CN104635233B - Objects in front state estimation and sorting technique based on vehicle-mounted millimeter wave radar - Google Patents
Objects in front state estimation and sorting technique based on vehicle-mounted millimeter wave radar Download PDFInfo
- Publication number
- CN104635233B CN104635233B CN201510085048.1A CN201510085048A CN104635233B CN 104635233 B CN104635233 B CN 104635233B CN 201510085048 A CN201510085048 A CN 201510085048A CN 104635233 B CN104635233 B CN 104635233B
- Authority
- CN
- China
- Prior art keywords
- objects
- motion
- state
- motion state
- speed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
Present invention is disclosed a kind of objects in front state estimation and sorting technique based on vehicle-mounted millimeter wave radar, side velocity information based on the limited objects in front motion measured directly of vehicle-mounted millimeter wave radar, establish the equation of motion of the objects in front under earth coordinates, using adaptive Kalman filter algorithm for estimating, objects in front motion state is real-time and accurately estimated；Then according to objects in front movable information, classified according to specified motion state division threshold speed and motion state transformation rule.Technical solution of the present invention it is real-time, substantially increase the accuracy of state estimation compared with the prior art, and specify strict objects in front motion state switch condition, ensure that classification accuracy.
Description
Technical field
The present patent application belongs to Radar Technology field, is related to objects in front state estimation and classification, available for advanced
Driver assistance system.
Background technology
In recent years, advanced driver assistance system ADAS has turned into the study hotspot of automotive safety technology.Its is main at present
Include adaptive cruise control system, Lane Departure Warning System, front collision early warning system etc..Advanced driver assistance system
Based on the advanced information sensing technology such as system and radar and computer vision, the driver comfort and vehicle of driver are improved
Driving safety.
Millimetre-wave radar is widely used in driver assistance system, for measure front medium and long distance in vehicle,
The targets such as barrier.For continuous wave millimetre-wave radar, its measuring principle is substantially：Emitter produces continuous high frequency persistent wave,
Its frequency carries out cyclically-varying in time.Reflect this period for returning to antenna through target again in radar wave propagation to target
Interior, now the frequency of emitter, which compares echo frequency, change, therefore becomes in mixer output and difference frequency electricity occur
Pressure.The relative distance of the difference frequency voltage directly between radar and objects ahead is related.And ought relative velocity not be between the two
0, due to Doppler effect, echo frequency also has with transmitter frequency on the basis of the foregoing difference frequency as caused by relative distance
Frequency-splitting changes, and the difference frequency is directly relevant with both relative velocities.But relative distance and the direct measurement of relative velocity
Based on vehicle-mounted millimeter wave radar fix system, objects in front motion state measured directly cannot be used directly for drive assist system,
Therefore need real-time and accurately to estimate objects in front motion state.
Vehicle condition parameter Estimation is early widely used among all kinds of control systems of automobile.Early stage is in Vehicle Stability Control
In system (Electronic Stability Program, ESP) research, the cars such as the wheel speed of low cost, yaw velocity are utilized
Information, recycle Kalman Filter Estimation algorithm, automobile side slip angle, coefficient of road adhesion etc. are difficult to direct measurement or
The higher information of vehicles of measurement cost is estimated, and is used for stabilitrak.Except being estimated using classical Kalman filtering
Calculating method, also introduce the state estimation algorithms such as particle filter, adaptive Kalman filter.Therefore, the application is direct in radar
On the basis of the objects in front movement state information of measurement, objects in front is based on using adaptive Kalman filter algorithm for estimating
The motion state of geodetic coordinates carries out accurate estimation in real time.
The classification of existing radar objects in front is classified according to its attribute mostly, be such as divided into wheeled vehicle, endless-track vehicle,
Pedestrian, trees etc..
The content of the invention
Aiming at for the present invention overcomes the shortcomings of above-mentioned prior art, before proposing a kind of radar based on vehicle-mounted millimeter wave
Square object body state estimation and sorting technique, to improve the estimation of objects in front motion state and classification accuracy.
The technical scheme that is achieved of above-mentioned first purpose of the present invention is：Front thing based on vehicle-mounted millimeter wave radar
Body estimation method of motion state, it is characterised in that：Based on the limited objects in front motion measured directly of vehicle-mounted millimeter wave radar
Side velocity information, establish the equation of motion of the objects in front under earth coordinates, estimated using adaptive Kalman filter
Algorithm, real-time and accurately estimate objects in front motion state.
Further, comprise the following steps：
I, it is by the equation of motion of the vehicle-mounted millimeter wave radar map objects in front under earth coordinates：
In above formula,For the acceleration of objects in front,For the speed of objects in front, xobj_R(t) it is
The distance of objects in front；
II, the equation of motion for characterizing objects in front：
Wherein Λ is object moving state sytem matrix in single reference axis；B is process noise matrix；W (t)=[wx
(t), wy(t)]T, wx(t)~N (0, σwx 2), wy(t)~N (0, σwy 2) it is separate random white noise process；
The discrete time model of objects in front side lengthwise movement equation is：
xk+1=diag [Φ, Φ] xk+ diag [G, G] wk
The observational equation of objects in front motion state is：
Z (t)=Cx (t)+v (t)
Wherein, z (t) is observing matrix；C is output state matrix；V (t)=[vx(t), vx(t), vy(t)]T, v (t)~N
(0, R) it is white Gaussian noise process；
Objects in front motion state observational equation discrete time model is：
zk=Cxk+vk, wherein, zkFor observation vector；vkFor Gaussian sequence；
III, the state equation using the discrete time equation of objects in front motion state equation as wave filter, with objects in front
The discrete time equation of motion state observational equation is the observational equation of wave filter, estimates to calculate using adaptive Kalman filter
Method, accurate estimation in real time is carried out to the motion state of objects in front, adaptive Kalman filter algorithm for estimating includes prediction, correction
It is as follows with noise three processes of estimation, detailed process：
First, process is predicted：
Status predication equation：Wherein, x (k) is state vector and the survey at k moment
Amount vector, A are systematic state transfer matrix, and q (k) is the average of system noise；
Error covariance predictive equation：P (k+1 | k)=Ap (k | k) AT+ Q (k), wherein, P is prediction covariance matrix；
Intermediate variable：Wherein, y (k) is measurement vector, and H is output
State matrix, r (k) are the average of observation noise, and b is forgetting factor；
2nd, trimming process：
Gain equation：K (k+1 | k)=P (k+1 | k) HT[HP(k+1|k)HT+R(k)]-1,
KkFor kalman gain matrix,
Filtering equations：X (k+1 | k+1)=x (k+1 | k)+K (k+1) ε (k+1),
Error covariance renewal equation：P (k+1 | k+1)=[I-K (k+1) H] P (k+1 | k),
3rd, noise estimation procedure：
The average and auto-covariance matrix of noise estimate that equation is：
Wherein, QkFor auto-covariance matrix.
The technical scheme that is achieved of above-mentioned second purpose of the present invention is：In the base of objects in front state estimation
On plinth, according to objects in front movable information, according to specified motion state division threshold speed and motion state transformation rule
Classified, the objects in front motion state is divided into fortune unfiled, static, in the same direction according to existing and historical movement state is specific
Dynamic, move toward one another, stoppings but motion and stopping but move toward one another is several before in the same direction before：
IV, by being defined as follows threshold speed, objects in front motion state is classified：
Vt- when objects in front speed is equal to or higher than the threshold value, objects in front is considered as to move in the same direction, and the threshold value is led to
Following formula is crossed to be dynamically updated in each control circulation：
Vt=Vminmoving+Vego*k1+aego*k2, wherein, VminmovingFor this vehicle speed it is relatively low when this car considered to be in row
Sail the threshold speed of state, VegoFor this vehicle speed of current time, aegoFor current time this car acceleration, k1And k2It is logical to need
Cross weights determined by real vehicle debugging；
Vx- when objects in front speed is equal to or less than the threshold value, objects in front is considered as move toward one another, and the threshold is set
Static parameter is set to, currency is -3m/s.
When objects in front speed is in VtWith VxBetween when, its motion state be it is static or stop determining whether need root
Carried out according to the historical movement state of the objects in front；
V, objects in front motion state is classified and the conversion formulation respective rule between motion state, between each state only
State Transferring can be carried out by the rule：
I), unfiled → static, in the same direction motion, move toward one another, if objects in front speed is in the range of in three circulations
And objects in front data can then realize that the motion state is changed by stably measured；
Ii), motion → stopping in the same direction but before motion in the same direction；Move toward one another → stopping but before motion in the same direction, front thing
In continuous two circulations of body speed all near 0, then it can realize that the motion state is changed；
Iii it is), static, stop but before in the same direction motion, stop but before move toward one another → in the same direction motion, move toward one another, even
Continuous three circulation objects in front speed is more than 0, then can realize that the motion state is changed.
The application implementation of objects in front state estimation of the present invention and sorting technique, it is protruded compared to prior art
Effect is：The accuracy of state estimation is substantially increased, and specifies strict objects in front motion state switch condition, is protected
Classification accuracy is demonstrate,proved.
Brief description of the drawings
Fig. 1 is based on vehicle-mounted millimeter wave radar objects in front state estimation and classification process block diagram for the present invention.
Fig. 2 is adaptive Kalman filter algorithm for estimating schematic diagram of the present invention.
Fig. 3 is objects in front motion state transformation rule schematic diagram of the present invention.
Embodiment
Classification of the application to objects in front is not based on its build-in attribute, is divided into if being directly based upon its motion state
Ganlei.Obtainable objects in front motion state includes side fore-and-aft distance, side longitudinal velocity, side longitudinal acceleration, azimuth etc..
In view of control system arithmetic speed and measurement error is likely to occur, the application selects the longitudinal velocity of objects in front and sets phase
Threshold speed is answered to be classified.Meanwhile strict objects in front motion state switch condition is formulated, ensure the accurate of classification
Property.
Just accompanying drawing in conjunction with the embodiments below, is described in further detail to the embodiment of the present invention, so that of the invention
Technical scheme is more readily understood, grasped.
Initially set up the motion state model of objects in front.The objects in front of vehicle-mounted millimeter wave radar mainly include vehicle,
Pedestrian, trees etc..These targets are generally minimum without movement in vertical direction or its movement in vertical direction speed.Therefore, originally
The motion state model established in invention ignores its movement in vertical direction, only focuses on motion of the objects in front in horizontal plane.Cause
This state estimation is just reduced to the state estimation of the objects in front in earth coordinates lower horizontal plane.When not considering car
When carrying the deformation of millimetre-wave radar support, it is believed that vehicle-mounted millimeter wave radar is fixed together with this car.Therefore, first really
The equation of motion for determining vehicle-mounted millimeter wave radar coordinate system relative to the earth is：
In above formula,For the acceleration of vehicle-mounted millimeter wave radar, avFor the acceleration of this car,For vehicle-mounted millimeter
The speed of ripple radar,For the initial velocity of this car, xR(t) it is the mounting distance of vehicle-mounted millimeter wave radar.
The equation of motion of the objects ahead under earth coordinates be：
In above formula,For the acceleration of objects in front, a is the acceleration figure of objects in front,For the speed of objects in front
Degree, x (t) are the distance of objects in front.
Going out the equation of motion of the objects in front under vehicle-mounted millimeter wave radar motion coordinate system by above-mentioned equation inference is：
In above formula,For the acceleration of objects in front,For the speed of objects in front, xobj_R(t) before being
The distance of square object body.
UsingThe motion state of objects in front, including its side fore-and-aft distance, speed and acceleration are described
Degree etc..The side lengthwise movement equation of objects ahead can be expressed as：
Wherein Λ is object moving state sytem matrix in single reference axis；B is process noise matrix；
W (t)=[wx(t), wy(t)]T, wx(t) N (0, σwx 2), wy(t) N (0, σwy 2) it is separate random white noise
Process.
The discrete time model of objects in front side lengthwise movement equation is：
xk+1=diag [Φ, Φ] xk+ diag [G, G] wk (5)
The observational equation of objects in front motion state is：
Z (t)=Cx (t)+v (t)
Wherein, z (t) is observing matrix；C is output state matrix；V (t)=[vx(t), vx(t), vy(t)]T, v (t)~N
(0, R) it is white Gaussian noise process.
Objects in front motion state observational equation discrete time model is：
zk=Cxk+vk (7)
Wherein, zkFor observation vector；vkFor Gaussian sequence.
For the objects in front motion state equation established, while in view of the various cars including vehicle-mounted millimeter wave radar
The statistical property of set sensor is difficult to predefine, therefore objects in front motion state is entered using adaptive Kalman filter algorithm
Row accurate estimation in real time.
Adaptive Kalman filter algorithm includes prediction, correction and noise and estimates three processes.
Reference picture 2, during prediction, the motion state based on current time, the motion state of subsequent time can be entered
Row prior estimate：
Status predication equation：
Wherein, x (k) is the state vector and measurement vector at k moment, and A is systematic state transfer matrix, and q (k) is system noise
The average of sound.
Error covariance predictive equation：
P (k+1 | k)=Ap (k | k) AT+Q(k) (9)
Wherein, P is prediction covariance matrix.
Intermediate variable：
Wherein, y (k) is measurement vector, and H is output state matrix, and r (k) is the average of observation noise, and b is forgetting factor.
In correcting process, the motion state observed is combined with the motion state pre-estimated, obtains posteriority
Estimation：
Gain equation：
K (k+1 | k)=P (k+1 | k) HT[HP(k+1|k)HT+R(k)]-1 (11)
KkFor kalman gain matrix.
Filtering equations：
X (k+1 | k+1)=x (k+1 | k)+K (k+1) ε (k+1) (12)
Error covariance renewal equation：
P (k+1 | k+1)=[I-K (k+1) H] P (k+1 | k) (13)
In noise estimation procedure, using estimating residual sequence estimation amendment observation noise and system noise covariance square
Battle array, realize the accurate estimation in real time to objects in front motion state：
Average and auto-covariance matrix the estimation equation of noise：
Wherein, QkFor auto-covariance matrix.
According to the equation of motion and observational equation of the foregoing objects in front under earth coordinates, formula (5) is Kalman filtering
The state equation of device, formula (7) are the measurement equation of Kalman filter.
The input information of wave filter includes the direct measurement information of vehicle-mounted millimeter wave radar, including objects in front it is relative away from
From, azimuth and relative velocity：
zk=[x, vr, y]T (15)
Have in the objects in front motion state obtained estimated by adaptive Kalman filter, choose the longitudinal direction of objects in front
Speed, classify for objects in front motion state.
Two threshold speeds are calculated first.
Vt- when objects in front speed is equal to or higher than the threshold value, objects in front is considered as to move in the same direction.The threshold value is led to
Following formula is crossed to be dynamically updated in each control circulation：
Vt=Vminmoving+Vego*k1+aego*k2 (16)
Wherein, VminmovingFor this vehicle speed when relatively low this car by task in the threshold speed of motion, VegoFor current time
This vehicle speed, aegoFor current time this car acceleration, k1And k2For need by real vehicle debug determined by weights.
Vx- when objects in front speed is equal to or less than the threshold value, objects in front is considered as move toward one another.The threshold value quilt
Static parameter is arranged to, currency is -3m/s.
When objects in front speed is between the two threshold speeds, its motion state is probably static or stopped.Enter one
Step judges to need the historical movement state according to the objects in front.
Therefore, in order to by objects in front carry out more accurately classification, it is necessary to objects in front motion state classify and move
Respective rule is formulated in conversion between state.Specifically, as shown in figure 3, when objects in front motion state meets following condition
When, corresponding motion state classification can be realized：
Motion state is unfiled i.e. no measurement data available enough before objects in front, while continuous three are followed
When its data of ring can be by stably measured, if its speed is in VtWith VxBetween, its motion state can convert to static；If its speed is small
In or equal to Vx, its motion state can be converted to opposite traveling；If its speed is more than or equal to Vt, its motion state can be converted to
Traveling in the same direction.
Motion state is motion in the same direction before objects in front, while is all existed in continuous two circulations of objects in front speed
0 nearby or in VtWith VxBetween when, its motion state can be converted to stopping but before in the same direction motion.
Motion state is move toward one another before objects in front, while is all existed in continuous two circulations of objects in front speed
0 nearby or in VtWith VxBetween when, its motion state can be converted to stopping but move toward one another before.
Move toward one another before motion state is static, stopped but moves or stop in the same direction before before objects in front,
Continuous three circulations objects in front speed is more than V simultaneouslyt, its motion state can be converted to motion in the same direction.
Move toward one another before motion state is static, stopped but moves or stop in the same direction before before objects in front,
Continuous three circulations objects in front speed is less than V simultaneouslyx, its motion state can be converted to move toward one another.
In summary, it is detailed description to the specific embodiment of the invention, this case protection domain is not limited in any way.
All technical methods formed using equivalent transformation or equivalent replacement, all fall within rights protection scope of the present invention.
Claims (1)
1. the objects in front motion state sorting technique based on vehicle-mounted millimeter wave radar, it is characterised in that：Moved in objects in front
On the basis of state estimation, according to objects in front movable information, according to specified motion state division threshold speed and motion
State transition rules are classified, the objects in front motion state according to existing and historical movement state it is specific be divided into do not divide
Class, motion static, in the same direction, move toward one another, stopping but before in the same direction motion and stop but move toward one another is several before：
IVth, by being defined as follows threshold speed, objects in front motion state is classified：
Vt- when objects in front speed is equal to or higher than the threshold value, objects in front is considered as to move in the same direction, and the threshold value is under
Formula is dynamically updated in each cycle：
Vt=Vminmoving+Vego*k1+aego*k2, wherein, VminmovingFor this vehicle speed it is relatively low when this car considered to be in traveling shape
The threshold speed of state, VegoFor this vehicle speed of current time, aegoFor current time this car acceleration, k1And k2To need to pass through reality
Weights determined by car debugging；
Vx- when objects in front speed is equal to or less than the threshold value, objects in front is considered as move toward one another, and the threshold value is set
For static parameter, currency is -3m/s；
When objects in front speed is in VtWith VxBetween when, its motion state be it is static or stop determining whether need according to should
The historical movement state of objects in front is carried out；
Vth, respective rule is formulated to the conversion between the classification of objects in front motion state and motion state, can only between each state
State Transferring is carried out by the rule：
I), unfiled → static, in the same direction motion, move toward one another, if objects in front speed is in the range of and preceding in three circulations
Square object speed can then realize that the motion state is changed by stably measured；
Ii), motion → stopping in the same direction but before in the same direction motion, objects in front speed it is continuous two circulation in all near 0, then can
Realize that the motion state is changed；
Iii it is), static, stop but before in the same direction motion, stop but before move toward one another → in the same direction motion, move toward one another, continuous three
Individual circulation objects in front speed is more than 0, then can realize that the motion state is changed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510085048.1A CN104635233B (en) | 2015-02-17 | 2015-02-17 | Objects in front state estimation and sorting technique based on vehicle-mounted millimeter wave radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510085048.1A CN104635233B (en) | 2015-02-17 | 2015-02-17 | Objects in front state estimation and sorting technique based on vehicle-mounted millimeter wave radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104635233A CN104635233A (en) | 2015-05-20 |
| CN104635233B true CN104635233B (en) | 2017-12-26 |
Family
ID=53214188
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510085048.1A Active CN104635233B (en) | 2015-02-17 | 2015-02-17 | Objects in front state estimation and sorting technique based on vehicle-mounted millimeter wave radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104635233B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102015108759A1 (en) * | 2015-06-03 | 2016-12-08 | Valeo Schalter Und Sensoren Gmbh | Method for classifying a longitudinally extended stationary object in a lateral surrounding area of a motor vehicle, driver assistance system and motor vehicle |
| CN105549003A (en) * | 2015-12-02 | 2016-05-04 | 华域汽车系统股份有限公司 | Automobile radar target tracking method |
| CN107783123B (en) * | 2016-08-25 | 2021-07-06 | 大连楼兰科技股份有限公司 | System and method for processing collision-proof millimeter wave radar signal in complex environment of unmanned vehicle |
| US10585185B2 (en) | 2017-02-03 | 2020-03-10 | Rohde & Schwarz Gmbh & Co. Kg | Security scanning system with walk-through-gate |
| CN108037502B (en) * | 2017-09-28 | 2021-10-29 | 南通大学 | Double-radar accurate positioning method for unmanned ship water quality detection operation path |
| CN108226924B (en) * | 2018-01-11 | 2020-11-10 | 南京楚航科技有限公司 | Automobile driving environment detection method and device based on millimeter wave radar and application of automobile driving environment detection method and device |
| CN109143221B (en) * | 2018-07-23 | 2022-06-10 | 奇瑞汽车股份有限公司 | Target tracking method and device |
| CN110007305B (en) * | 2019-04-15 | 2021-01-29 | 北京行易道科技有限公司 | Method, device, server and storage medium for determining vehicle front target |
| CN110208793B (en) * | 2019-04-26 | 2022-03-11 | 纵目科技（上海）股份有限公司 | Auxiliary driving system, method, terminal and medium based on millimeter wave radar |
| CN110300230B (en) * | 2019-07-01 | 2021-03-19 | 腾讯科技（深圳）有限公司 | Application control method, device, storage medium and terminal |
| CN110940979A (en) * | 2019-10-28 | 2020-03-31 | 杭州飞步科技有限公司 | Obstacle detection method, apparatus, device, and storage medium |
| CN111366901B (en) * | 2020-03-05 | 2023-11-10 | 北京小马慧行科技有限公司 | Method and device for calibrating azimuth angle installation deviation of vehicle-mounted millimeter wave radar |
| CN112356845B (en) * | 2020-11-19 | 2022-02-18 | 中国第一汽车股份有限公司 | Method, device and equipment for predicting motion state of target and vehicle |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH04315083A (en) * | 1991-04-12 | 1992-11-06 | Japan Radio Co Ltd | Obstacle detector for car |
| JPH11345394A (en) * | 1998-06-03 | 1999-12-14 | Mitsubishi Electric Corp | Ambient monitoring device for vehicle |
| SE9902141L (en) * | 1999-06-08 | 2000-12-09 | Celsiustech Electronics Ab | Procedure in a radar system |
| WO2004008648A2 (en) * | 2002-07-15 | 2004-01-22 | Automotive Systems Laboratory, Inc. | Road curvature estimation and automotive target state estimation system |
| CN102538781B (en) * | 2011-12-14 | 2014-12-17 | 浙江大学 | Machine vision and inertial navigation fusion-based mobile robot motion attitude estimation method |
-
2015
- 2015-02-17 CN CN201510085048.1A patent/CN104635233B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104635233A (en) | 2015-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104635233B (en) | Objects in front state estimation and sorting technique based on vehicle-mounted millimeter wave radar | |
| KR102497330B1 (en) | Control of autonomous vehicle based on determined yaw parameter(s) of additional vehicle | |
| Jo et al. | Tracking and behavior reasoning of moving vehicles based on roadway geometry constraints | |
| JP6432116B2 (en) | Vehicle position specifying device, vehicle control system, vehicle position specifying method, and vehicle position specifying program | |
| US9199643B1 (en) | Sensor odometry and application in crash avoidance vehicle | |
| CN111458700B (en) | Method and system for vehicle mapping and positioning | |
| KR20200139779A (en) | Autonomous vehicle control based on environmental object classification determined using phase coherent lidar data | |
| Thormann et al. | Extended target tracking using Gaussian processes with high-resolution automotive radar | |
| CN110874642B (en) | Learning device, learning method, and storage medium | |
| JPWO2018131090A1 (en) | Vehicle control device, vehicle control method, and vehicle control program | |
| CN106476728A (en) | Motion compensation for vehicle-mounted vehicle sensors | |
| US20180095103A1 (en) | State calculation apparatus, state calculation method, and recording medium storing program for moving object | |
| CN111942378A (en) | Vehicle control device, vehicle control method, and storage medium | |
| RU2626424C1 (en) | Device and method for evaluation of vehicle attitude angle and position | |
| EP4260092A1 (en) | Method, apparatus and radar system for tracking objects | |
| US10839678B2 (en) | Vehicle identifying device | |
| US10976426B2 (en) | Apparatus and method for ascertaining object kinematics of a movable object | |
| TWI680898B (en) | Light reaching detection device and method for close obstacles | |
| Kamann et al. | Object tracking based on an extended Kalman filter in high dynamic driving situations | |
| US11693110B2 (en) | Systems and methods for radar false track mitigation with camera | |
| JP7138151B2 (en) | VEHICLE CONTROL DEVICE, VEHICLE CONTROL METHOD, AND PROGRAM | |
| CN110678776B (en) | System for enhanced object tracking | |
| US20230182774A1 (en) | Autonomous driving lidar technology | |
| CN116022163A (en) | Automatic driving vehicle scanning matching and radar attitude estimator based on super local subgraph | |
| CN115082562A (en) | External parameter calibration method, device, equipment, server and vehicle-mounted computing equipment |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210928 Address after: 543000 building 4, Yingtian Intelligent Manufacturing Industrial Park, high tech Zone, Wanxiu District, Wuzhou City, Guangxi Zhuang Autonomous Region Patentee after: Guangxi Jingzhi Automobile Technology Co.,Ltd. Address before: 215134 No. 81 Weizhong Road, Weitang Town, Xiangcheng District, Suzhou City, Jiangsu Province Patentee before: SUZHOU ANZHI AUTO PARTS Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Estimation and classification of moving state of objects in front based on vehicle-mounted millimeter wave radar Effective date of registration: 20230112 Granted publication date: 20171226 Pledgee: Wuzhou Small and Micro Enterprises Financing Guarantee Co.,Ltd. Pledgor: Guangxi Jingzhi Automobile Technology Co.,Ltd. Registration number: Y2023450000013 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |