CN110058239A - A kind of vehicle-mounted millimeter wave radar installations and object detection method - Google Patents
A kind of vehicle-mounted millimeter wave radar installations and object detection method Download PDFInfo
- Publication number
- CN110058239A CN110058239A CN201910355156.4A CN201910355156A CN110058239A CN 110058239 A CN110058239 A CN 110058239A CN 201910355156 A CN201910355156 A CN 201910355156A CN 110058239 A CN110058239 A CN 110058239A
- Authority
- CN
- China
- Prior art keywords
- target
- fence
- vehicle
- static
- signal
- 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.)
- Granted
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
- 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
The present invention relates to a kind of vehicle-mounted millimeter wave radar installations and object detection methods, including RF front-end module, power module and signal processing module, the power module is that RF front-end module and signal processing module are powered, the RF front-end module, for emitting millimeter-wave signal, receive the echo-signal of target, and the echo-signal of target is pre-processed, the signal processing module includes: distance and speed acquiring unit, for the velocity information and range information according to pretreated signal acquisition target;Position acquisition unit, for the location information according to pretreated signal acquisition target;Object detection unit for filtering out the fence on track both sides according to velocity information, range information and the location information of obtained target, and extracts the target in fence as effective target.The present invention can effectively filter out the fence on rail traffic both sides, carry out decision convenient for user.
Description
Technical field
The present invention relates to millimetre-wave radar detection technology fields, more particularly to a kind of vehicle-mounted millimeter wave radar installations and mesh
Mark detection method.
Background technique
Radar is the method discovery target using radio and the spatial position for detecting target.Radar can according to signal form
It is divided into pulse radar, continuous wave radar, pulse compression radar and frequency agile radar.And CW with frequency modulation (FMCW) radar is because of knot
Structure is simple, the features such as being easy modulation, is at low cost, signal processing is simple is applied in car radar field.The continuous quilt of frequency modulation
Radar obtains the information of target by the difference on the frequency of the echo-signal frequency of any time and transmitting signal.Millimetre-wave radar is
Refer to radar tooling millimeter wave band detection radar, the wavelength of millimeter wave between centimeter wave and microwave, wavelength be 1~
10mm。
With wisdom traffic and unpiloted development, the development of active probe class sensor has been pushed.And millimeter wave thunder
It is small with antenna aperture, with roomy up to the important a part for being automobile active detector, strong antijamming capability, not by light
It influences, has the characteristics that all weather operations, therefore obtained the units such as every country government department, research institute, colleges and universities and enterprise
Pay much attention to.But current millimetre-wave radar technology is not mature enough, mostly rests on ranging, tests the speed with angle measurement functionally,
Precision is also not high enough, and it is pre- to be not met by adaptive learning algorithms (Adaptive Cruise Control, ACC), front truck anticollision
Alert system (Forward Collision Warning System, FCWS) and advanced driving assistance system (Advanced
Driver Assistance Systems, ADAS) application.
In recent years, millimetre-wave radar obtains certain application in fields such as tramcar, train, cloud rails.But because
The particularity of the vehicle running rail, millimetre-wave radar will lead to decision terminal mistake when detecting the fence target on track both sides
Sentence.
Summary of the invention
Technical problem to be solved by the invention is to provide a kind of vehicle-mounted millimeter wave radar installations and object detection method, energy
The fence on rail traffic both sides is enough effectively filtered out, carries out decision convenient for user.
The technical solution adopted by the present invention to solve the technical problems is: providing a kind of vehicle-mounted millimeter wave radar installations, wraps
RF front-end module, power module and signal processing module are included, the power module is RF front-end module and signal processing mould
Block is powered, the RF front-end module, for emitting millimeter-wave signal, receives the echo-signal of target, and to target
Echo-signal is pre-processed, and the signal processing module includes: distance and speed acquiring unit, for according to pretreated
The velocity information and range information of signal acquisition target;Position acquisition unit, for according to pretreated signal acquisition target
Location information;Object detection unit, for being according to velocity information, range information and the location information determination of obtained target
Bend or straight way, and the fence on track both sides is filtered out, the target in fence is extracted as effective target.
Data associating unit is provided with after the position acquisition unit, the data associating unit cannot by target range
The principle of mutation solves the problems, such as velocity ambiguity.
The velocity information for the target that the object detection unit will test and the velocity information of vehicle compare, judgement
Show whether the target detected is static target, if it is static target, determines that current orbit is straight by RANSAC algorithm
Trajectory or bend track, and determine column location.
If not static target, then judge target abscissa value whether the fixed value between the fence less than 1/2, if
It is then for effective target, is otherwise invalid targets.
After determining current orbit, according to the fixed value between fence of the abscissa value of the location information of target equal to 1/2
Column location is determined, if the fixed value between fence of the abscissa value of target less than 1/2, for effective target.
The RANSAC algorithm process process are as follows: two o'clock building linear equation is arbitrarily found out, all static targets are traversed,
Threshold value is set, if the number of the static target on the straight line found is greater than threshold value, determines that track at this time is straight line
Otherwise track is bend track, and using the straight line of a cover-most static target as fence.
According to the obtained fence, it is located in fence if target is shown, which is effective target, if target is shown
Outside fence, then it is considered as invalid targets.
The technical solution adopted by the present invention to solve the technical problems is: also providing a kind of vehicle-mounted millimeter wave radar installations
Object detection method, comprising the following steps:
(1) radar emission electromagnetic signal, and the echo-signal of target is received, and located in advance to the echo-signal of target
Reason;
(2) according to velocity information, range information and the location information of pretreated signal acquisition target;
(3) it is bend or straight way according to velocity information, range information and the location information determination of obtained target, and filters
Except the fence on track both sides, the target in fence is extracted as effective target.
The wherein specific steps of target acquisition are as follows:
(3.1) velocity information of target and the velocity information of vehicle that will test in step (3) are compared, is judged
Show whether the target detected is static target;
(3.2) if it is static target, then determine that current orbit is rectilinear orbit or bend rail by RANSAC algorithm
Road, and determine column location;
It is determined after determining current orbit according to the fixed value between fence of the abscissa value of target position information equal to 1/2
Column location, if the fixed value between fence of the abscissa value of target less than 1/2, for effective target;
(3.2.1) RANSAC algorithm process process are as follows: arbitrarily find out two o'clock building linear equation, traverse all static mesh
Threshold value is arranged in mark, if the number of the static target on the straight line found is greater than threshold value, determines that track at this time is straight
Otherwise trajectory is bend track, and using the straight line of a cover-most static target as fence;
(3.3) if not static target, then judge target abscissa value whether the fixation between the fence less than 1/2
Value, is if it is effective nonstatic target, is otherwise invalid targets.
Beneficial effect
Due to the adoption of the above technical solution, compared with prior art, the present invention having the following advantages that and actively imitating
Fruit: the present invention not only has the function of testing the speed, ranging, angle measurement, the also function with target following, target classification and Data Post
Can, the present invention passes through the type for judging track, whether in orbit to judge target using different methods according to classification of track, together
When increase the fence function of filtering out track both sides, so as to accurately determine target position, classification and motion state, and
It is communicated with vehicle, and then realizes the function of active forewarning, carries out decision for vehicle sensory perceptual system.
Detailed description of the invention
Fig. 1 is the system block diagram of embodiment of the present invention;
Fig. 2 is the signal processing flow figure in embodiment of the present invention;
Fig. 3 is display module schematic diagram in embodiment of the present invention;
Fig. 4 is data management module schematic diagram in embodiment of the present invention;
Fig. 5 is alarm module schematic diagram in embodiment of the present invention;
Fig. 6 is rectilinear orbit flow chart of data processing figure in embodiment of the present invention;
Fig. 7 is object detection process flow chart in embodiment of the present invention.
Specific embodiment
Present invention will be further explained below with reference to specific examples.It should be understood that these embodiments are merely to illustrate the present invention
Rather than it limits the scope of the invention.In addition, it should also be understood that, after reading the content taught by the present invention, those skilled in the art
Member can make various changes or modifications the present invention, and such equivalent forms equally fall within the application the appended claims and limited
Range.
Embodiments of the present invention are related to a kind of vehicle-mounted millimeter wave radar installations, as shown in Figure 1, comprising power module, penetrating
Frequency front end, signal processing module, communication module, data management module, fault alarm module and display module.Wherein, power supply mould
Block mainly provides corresponding work for modules such as signal processing module, radio-frequency front-end, display module, communication module and data managements
Make power supply;Radio-frequency front-end is mainly the electromagnetic wave for emitting certain frequency, and echo letter is generated when the electromagnetic wave and target surface contact
Number, receiving antenna, which receives the echo-signal and then passes through signal conditioning circuit, is converted to intermediate-freuqncy signal, and intermediate-freuqncy signal passes through low pass
Filter circuit (filtering out the target intermediate-freuqncy signal other than radar detection maximum distance) enters in analog-digital converter ADC;Maximum measure distance
Range threshold is related with the frequency of intermediate-freuqncy signal, and in order to avoid the generation of range ambiguity phenomenon, the present apparatus carries out intermediate-freuqncy signal
The processing of low-pass filter makes the frequency of its output signal within the half of ADC sample frequency.Signal processing module is mainly
Fast Fourier Transform (FFT) (FFT), CFAR detection (CFAR), digital beam froming are successively carried out to the collected data of ADC
(DBF), target following, target detection (RANSAC) and target classification (clustering).Display module includes that target range is aobvious
Show, target velocity is shown, target angle is shown, target category is shown, dbjective state is shown, shows setting options;Fault alarm
Module includes audible alarm, light warning, noise elimination button and reset button;Data management includes data storage, data-printing, number
According to analysis and data dump.
For signal processing module as shown in Fig. 2, the FFT unit in the module is mainly handled adc data, FFT is mono-
Member realizes the conversion of time domain to frequency domain, obtains the range information of target by one-dimensional Fast Fourier Transform (FFT), then fast by two dimension
Fast Fourier transformation obtains the velocity information of target.Spectral leakage in order to prevent also needs to carry out windowing operation, and adding window has Hamming
The methods of window, Chebyshev window, Blackman window.Present embodiment uses Chebyshev window.
Contain certain false-alarm and clutter information in the velocity information and range information of the target that FFT unit is calculated,
CFAR processing need to be carried out to the result of FFT unit, be the information of real goal by the result that CFAR is handled;This embodiment party
In formula, in order to reduce shadowing effect, range information is handled using OS-CFAR algorithm (Ordered Statistic class CFAR detection);
To reduce operand, velocity information is handled using CA-CFAR algorithm (cell-average method CFAR detection).
It, need to be to target in order to accurately obtain the specific location of target when the speed of two or more targets is with as
Carry out angle measurement.DBF, which is mainly calculated, finds out target with respect to radar angular position, and then the multiple targets of difference and determination are multiple
The specific location of target.
Since the road conditions of vehicle driving are complex, data associating unit joined after DBF, in order to further solve
Whether certainly velocity ambiguity problem, the data associating unit occur velocity ambiguity phenomenon in the target velocity that uncertain radar returns
In the case where, according to the principle that the distance of target cannot be mutated, it is accurate to be carried out calculating according to the formula of speed and distance
Target velocity.Assuming that the position where t moment target is Rt, the position at t-1 moment is Rt-1, VtIt is returned for the radar of t moment
Speed, Vt-1For the speed that the radar at t-1 moment returns, Δ t is the time difference.The formula of speed and distance is Rt=Rt-1+
Vt-1* Δ t, it is known that For the true velocity at the target t-1 moment being calculated according to target position, because
Radar refresh rate is Millisecond, and Δ t is sufficiently small, it is believed that the speed at target t-1 moment and the speed of t moment are essentially identical, this
When compareThe V returned with radartDifference it is whether very big, when velocity ambiguity difference be can be very big.If target value
VtWith Vt-1Difference is very big, then it is assumed that for false-alarm or produces velocity ambiguity, need to reject the target.
The main purpose of the object detection unit of the present apparatus is the fence for filtering out track both sides, and the target in extraction fence is simultaneously
It exports to user, detection process is as follows: the target detection of rail traffic, including determine that track is straight line or bend, according to
Whether in orbit the difference of rectilinear orbit and bend judges target using different methods, and then exports effective target.
The object detection method of the vehicle-mounted millimeter wave radar installations of present embodiment, comprising the following steps: (1) radar emission
Millimeter-wave signal, and the echo-signal of target is received, and pre-process to the echo-signal of target;(2) after according to pretreatment
Signal acquisition target velocity information, range information and location information;(3) according to velocity information, the distance of obtained target
Information and location information filter out the fence on track both sides, and extract the target in fence as effective target.It is specific as follows:
Judge in the whether in-orbit road fence of target, detections of radar to target information have distance, speed and angle, away from
(R can be used from, speed and angle informationx、Ry, Vx、Vy, Ax、Ay) indicate.It is communicated, is introduced whole with vehicle by the CAN interface of radar
The speed of vehicle compares the speed of the speed of detections of radar to target and vehicle, and then can determine whether to obtain detections of radar
Whether target is static target.When the target velocity of detections of radar is equal and contrary with vehicle speed, then the target is judged to
For static target (static relative to the earth).For vehicle when along rail running, the fence spacing on track both sides is fixed value r, mesh
Mark present position is Rx(after target location vector decomposes, the distance of x-axis direction).When detection, first determine whether target is opposite
It is greatly static, if not static, judge target whether Rx< r/2 exports target if it is, being effective target, if
It is not exported for invalid targets then.If static target, then by RANSAC algorithm (random sampling unification algorism)
Reason, process flow are as follows: firstly, arbitrarily finding out two o'clock (Rx1、Ry1)、(Rx2、Ry2) two o'clock building y=ax+b linear equation;Its
It is secondary, all static targets are traversed, one is found out and covers the straight line of most static targets as fence.Only it is to speed herein
0 target is traversed, it is possible to reduce traversal time, its purpose is to judge to be currently at operation on rectilinear orbit still to exist
It is run on bend track and whether static target is fence.One RANSAC threshold value α is set at this time, if in this straight line
On number of targets β be greater than α, it is determined that track at this time be rectilinear orbit, straight line y=ax+b can be found out at this time and be determined as grid
Column, and filtered out and do not shown.When the static target of detection is in fence, then the target is exported to user, if static target
Outside fence, then filters out and do not show.As shown in fig. 7, straight line when driving, the criterion of effective target is, target speed
Spend V=0, Rx=r/2, then target is fence, does not export on the straight line for meeting the detection of RANSAC algorithm;If Rx< r/2 then mesh
It is marked in fence, then target is effective target, exports the target.If the number of targets β on this straight line is less than α, it is determined that this
When enter bend track, the criterion of effective target is target velocity V=0, Rx=r/2, then target is meeting RANSAC
On the straight line of algorithm detection, it is fence, does not export;If RxThen target is in fence by < r/2, then target is effective target, output
The target.
As shown in fig. 6, only first can judge static target, all static targets are traversed according to RANSAC, obtain covering quiet
The only most straight line fence of target, and determine track for after rectilinear orbit, it is not necessary to the R of moving targetx< r/2 judged,
According to straight line fence, in fence movement or static target show, as effective target, to fence outside movement or
Person's static target is not shown, as invalid targets, and is filtered out fence, is prevented false alarm.I.e. when detecting target, no matter mesh
It is designated as static target or moving target, if the target, not in fence, which does not export to user, i.e., not in display screen
Upper output;If target in fence, exports the target.The R by comparing target is not needed at this timexWith the pass of r/2
System is to judge whether moving target or static target are effective target.
The present invention estimates the parameter of mathematical model by the way of iteration from one group comprising discrete being observed in data,
The target static relative to speed is traversed, to reduce the number of iterations and then promote calculating speed.When detecting opposite thunder
When up to static and target in the same direction, RANSAC algorithm can be used, the target is handled, and then can be to user
Fence is filtered out to prevent false alarm.Since the guardrail/fence on field of track traffic both sides is generally metal material, detections of radar is arrived
After fence/guardrail, easily there is false judgment (false alarm).The inventors found that the fence spacing on track both sides is
Fixed value, therefore after first can judging fence, and then judge whether target in fence determines whether target is effective, that is, helps
Target is judged whether in fence/guardrail, if being judged to effective target in guardrail/fence and exporting to vehicle.If not
In fence/guardrail, invalid targets are judged to, are not exported to vehicle.
In order to the specific location of real-time detection target, tracking processing need to be carried out to target.Present apparatus track algorithm packet
Contain but be not limited to expanded Kalman filtration algorithm (Extended Kalman Filter, EKF), the algorithm to the distance of target,
Speed, angle carry out estimation tracking prediction, so can real-time tracking target operation conditions.
The present apparatus also includes target classification unit, can be according to the Radar Cross Section (Radar of the target of detection
Cross-Section, RCS) size distinguish the targets such as pedestrian, bicycle, motorcycle, car, truck, and then showing
It carries out being displayed for vehicle progress decision in module.The device is not limited only to realize the classification of target by RCS, can also construct
It the library of the targets such as pedestrian, bicycle, motorcycle, car, truck can be by the method tabled look-up to target when detecting target
It is accurately identified;The progress of the complicated algorithms such as deep learning, machine learning and neural network can also be used in the method for target identification
The identification of the targets such as people, bicycle, motorcycle, car, truck.
Display module is as shown in figure 3, the module distinguishes speed, distance, angle, classification and the state of displaying target.Target
State include stationary state, far from state, close to state;Show that quantity, setting that setting options are setting displaying target are aobvious
Show the distance range for detecting target, the angular range that displaying target is set, the status display that displaying target is set, setting display
The information such as the classification of target.
Data management module as shown in figure 4, in the module data storage be the FFT saved in adc data and signal processing
(Fourier transformation), CFAR (CFAR detection), tracking (tracking target), the data of clustering (target classification);
Data-printing is mainly the data of FFT, CFAR, tracking, clustering in difference print data memory module;Data
Removing is the data removed in memory module, in order to which more data are stored;Data analysis be analysis adc data, FFT,
The data of CFAR, tracking, RANSAC, clustering.
Alarm module as shown in figure 5, the module can alarm to the failure of modules, and by audible alarm, no
With color light, alert notice user is checked.According to different fault signatures, can export on a display screen error code for
User is with reference to simultaneously fault location.The device further includes noise elimination button and reset button, after pressing noise elimination button, sound report
It is alert to eliminate;After pressing reset button, system restarting.The module is not limited only to fault alarm, can also be to closer away from radar
Target alarm, to prompt user to carry out judging whether deceleration/parking.
Communication module in the device is connect by communication interface with full-vehicle control module, and in real time the target detected
Data transmission is to full-vehicle control module.Full-vehicle control module can also be communicated by communication interface with radar installations.In addition, logical
The modes such as controller local area network (ControllerAreaNetwork, CAN) communication, network communication can be used in letter mode.
It is noted that the device has function for monitoring, and self monitoring, i.e., automatic therefore fail-safe detection design.It can
Automatic fault detection, and error code is exported automatically, and then obtains fault message.The device is not limited to use in tramcar field,
Apply also for the field of track traffic such as railway, subway.The working frequency of the device is not limited to 77GHz, can also be using 24GHz etc.
The millimetre-wave radar of frequency range.The power module of the device give respectively display module, alarm module, data processing module, at signal
Manage module and communication module power supply.Power module can be communicated with central processing unit, according to the working condition of each module into
And the output of power module is controlled, and then reduce the power consumption of the device.The waveform that the device uses is not limited to sawtooth wave, can also adopt
With the combination of triangular wave, frequency shift keying (FSK), phase-shift keying (PSK) (PSK) or multiple waveforms.
Claims (9)
1. a kind of vehicle-mounted millimeter wave radar installations, including RF front-end module, power module and signal processing module, the power supply
Module is that RF front-end module and signal processing module are powered, which is characterized in that the RF front-end module, for emitting
Millimeter-wave signal receives the echo-signal of target, and pre-processes to the echo-signal of target, the signal processing module packet
It includes: distance and speed acquiring unit, for the velocity information and range information according to pretreated signal acquisition target;Position
Acquiring unit, for the location information according to pretreated signal acquisition target;Object detection unit is obtained for basis
Velocity information, range information and the location information determination of target are bend or straight way, and filter out the fence on track both sides, are extracted
Target in fence is as effective target.
2. vehicle-mounted millimeter wave radar installations according to claim 1, which is characterized in that be arranged after the position acquisition unit
There is data associating unit, the data associating unit solves asking for velocity ambiguity by the principle that target range cannot be mutated
Topic.
3. vehicle-mounted millimeter wave radar installations according to claim 1, which is characterized in that the object detection unit will test
To the velocity information of target and the velocity information of vehicle compare, judgement show whether the target detected is static mesh
Mark determines that current orbit is rectilinear orbit or bend track by RANSAC algorithm, and determine fence if it is static target
Position.
4. vehicle-mounted millimeter wave radar installations according to claim 3, which is characterized in that if not static target, then sentence
The abscissa value of disconnected target whether the fixed value between the fence less than 1/2, be if it is effective target, be otherwise invalid mesh
Mark.
5. vehicle-mounted millimeter wave radar installations according to claim 3, which is characterized in that after determining current orbit, according to
Fixed value between fence of the abscissa value of the location information of target equal to 1/2 determines column location, if the abscissa of target
The fixed value being worth between the fence less than 1/2, then be effective target.
6. vehicle-mounted millimeter wave radar installations according to claim 3, which is characterized in that the RANSAC algorithm process process
Are as follows: two o'clock building linear equation is arbitrarily found out, all static targets is traversed, threshold value is set, if on the straight line found
When the number of static target is greater than threshold value, determines that track at this time is rectilinear orbit, be otherwise bend track, and one is covered
The straight line of most static targets is covered as fence.
7. vehicle-mounted millimeter wave radar installations according to claim 6, which is characterized in that according to the obtained fence, if
Target, which is shown, to be located in fence, then the target is effective target, is located at outside fence if target is shown, is considered as invalid targets.
8. a kind of object detection method of vehicle-mounted millimeter wave radar installations, which comprises the following steps:
(1) radar emission electromagnetic signal, and the echo-signal of target is received, and pre-process to the echo-signal of target;
(2) according to velocity information, range information and the location information of pretreated signal acquisition target;
(3) it is bend or straight way according to velocity information, range information and the location information determination of obtained target, and filters out rail
The fence on road both sides extracts the target in fence as effective target.
9. the object detection method of vehicle-mounted millimeter wave radar installations according to claim 8, which is characterized in that target acquisition
Specific steps are as follows:
(3.1) velocity information of target and the velocity information of vehicle that detect in step (3) are compared, judgement obtains inspection
Whether the target measured is static target;
(3.2) if it is static target, then determine that current orbit is rectilinear orbit or bend track by RANSAC algorithm, and
Determine column location;
Fence is determined according to the fixed value between fence of the abscissa value of target position information equal to 1/2 after determining current orbit
Position, if the fixed value between fence of the abscissa value of target less than 1/2, for effective target;
(3.2.1) RANSAC algorithm process process are as follows: two o'clock building linear equation is arbitrarily found out, all static targets are traversed,
Threshold value is set, if the number of the static target on the straight line found is greater than threshold value, determines that track at this time is straight line
Otherwise track is bend track, and using the straight line of a cover-most static target as fence;
(3.3) if not static target, then judge target abscissa value whether the fixed value between the fence less than 1/2, such as
Fruit is then for effective nonstatic target, is otherwise invalid targets.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910355156.4A CN110058239B (en) | 2019-04-29 | 2019-04-29 | Vehicle-mounted millimeter wave radar device and target detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910355156.4A CN110058239B (en) | 2019-04-29 | 2019-04-29 | Vehicle-mounted millimeter wave radar device and target detection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110058239A true CN110058239A (en) | 2019-07-26 |
CN110058239B CN110058239B (en) | 2023-08-01 |
Family
ID=67321551
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910355156.4A Active CN110058239B (en) | 2019-04-29 | 2019-04-29 | Vehicle-mounted millimeter wave radar device and target detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110058239B (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111198370A (en) * | 2020-01-02 | 2020-05-26 | 北京百度网讯科技有限公司 | Millimeter wave radar background detection method and device, electronic equipment and storage medium |
CN111289980A (en) * | 2020-03-06 | 2020-06-16 | 成都纳雷科技有限公司 | Roadside stationary object detection method and system based on vehicle-mounted millimeter wave radar |
CN111323757A (en) * | 2019-12-30 | 2020-06-23 | 北京海兰信数据科技股份有限公司 | Target detection method and device for marine radar |
CN112130136A (en) * | 2020-09-11 | 2020-12-25 | 中国重汽集团济南动力有限公司 | Traffic target comprehensive sensing system and method |
CN112198503A (en) * | 2020-10-16 | 2021-01-08 | 无锡威孚高科技集团股份有限公司 | Target track prediction optimization method and device and radar system |
CN112233416A (en) * | 2020-09-17 | 2021-01-15 | 北京聚利科技有限公司 | Traffic flow detection method and device |
CN112313539A (en) * | 2019-11-26 | 2021-02-02 | 深圳市大疆创新科技有限公司 | Guardrail detection method and equipment, storage medium and movable platform |
WO2021082148A1 (en) * | 2019-10-31 | 2021-05-06 | 惠州市德赛西威智能交通技术研究院有限公司 | Target detection method for vehicle-mounted millimeter wave radar, and vehicle-mounted radar system therefor |
CN112799044A (en) * | 2020-12-30 | 2021-05-14 | 南京市德赛西威汽车电子有限公司 | BSD radar-based simple and efficient railing identification method |
CN113167886A (en) * | 2021-03-02 | 2021-07-23 | 华为技术有限公司 | Target detection method and device |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07309238A (en) * | 1994-05-17 | 1995-11-28 | Ishikawajima Harima Heavy Ind Co Ltd | Rolling stock having safety confirmation function |
CN1152958A (en) * | 1994-07-18 | 1997-06-25 | 西门子公司 | Arrangement for detecting objects in a region to be monitored |
CN103235310A (en) * | 2013-03-26 | 2013-08-07 | 北京理工雷科电子信息技术有限公司 | Vehicular millimeter-wave train collision avoidance radar system |
CN106458237A (en) * | 2014-05-20 | 2017-02-22 | 株式会社日立国际电气 | Wireless communication system, wireless communication device, wireless communication method, movable fence control system, communication device, and movable fence device |
CN107064952A (en) * | 2017-04-21 | 2017-08-18 | 中国科学院半导体研究所 | Railway track barrier monitors the device of identification automatically |
CN109164443A (en) * | 2018-08-27 | 2019-01-08 | 南京微达电子科技有限公司 | Rail track foreign matter detecting method and system based on radar and image analysis |
CN109684921A (en) * | 2018-11-20 | 2019-04-26 | 吉林大学 | A kind of road edge identification and tracking based on three-dimensional laser radar |
-
2019
- 2019-04-29 CN CN201910355156.4A patent/CN110058239B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07309238A (en) * | 1994-05-17 | 1995-11-28 | Ishikawajima Harima Heavy Ind Co Ltd | Rolling stock having safety confirmation function |
CN1152958A (en) * | 1994-07-18 | 1997-06-25 | 西门子公司 | Arrangement for detecting objects in a region to be monitored |
CN103235310A (en) * | 2013-03-26 | 2013-08-07 | 北京理工雷科电子信息技术有限公司 | Vehicular millimeter-wave train collision avoidance radar system |
CN106458237A (en) * | 2014-05-20 | 2017-02-22 | 株式会社日立国际电气 | Wireless communication system, wireless communication device, wireless communication method, movable fence control system, communication device, and movable fence device |
CN107064952A (en) * | 2017-04-21 | 2017-08-18 | 中国科学院半导体研究所 | Railway track barrier monitors the device of identification automatically |
CN109164443A (en) * | 2018-08-27 | 2019-01-08 | 南京微达电子科技有限公司 | Rail track foreign matter detecting method and system based on radar and image analysis |
CN109684921A (en) * | 2018-11-20 | 2019-04-26 | 吉林大学 | A kind of road edge identification and tracking based on three-dimensional laser radar |
Non-Patent Citations (5)
Title |
---|
NORIYUKI SHINODA 等: "FUNDAMENTAL EXPERIMENT FOR UTILIZING LIDAR SENSOR FOR RAILWAY", 《ENVIRONMENTAL SCIENCE》 * |
曲越: "城轨列车非接触式障碍物检测系统的研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
武永斌 等: "机载 LiDAR铁路测绘关键技术及应用", 《测绘通报》 * |
翟光耀: "基于毫米波雷达与机器视觉信息融合的有轨电车障碍物检测", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
陆德彪 等: "基于深度数据的车辆目标检测与跟踪方法", 《交通运输系统工程与信息》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021082148A1 (en) * | 2019-10-31 | 2021-05-06 | 惠州市德赛西威智能交通技术研究院有限公司 | Target detection method for vehicle-mounted millimeter wave radar, and vehicle-mounted radar system therefor |
CN112313539B (en) * | 2019-11-26 | 2024-03-15 | 深圳市大疆创新科技有限公司 | Guardrail detection method and equipment, storage medium and movable platform |
CN112313539A (en) * | 2019-11-26 | 2021-02-02 | 深圳市大疆创新科技有限公司 | Guardrail detection method and equipment, storage medium and movable platform |
CN111323757A (en) * | 2019-12-30 | 2020-06-23 | 北京海兰信数据科技股份有限公司 | Target detection method and device for marine radar |
CN111323757B (en) * | 2019-12-30 | 2022-04-05 | 北京海兰信数据科技股份有限公司 | Target detection method and device for marine radar |
CN111198370A (en) * | 2020-01-02 | 2020-05-26 | 北京百度网讯科技有限公司 | Millimeter wave radar background detection method and device, electronic equipment and storage medium |
CN111289980A (en) * | 2020-03-06 | 2020-06-16 | 成都纳雷科技有限公司 | Roadside stationary object detection method and system based on vehicle-mounted millimeter wave radar |
CN111289980B (en) * | 2020-03-06 | 2022-03-08 | 成都纳雷科技有限公司 | Roadside stationary object detection method and system based on vehicle-mounted millimeter wave radar |
CN112130136A (en) * | 2020-09-11 | 2020-12-25 | 中国重汽集团济南动力有限公司 | Traffic target comprehensive sensing system and method |
CN112130136B (en) * | 2020-09-11 | 2024-04-12 | 中国重汽集团济南动力有限公司 | Comprehensive perception system and method for traffic targets |
CN112233416A (en) * | 2020-09-17 | 2021-01-15 | 北京聚利科技有限公司 | Traffic flow detection method and device |
CN112198503A (en) * | 2020-10-16 | 2021-01-08 | 无锡威孚高科技集团股份有限公司 | Target track prediction optimization method and device and radar system |
CN112799044A (en) * | 2020-12-30 | 2021-05-14 | 南京市德赛西威汽车电子有限公司 | BSD radar-based simple and efficient railing identification method |
WO2022183369A1 (en) * | 2021-03-02 | 2022-09-09 | 华为技术有限公司 | Target detection method and apparatus |
CN113167886B (en) * | 2021-03-02 | 2022-05-31 | 华为技术有限公司 | Target detection method and device |
CN113167886A (en) * | 2021-03-02 | 2021-07-23 | 华为技术有限公司 | Target detection method and device |
Also Published As
Publication number | Publication date |
---|---|
CN110058239B (en) | 2023-08-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110058239A (en) | A kind of vehicle-mounted millimeter wave radar installations and object detection method | |
Urazghildiiev et al. | Vehicle classification based on the radar measurement of height profiles | |
EP1681583B1 (en) | Vehicle radar process | |
CN105261215B (en) | Intelligent transportation behavior cognitive method and system based on microwave | |
CN110531336A (en) | A kind of object detection recognition methods and system | |
CN106405556B (en) | Vehicle target information detection identifying system and its signal processing method | |
CN110832340B (en) | System for detecting moving objects | |
CN103339525B (en) | Method and device for monitoring variations in terrain | |
CN109814080B (en) | Millimeter wave radar target tracking and confidence coefficient algorithm and device thereof | |
CN109324317A (en) | Millimetre-wave radar system and its positioning-speed-measuring method | |
Diewald et al. | Radar-interference-based bridge identification for collision avoidance systems | |
CN105334514A (en) | Tramcar radar video compound early warning crashproof system and method | |
EP3740784B1 (en) | Method and device for detecting critical transverse movements | |
EP3349038A1 (en) | Method to classify objects in a vehicle environment based on radar detections | |
Hyun et al. | Design and development of automotive blind spot detection radar system based on ROI pre-processing scheme | |
CN110161505A (en) | One kind being based on millimetre-wave radar rear anti-crash method for early warning | |
EP4343369A2 (en) | Interference signal detection method and apparatus, and integrated circuit, radio device and terminal | |
CN109932695A (en) | A kind of method and device improving object recognition speed | |
CN102540163A (en) | Helicopter classifying and identifying method based on conventional coherent radar | |
CN104977566B (en) | Signal processing method and device for a frequency modulated continuous wave radar sensing system | |
CN210572736U (en) | Low, small and slow target automatic detection, tracking and identification system | |
CN115061113A (en) | Target detection model training method and device for radar and storage medium | |
JPH0883400A (en) | Obstacle discrimination system for vehicle | |
CN111406224B (en) | Target credibility determining method, target identifying method, system, vehicle and storage medium | |
CN112098969B (en) | Target detection and early warning optimization method for millimeter wave large vehicle blind area radar |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |