HI };
S3-2-2:Determine membership function;By taking Triangleshape grade of membership function and trapezoidal membership function as an example, degree of membership letter
Several formula are expressed as:
S3-2-3:Establish fuzzy rule base;Empirical rule is derived according to the traveling rule of automobile and driver experience.
If longitudinally opposed distance is remote, longitudinally relative speed is slow, and such case can be described as target carriage road in adjacent lane and travel,
No lane-change cuts the possibility in this track, continues using the nearest vehicle of front fore-and-aft distance in this track as common-denominator target;If longitudinal
Relative distance is remote, and slow in longitudinally relative speed, such case can be described as target carriage and be travelled in adjacent road, no lane-change
Cut the possibility in this track;Continue using the nearest vehicle of front fore-and-aft distance in this track as common-denominator target;If it is longitudinally opposed away from
From being remote, longitudinally relative speed is medium, and such case can be described as target carriage and travel and have certain in adjacent lane
Longitudinal velocity, the possibility gender for cutting this track is relatively low, continues with the nearest vehicle of front fore-and-aft distance in this track to be key
Target;Such reasoning is gone down, and can be obtained 25 control rules, is summarized as follows table
1 fuzzy reasoning table of table
S3-2-4:Fuzzy reasoning;Fuzzy reasoning is exactly to carry out fuzzy logic operation to control system using fuzzy rule.
S3-2-5:Defuzzification;Select gravity model appoach to carry out fuzzy judgment to output quantity, be primarily due to gravity model appoach and can make be
System output is more smooth;After de-fuzzy, system output amount is subjected to section conversion, you can obtain target carriage and cut this track
Probability;
The expression formula of gravity model appoach is:
S3-3:When this track duck vehicle, trailer-mounted radar electronic system is by the Target in front of this track
For common-denominator target;When this track duck vehicle, and there is target vehicle to be moved to this track and p >=0.51 in adjacent lane
When determine common-denominator target by judging this track objects ahead vehicle and cut the size of fore-and-aft distance of the vehicle away from this vehicle again,
Common-denominator target is constant if the target carriage distance in this track is close, becomes to the nearly common-denominator target vehicle of target carriage distance of incision
The vehicle of incision;In front of this track without this vehicle of target vehicle in cruise when, target carriage in adjacent lane is to this vehicle
Road moves, this vehicle is considered that common-denominator target enters tracing mode by cruise if p >=0.51, if p<Then think when 0.51
This vehicle does not cut into this track
S4:Early warning is carried out according to the approximation ratio of Ben Che and target vehicle, to ensure vehicle driving safety.According to radar
Whether the distance between this vehicle and common-denominator target vehicle that electronic system measures and the distance meet early-warning conditions, and decision is always
No unlatching alarm.
S4-1:The distance between this vehicle and common-denominator target vehicle be less than or equal to pre-determined threshold apart from when, open alarm, remind
Driver is slowed down or brake;Otherwise, it is re-executed from step S1.
A kind of trailer-mounted radar Department of Electronics for approaching detection and target-recognition based on fuzzy logic and Extended Kalman filter
System, it is characterised in that including:
This car state estimation module carries out the estimation of this car state, to the sideway of this vehicle for being based on Extended Kalman filter
Angular speed and side slip angle are estimated, and then realize the real-time online estimation of road curvature;Including:Degree of freedom mould
Pattern block and Kalman filtering module, major function are:In conjunction with automobile two-freedom model, application extension Kalman filtering is calculated
Method establishes vehicle state estimation process;
The filter state equation and observational equation of vehicle two-freedom model module are as follows:
State equation:
Observational equation:
Wherein, ω is yaw velocity, and β is side slip angle, and δ is front wheel angle, w1、w2For system noise, v1、v2For
Sensor measurement noise, w1、w2、v1、v2For mutually independent white noise;
The foundation of the state equation and measurement equation of Extended Kalman filter module;
State equation:xk+1=f (xk,uk,wk)··················(7)
Observational equation:yk+1=h (xk,vk)····················(8)
State variable:[ω, β]
Control input quantity:[δ]
Both sides export:
Road curvature estimation module is judged by obtaining the curvature for being presently in road to the estimation of road curvature
The road travelled is straight way or bend;Including curvature estimation module, road condition judgment module;
Curvature estimation module, by steering wheel angle, yaw velocity, side acceleration and each wheel difference in wheel
One or more calculates road curvature;
Road condition judgment module judges that road is straight way or bend, when curvature value is more than predetermined according to curvature
When value, it is believed that present road is straight way, when curvature value is less than predetermined value, it is believed that present road is bend;
It is described that road curvature is calculated by steering wheel angle, curvature is calculated using formula (9),Its
Middle δsSteering wheel, i are steering mechanism's speed ratio, dxIt is the wheelspan of front and back wheel;
It is described that road curvature is calculated by side velocity, curvature is calculated using formula (10),Wherein
ayIt is side acceleration, v is car speed;
It is described that curvature is calculated using formula (11) by wheel speed calculation road curvature,Wherein
vrIt is wheel velocity, d is lane width, to avoid the influence of transmission, v and v as possiblerIt is obtained from nonpowered axle, the method need to be estimated
Meter or with sensor measurement lane width;
It is described that road curvature is calculated by yaw velocity, curvature is calculated using formula (12),Its
Middle ω is yaw velocity, and v is car speed;
Target-recognition module differentiates under the different situations of straight way and bend, establishing corresponding common-denominator target respectively
Model differentiates the common-denominator target in front;
Bend target-recognition module, when judge present road for bend when, using radar installation site as origin o, x-axis is parallel
It being directing forwardly in ground, y-axis is directed toward on the left of driver,
As shown in figure 2 above, when Ben Che and target carriage are all when negotiation of bends is turned to the left, target can be obtained by radar
Vehicle (arrives the azimuth of target closest approach line, relatively relative to the distance ds of this vehicle, target carriage relative to the azimuth with this vehicle
In sensor be directed toward) θ obtain target carriage with respect to this vehicle diametrically distance;
It can be obtained by geometrical relationship in upper figure:
dsy=dssin θ
(13);
dsx=dscos θ
(14);
Symbol definition is as follows in figure:
ds:Target detection range;θ:Target relative sensors coordinate system azimuth;dsx:Target detection range is in sensor
The component in the directions x in coordinate system;dsy:The component in target detection range directions y in sensor coordinate system;ρ:This vehicle travels rail
The radius of curvature of mark;ρ1:The radius of curvature of target carriage driving trace;ρoffset:Radial distance of the target carriage to this track;
By formula (15) to this track of objects ahead spacing diametrically distance ρoffsetIt is calculated, before can obtaining
Each diametrically distance of the target vehicle away from this vehicle driving trace center line in side;Each target can be judged with respect to this track accordingly
Position relationship, i.e. target are in this track or in adjacent lane;
When target carriage is wide less than half track with respect to the radial distance in this track i.e.:
Wherein, l is that track is wide;When meeting formula (15), then it is assumed that target and this parking stall are latent in identical track
Common-denominator target;
Straight way target-recognition module takes the laterally opposed speed and adjacent lane objects ahead of adjacent lane objects ahead vehicle
Input variable of the laterally opposed distance of vehicle as fuzzy controller, target vehicle cut the possibility in this track as Fuzzy Control
The output variable of device processed;Main includes being blurred module, membership function determining module, fuzzy rule base to establish module, obscure
Reasoning module and defuzzification module;
It is blurred module;Laterally opposed speed, laterally opposed distance and target carriage cut the Linguistic Value of the probability in this track
Respectively:
Relative velocity:It is slow, and in it is slow, it is medium, in it is fast, it is fast };
v:{ SL, MS, MD, MQ, QU };
Relative distance:It is close, and in it is close, it is medium, in it is remote, it is remote };
d:{ NE, MN, MD, MF, FA };
Cut probability:It is low, and in low, medium, middle height, it is high };
P:{ LO, ML, MD, MH, HI };
Membership function determining module;By taking Triangleshape grade of membership function and trapezoidal membership function as an example, membership function
Formula be expressed as:
Fuzzy rule base establishes module;Empirical rule is derived according to the traveling rule of automobile and driver experience.If
Longitudinally opposed distance is remote, and longitudinally relative speed is slow, and such case can be described as target carriage road in adjacent lane and travel, nothing
Lane-change cuts the possibility in this track, continues using the nearest vehicle of front fore-and-aft distance in this track as common-denominator target;If longitudinal phase
It is remote to adjust the distance, and slow in longitudinally relative speed, such case can be described as target carriage and be travelled in adjacent road, and no lane-change is cut
Enter the possibility in this track;Continue using the nearest vehicle of front fore-and-aft distance in this track as common-denominator target;If longitudinally opposed distance
It is remote, longitudinally relative speed is medium, and such case can be described as target carriage and be travelled in adjacent lane and have certain indulge
To speed, the possibility gender for cutting this track is relatively low, continues with the nearest vehicle of front fore-and-aft distance in this track to be crucial mesh
Mark;Such reasoning is gone down, and can be obtained 25 control rules, is summarized as follows table 2
Table 2
Fuzzy reasoning module;Fuzzy reasoning is exactly to carry out fuzzy logic operation to control system using fuzzy rule.
Defuzzification module;Gravity model appoach is selected to carry out fuzzy judgment to output quantity, system can be made by being primarily due to gravity model appoach
Output is more smooth;After de-fuzzy, system output amount is subjected to section conversion, you can obtain target carriage and cut the general of this track
Rate;
The expression formula of gravity model appoach is:
Straight way module discrimination module;When this track duck vehicle, trailer-mounted radar electronic system will be before this track
The Target of side is common-denominator target;When this track duck vehicle, and there is target vehicle to this track in adjacent lane
Movement and when p >=0.51 again by judge this track objects ahead vehicle and cut the size of fore-and-aft distance of the vehicle away from this vehicle come
Determine common-denominator target, common-denominator target is constant if the target carriage distance in this track is close, close to the target carriage distance of incision
Common-denominator target vehicle becomes the vehicle of incision;When there is no this vehicle of target vehicle in cruise in front of this track, in adjacent lane
Target carriage moved to this track, this vehicle is considered that common-denominator target enters tracing mode by cruise if p >=0.51, if
p<Then think that this vehicle does not cut into this track when 0.51
Warning module carries out early warning, to ensure vehicle driving safety according to the approximation ratio of Ben Che and target object.Root
It is old whether the distance between this vehicle and common-denominator target vehicle for being measured according to radar electric system and the distance meet early-warning conditions
Decide whether to open alarm;Include mainly distance calculation module, control module, alarm device;Distance calculation module
Further, in this car state estimation module further include two degrees of freedom car model module.
It is tested for the common-denominator target vehicle discriminating in the case of two kinds of straight way and bend.
Simulation parameter is arranged under straight way operating mode:The wide 4m in track, this vehicle are at the uniform velocity travelled with the speed of 20m/s;Target carriage 1 with
The speed of 20m/s is in the left-hand lane away from this vehicle lateral distance 4m, traveling ahead of the fore-and-aft distance away from this vehicle 50m;Target carriage 2 with
The speed of 20m/s in the right-hand lane fore-and-aft distance away from this vehicle 4m away from the traveling ahead of this vehicle 70m;Target carriage 3 is with 20m/s's
Speed travels in the front away from this vehicle 90m.
Under straight way operating mode other than quickly calculating each target carriage at a distance of the distance of the center line in this track, also want
It quickly recognizes target carriage in adjacent lane and carrys out the probability for cutting this track.The information that radar is obtained is counted by formula (15)
It calculates and obtains the distance that a target carriage travels trace centerline away from this vehicle, as shown in Figure 5.Radar is obtained to the transverse movement of target carriage
Speed and laterally opposed distance are input to the incision probability that target carriage is calculated in fuzzy controller, to quickly recognize front
Common-denominator target.
Gradually determine that critical value, target carriage think that this track is transported with the side velocity of 1m/s respectively by Multi simulation running experiment
The side velocity of dynamic 4s, 0.8m/s move the side that the side velocity of 5s, 0.7m/s moves to this track 2s, 0.5m/s to this track
2s is moved to speed to this track.Sensor acquisition target carriage vehicle data, which is input in fuzzy controller, obtains target carriage not
The probability in this track is cut under same traffic scene, as shown in Figure 4.
As seen in Figure 4:Cutting vehicle 1 and cutting the minimum stationary value of 2 probability of vehicle is not cut 0.51 or more
It is below 0.51 to enter vehicle 1 and be not cut into the maximum value of 2 probability of vehicle.Accordingly, it is determined that going out whether target carriage cuts this track
Probability critical value be 0.51.
By in the information input to common-denominator target arbiter of the target carriage 1 obtained by radar and target carriage 2, calculating is arrived
The probability in this track of target carriage 1 and the incision of target carriage 2.As shown in fig. 6, solid line indicates target carriage 1 during the motion in figure
Cut the probability situation of change in this track.Variation tendency by curve and the set target carriage in emulation duty parameter setting
Known to 1 motion state:In 5s, target carriage 1 starts to start to be become larger rapidly by smaller value to this track incision curve, passes through
The smaller time reaches critical value 0.51, and curve values become smaller rapidly again after completing lane-change to target carriage 1 drops to critical value or less
It is most steady to be scheduled on 0.1 or so eventually.What fuzzy controller can be promptly and accurately as seen through the above analysis, which determine target carriage 1, cuts
The fact that enter this track.
Dotted line indicates that target carriage 2 cuts the probability situation of change in this track during the motion in figure.Pass through the change of curve
Change trend and emulation duty parameter setting in set target carriage 2 motion state known to:In 5s, target carriage 2 start to
A left side travels duration 2s with the side velocity of 0.5m/s, and does not enter this track and can regard target carriage 2 as in oneself track
Carry out the simple adjustment of transport condition.Curve tendency is seen it is found that curve has the rising of very little not change significantly in 5s,
And not up to critical value.After 2s after the completion of the adjustment of 2 transport condition of target carriage, curve has decline again, and finally stablizes on 0.1 left side
It is right.Fuzzy controller, which can accurately determine target carriage 2, as seen through the above analysis can not possibly cut this track.
By Fig. 5 and Fig. 6 we can see that target carriage 1 has cut this track, and target carriage 2 does not cut this vehicle
Road.Therefore, front common-denominator target has target carriage 3 to become target carriage 1.When meeting p >=0.51, then it is assumed that target carriage will be cut into
It is potential common-denominator target in this track.The chosen distance from the objects ahead vehicle in these potential targets and this track
The nearest target of this vehicle is exactly the common-denominator target for the adaptive cruise front side to be differentiated herein.The ID of this target is exported, and
Its corresponding radar data (distance and bearing angle etc.) is sent into radar electric system.
Experiment in the case of bend
(1) straight way is into bend
Parameter setting:This vehicle is on straight way, and target carriage 1 is in straight way, target carriage 2 and target carriage 3 and is in bend.Track
Wide 4m, this vehicle are at the uniform velocity travelled with the speed of 20m/s;Target carriage 1 is with the speed of 20m/s in the left side vehicle away from this vehicle radial distance 4m
Road, traveling ahead of the distance away from this vehicle 60m;Target carriage 2 with the speed of 20m/s the right-hand lane distance away from this vehicle 4m away from this
The traveling ahead of vehicle 70m;Target carriage 3 with the speed of 20m/s in the identical lanes away from this vehicle 90m, as shown in Figure 7.
(2) in bend
Parameter setting:Target carriage 1, target carriage 2 and target carriage 3 are all on bend, and the wide 4m in track, this vehicle is with the speed of 20m/s
Degree at the uniform velocity travels;Target carriage 1 with the speed of 20m/s in the left-hand lane away from this vehicle radial distance 4m, before distance is away from this vehicle 60m
Side's traveling;Target carriage 2 is with the speed of 20m/s in the right-hand lane away from this vehicle 4m, traveling ahead of the distance away from this vehicle 70m;Mesh
Vehicle 3 is marked with the speed of 20m/s in the identical lanes away from this vehicle 90m, as shown in Figure 8.
(3) bend lane-change
Parameter setting:Bend lane-change operating mode simulation parameter is arranged:Target carriage 1, target carriage 2 and target carriage 3 all on bend,
The wide 4m in track, this vehicle are at the uniform velocity travelled with the speed of 20m/s;Target carriage 1 is with the speed of 20m/s on the left side away from this vehicle radial distance 4m
Side track, traveling ahead of the distance away from this vehicle 60m;Target carriage 2 with the speed of 20m/s in the right-hand lane away from this vehicle 4m, away from
The traveling ahead of this vehicle of separation 70m, starts the track that this vehicle traveling is cut with the radial velocity of 1m/s after travelling 5s, and 4s is completed
Entire lane-change process;Target carriage 3 with the speed of 20m/s in the identical lanes away from this vehicle 90m, as shown in Figure 9.
(4) it goes off the curve
Parameter setting:Operating mode of going off the curve simulation parameter is arranged:This vehicle is also on bend, and target carriage 3 has then been driven into
Straight way, target carriage 1 and target carriage 2 are all also in bend.Target carriage 2 and target carriage 1 drive into straight way after 1s and 2s respectively.Track is wide
4m, this vehicle are at the uniform velocity travelled with the speed of 20m/s;Target carriage 1 is with the speed of 20m/s in the left side vehicle away from this vehicle radial distance 4m
Road, traveling ahead of the distance away from this vehicle 60m;Target carriage 2 with the speed of 20m/s in the right-hand lane away from this vehicle 4m, distance away from
The traveling ahead of this vehicle 70m;Target carriage 3 with the speed of 20m/s in the identical lanes away from this vehicle 90m, as shown in Figure 10.
Simulation result
(1) enter bend operating mode simulation analysis:
As shown in Figure 7:Target carriage 1 is when left-hand lane travels beginning and does not enter bend, the closest approach of radar surveying
It is the right rear side boundary point of target carriage 1.And after entering bend, radar surveying is its rear midpoint;Target carriage 2 is in right side vehicle
Road traveling does not drive into bend when starting, and the closest approach of radar surveying is the left rear side boundary point of target carriage 2.Into bend
Afterwards, the point of radar surveying does not change.The target carriage 3 in lanes identical with this vehicle, beginning of target carriage 3 comes into curved
The closest approach in road, radar surveying is the rear midpoint of target carriage 3.
As shown in Figure 11:The distance difference away from this vehicle driving trace center line through the calculated target carriage of target-recognition model
For, when not entering bend, target carriage 1, target carriage 2 and target carriage 3 away from Ben Cheche driving trace center lines be respectively 3.2m, 3.2m and
0.2m or so.Into after bend, distance of the target carriage away from this vehicle driving trace becomes 3.75m, 3.25m and 0.9m or so respectively.
By with the radial distance of target carriage 1, target carriage 2 and target carriage 3 away from this track in emulation operating mode setting relatively after, can be with
It obtains:When entering curved, algorithm taken herein can recognize that front is in the target in this track.
(2) bend operating mode simulation analysis
As shown in Figure 8:This vehicle and target carriage have all driven into bend, and the closest approach of radar surveying is target carriage 1 respectively
Rear midpoint nearby, the left rear side boundary point of the left rear side boundary point of target carriage 2 and target carriage 3, and entirely travelling
The closest approach of radar surveying does not convert in the process.
As shown in Figure 12:The distance difference away from this vehicle driving trace center line through the calculated target carriage of target-recognition model
For 3.75m, 3.2m and 0.9m or so, the target carriage 1, target carriage 2 and target carriage 3 that center are being set away from this vehicle with emulation operating mode
After the radial distance of driving trace center line compares and considers with the closest approach of the target carriage of radar surveying, it can obtain
Go out:In bend, method taken herein can recognize that front is in the target in this track.
(3) bend lane-change operating mode simulation analysis
By Fig. 9 (a) and (b):This vehicle and target carriage are all in negotiation of bends, and target carriage 2 is after travelling a period of time
It completes lane-change and cuts this track.The closest approach of radar surveying be respectively target carriage 1 rear midpoint nearby, the rear of target carriage 2
The left rear side boundary point of left border point and target carriage 3, the closest approach of radar surveying after completing lane-change of target carriage 2 do not become
Change.
As shown in Figure 13:Target carriage 1, target carriage 2 and target carriage 3 are calculated by target-recognition model to travel away from this vehicle
Track center line is respectively 3.75m, 3.2m and 0.9m or so, at this point, common-denominator target is target carriage 3.When 5s, target carriage 2 is opened
Beginning lane-change can be seen that distance of the target carriage 2 away from this lane center tapers by dotted line in figure, and lane-change mesh is completed after 4s
Spacing this track center line 0.9m or so is marked, at this point, common-denominator target becomes target carriage 2.
(4) go out curved operating mode simulation analysis
As shown in Figure 10:After target carriage and Ben Che are all in bend 2s when beginning, mesh vehicle initially enters bend.Radar is surveyed
The closest approach of amount becomes 1 right rear side boundary point of target carriage from 1 rear midpoint of target carriage;The closest approach of 2 radar surveying of target carriage
It is always the left rear side boundary point of target carriage 2;It is the rear of target carriage 3 left side when closest approach of 3 radar surveying of target carriage starts
Lateral boundaries point, after become rear midpoint.
As shown in Figure 14:When not entering straight way, the target carriage being calculated through target-recognition model is travelled away from this vehicle
The distance of track center line in bend calculated distance relatively coincide, through the calculated target carriage of algorithm away from this vehicle travel
The distance of track center line is respectively 3.75m, 3.2m and 0.9m or so.After target carriage successively enters bend, and radar surveying
The closest approach of target carriage is varied from, and through the calculated target carriage of algorithm, the distance away from this vehicle driving trace center line is respectively
3.2m, 3.2m and 0.2m or so, compared with the target carriage 1,2,3 in emulation operating mode setting away from the radial distance in this track
Afterwards, it can be deduced that:When going out curved, method taken herein can recognize that front is in the target in this track.
It to sum up analyzes into bend, the bend that exists together, bend lane-change and simulation result of going off the curve, we can see that institute of the present invention
The method of use accurate can calculate relatively radially distance of the target carriage away from this vehicle driving trace in different tracks, according to
This determines the common-denominator target of adaptive cruise front side, and then according to approaching distance and carry out between Ben Che and common-denominator target
Early warning is to ensure the safety of this vehicle traveling.
Claims (9)
1. one kind approaching detection and target-recognition method, it is characterised in that including with step:
S1:The estimation of this car state is carried out based on Extended Kalman filter, the yaw velocity and side slip angle to this vehicle carry out
Estimation, and then realize the real-time online estimation of road curvature;
S2:Judge that travelled road is straight way by obtaining the curvature for being presently in road to the estimation of road curvature
Or bend;
S3:For under the different situations of straight way and bend, establishing corresponding common-denominator target discrimination model respectively, to the pass in front
Key target is differentiated;
S4:Early warning is carried out according to the approximation ratio of Ben Che and target object, to ensure vehicle driving safety.
2. one kind is according to claim 1 to approach detection and target-recognition method, it is characterised in that this vehicle in step S1
The model of the vehicle used when state estimation is two degrees of freedom car model.
3. one kind is according to claim 1 or 2 to approach detection and target-recognition method, it is characterised in that:Two degrees of freedom vapour
Vehicle model does not consider the influence of steering, and by front wheel angle (steering wheel angle) as inputting, vehicle body only does plane motion, preceding
Constant into speed, tire cornering characteristics are in the range of linearity, do not consider the influence of air drag.
4. one kind is according to claim 1 or 2 or 3 to approach detection and target-recognition method, patrolled using fuzzy in step S3
It collects algorithm to differentiate common-denominator target, the fuzzy logic algorithm needs to use gravity model appoach.
5. a kind of approaching detection and target-recognition method, vehicle two degrees of freedom according to any one of claim 1-4
The filter state equation and observational equation of model are as follows:
State equation:
Observational equation:
Wherein, ω is yaw velocity, and β is side slip angle, and δ is front wheel angle, w1、w2For system noise, v1、v2For sensor
Measurement noise, w1、w2、v1、v2For mutually independent white noise.
6. a kind of using the trailer-mounted radar Department of Electronics for approaching detection and target-recognition method described in claim 1-5 any one
System, it is characterised in that including:
This car state estimation module carries out the estimation of this car state for being based on Extended Kalman filter, to the yaw angle speed of this vehicle
Degree and side slip angle are estimated, and then realize the real-time online estimation of road curvature;
Road curvature estimation module is judged to be gone by obtaining the curvature for being presently in road to the estimation of road curvature
The road sailed is straight way or bend;
Target-recognition module, under the different situations of straight way and bend, establishing corresponding common-denominator target discrimination model respectively,
The common-denominator target in front is differentiated.
Warning module carries out early warning, to ensure vehicle driving safety according to the approximation ratio of Ben Che and target object.
7. a kind of trailer-mounted radar electronic system according to claim 6, it is characterised in that:In this car state estimation module
It further include two degrees of freedom car model module.
8. the trailer-mounted radar electronic system described in a kind of according to claim 6 or 7, it is characterised in that:Two degrees of freedom car model
Module does not consider the influence of steering, and by front wheel angle (steering wheel angle) as inputting, vehicle body only does plane motion, advances
Speed is constant, and tire cornering characteristics are in the range of linearity, do not consider the influence of air drag.
9. a kind of trailer-mounted radar electronic system according to claim 6-8 any one, it is characterised in that:Vehicle two is freely
Filter state equation and the observational equation for spending model are as follows:
State equation:
Observational equation:
Wherein, ω is yaw velocity, and β is side slip angle, and δ is front wheel angle, w1、w2For system noise, v1、v2For sensor
Measurement noise, w1、w2、v1、v2For mutually independent white noise.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710358174.9A CN108569289A (en) | 2017-05-19 | 2017-05-19 | A kind of trailer-mounted radar and its approach detection and target-recognition method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710358174.9A CN108569289A (en) | 2017-05-19 | 2017-05-19 | A kind of trailer-mounted radar and its approach detection and target-recognition method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108569289A true CN108569289A (en) | 2018-09-25 |
Family
ID=63576052
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710358174.9A Pending CN108569289A (en) | 2017-05-19 | 2017-05-19 | A kind of trailer-mounted radar and its approach detection and target-recognition method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108569289A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109693669A (en) * | 2018-12-29 | 2019-04-30 | 北京经纬恒润科技有限公司 | It is a kind of to determine recently in the method and system of diameter front truck |
CN111016919A (en) * | 2018-10-10 | 2020-04-17 | 现代自动车株式会社 | Device and method for identifying a close cut into a vehicle and vehicle comprising such a device |
CN111487659A (en) * | 2019-01-28 | 2020-08-04 | 广州市中海达测绘仪器有限公司 | State recognition method and device, computer equipment and storage medium |
CN111645679A (en) * | 2019-03-04 | 2020-09-11 | Aptiv技术有限公司 | Side impact risk estimation system for vehicle |
WO2021144105A1 (en) * | 2020-01-16 | 2021-07-22 | Valeo Schalter Und Sensoren Gmbh | Method for determining the current angle of lateral inclination of a roadway |
CN114043993A (en) * | 2022-01-13 | 2022-02-15 | 深圳佑驾创新科技有限公司 | Key target selection method and device suitable for intelligent driving vehicle |
CN115402307A (en) * | 2022-11-03 | 2022-11-29 | 天津所托瑞安汽车科技有限公司 | Method for early warning of cut-in of bypass vehicle, electronic device and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5510990A (en) * | 1992-12-08 | 1996-04-23 | Nippondenso Co., Ltd. | System for distinguishing a vehicle traveling ahead based on an adjustable probability distribution |
CN1668938A (en) * | 2002-07-15 | 2005-09-14 | 汽车系统实验室公司 | Road curvature estimation and automotive target state estimation system |
CN106114511A (en) * | 2016-07-21 | 2016-11-16 | 辽宁工业大学 | A kind of automobile cruise system core target identification method |
-
2017
- 2017-05-19 CN CN201710358174.9A patent/CN108569289A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5510990A (en) * | 1992-12-08 | 1996-04-23 | Nippondenso Co., Ltd. | System for distinguishing a vehicle traveling ahead based on an adjustable probability distribution |
CN1668938A (en) * | 2002-07-15 | 2005-09-14 | 汽车系统实验室公司 | Road curvature estimation and automotive target state estimation system |
CN106114511A (en) * | 2016-07-21 | 2016-11-16 | 辽宁工业大学 | A kind of automobile cruise system core target identification method |
Non-Patent Citations (1)
Title |
---|
朱甲林: "基于雷达信息巡航车前方关键目标的判别研究", 《中国优秀硕士学位论文全文数据库工程科技II辑》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111016919A (en) * | 2018-10-10 | 2020-04-17 | 现代自动车株式会社 | Device and method for identifying a close cut into a vehicle and vehicle comprising such a device |
CN109693669A (en) * | 2018-12-29 | 2019-04-30 | 北京经纬恒润科技有限公司 | It is a kind of to determine recently in the method and system of diameter front truck |
CN111487659A (en) * | 2019-01-28 | 2020-08-04 | 广州市中海达测绘仪器有限公司 | State recognition method and device, computer equipment and storage medium |
CN111645679A (en) * | 2019-03-04 | 2020-09-11 | Aptiv技术有限公司 | Side impact risk estimation system for vehicle |
WO2021144105A1 (en) * | 2020-01-16 | 2021-07-22 | Valeo Schalter Und Sensoren Gmbh | Method for determining the current angle of lateral inclination of a roadway |
CN114043993A (en) * | 2022-01-13 | 2022-02-15 | 深圳佑驾创新科技有限公司 | Key target selection method and device suitable for intelligent driving vehicle |
CN114043993B (en) * | 2022-01-13 | 2022-04-29 | 深圳佑驾创新科技有限公司 | Key target selection method and device suitable for intelligent driving vehicle |
CN115402307A (en) * | 2022-11-03 | 2022-11-29 | 天津所托瑞安汽车科技有限公司 | Method for early warning of cut-in of bypass vehicle, electronic device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108569289A (en) | A kind of trailer-mounted radar and its approach detection and target-recognition method | |
CN108313054B (en) | The autonomous lane-change decision-making technique of automatic Pilot and device and automatic driving vehicle | |
CN106114511B (en) | A kind of automobile cruise system core target identification method | |
JP7003423B2 (en) | Vehicle driving control device | |
CN108437986A (en) | Vehicle drive assist system and householder method | |
CN109032131A (en) | A kind of dynamic applied to pilotless automobile is overtaken other vehicles barrier-avoiding method | |
CN104192144A (en) | Automobile active anti-collision curve false-alarm eliminating method | |
CN108594812A (en) | A kind of intelligent vehicle smooth track planing method of structured road | |
CN105427669A (en) | Anti-collision early warning method based on DSRC vehicle-to-vehicle communication technology | |
CN110816540B (en) | Traffic jam determining method, device and system and vehicle | |
Nobe et al. | An overview of recent developments in automated lateral and longitudinal vehicle controls | |
CN105000019A (en) | Method and system for detecting, tracking and estimating stationary roadside objects | |
CN105809130A (en) | Binocular depth perception-based vehicle travelable area calculation method | |
CN110979339B (en) | Front road form reconstruction method based on V2V | |
CN109131312A (en) | A kind of intelligent electric automobile ACC/ESC integrated control system and its method | |
CN112249008B (en) | Unmanned automobile early warning method aiming at complex dynamic environment | |
CN102795225A (en) | Method for detecting disturbance state of driver by utilizing driver-side longitudinal control model | |
CN104635233A (en) | Method for estimating and classifying motion states of front object based on vehicle-mounted millimeter wave radar | |
CN105631217A (en) | Vehicle self-adaptive virtual lane based front effective target selection system and method | |
CN106569214A (en) | Method and system for processing vehicle-mounted radar data of adaptive cruise vehicle in conjunction with navigation information | |
CN113291318B (en) | Unmanned vehicle blind area turning planning method based on partially observable Markov model | |
CN113895459B (en) | Barrier screening method and system | |
Zhang et al. | A framework for turning behavior classification at intersections using 3D LIDAR | |
CN115503700A (en) | Active collision avoidance method based on multivariate risk fusion potential field and vehicle-mounted electronic equipment | |
CN114735018A (en) | Single-radar-based front most-dangerous target vehicle screening and judging method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180925 |
|
RJ01 | Rejection of invention patent application after publication |