WO2021097880A1 - 一种基于车载角雷达的镜像目标去除方法 - Google Patents

一种基于车载角雷达的镜像目标去除方法 Download PDF

Info

Publication number
WO2021097880A1
WO2021097880A1 PCT/CN2019/121202 CN2019121202W WO2021097880A1 WO 2021097880 A1 WO2021097880 A1 WO 2021097880A1 CN 2019121202 W CN2019121202 W CN 2019121202W WO 2021097880 A1 WO2021097880 A1 WO 2021097880A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
radar
mirror
vehicle
reflecting surface
Prior art date
Application number
PCT/CN2019/121202
Other languages
English (en)
French (fr)
Inventor
阮洪宁
陈丽
唐恺
伍小军
Original Assignee
惠州市德赛西威汽车电子股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 惠州市德赛西威汽车电子股份有限公司 filed Critical 惠州市德赛西威汽车电子股份有限公司
Publication of WO2021097880A1 publication Critical patent/WO2021097880A1/zh

Links

Images

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
    • 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
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/023—Interference mitigation, e.g. reducing or avoiding non-intentional interference with other HF-transmitters, base station transmitters for mobile communication or other radar systems, e.g. using electro-magnetic interference [EMI] reduction techniques

Definitions

  • the invention relates to the technical field of radar target detection, in particular to a mirror target removal method based on a vehicle-mounted angle radar.
  • Millimeter wave radars are usually installed in the four corners of the front and rear of the car, forming the four corner radars at the front and rear of the car.
  • the car uses four-corner radar to perceive the surrounding environment of the car body, especially the targets in the blind area of the car's perspective, and analyze the potential hazards that may occur to realize blind spot detection, lane change assistance, rear cross warning, rear collision warning, door opening warning, etc. Function to provide important information for the driver’s decision to improve the safety of assisted and autonomous driving.
  • the vehicle-mounted millimeter-wave radar has relatively sensitive detection capabilities for targets, the coverage of a single radar is limited, and it is still not able to fully cope with the complex actual scenes.
  • the "target” judged by the radar based on the echo generated by the multipath phenomenon is not actually at the position of the real target, but in other directions (or A false mirror target at a greater radial distance or other angles). If this mirrored target appears in a certain functional alarm area of a radar on a certain side, and no screening measures are taken, then the radar may make a wrong judgment and trigger unnecessary alarms.
  • the rear cross warning function needs to check whether there is a laterally driving vehicle in the reverse trajectory behind the vehicle.
  • the corner radar set on the right side of the vehicle will detect it
  • the real target vehicle enters the rear cross warning zone and alarms.
  • the electromagnetic waves emitted by the left radar are repeatedly between the stationary car and the real target.
  • the left radar After reflection, it will be received by the left radar again, causing the left radar to also detect that there is a "target" on the left side of the vehicle toward the vehicle, that is, the mirror image target. Therefore, the left radar will also send out an alarm at the same time, which makes the radar alarm system mistakenly believe that there are targets driving on both sides of the vehicle and make a wrong judgment.
  • the "target" detected by the left radar of the vehicle does not exist, it is a mirrored target generated by the real target in the field of view of the left radar. The existence of the mirror target will affect the accuracy and confidence of the radar detection, and even provide the driver with wrong information, misleading the driver's decision-making, and is not conducive to improving the safety of assisted and autonomous driving.
  • the present invention provides a method for removing a mirror image target based on a vehicle-mounted angle radar. Based on a radar set at the four corners of the vehicle, the method includes the following steps:
  • the stationary reflector is detected by the first radar, and the stationary reflector is located on the same side of the vehicle as the first radar, and the reflecting surface of the stationary reflector relative to the first radar is analyzed;
  • the second target is considered to be a mirror image of the first target, and the second target is ignored; otherwise, it is regarded as a non-mirrored target.
  • the detecting the stationary reflector includes the following steps:
  • the target object is regarded as a static reflector; otherwise, the target object is ignored.
  • the reflecting surface refers to the ability to reflect the transmitted signal of the first radar to the first target, and reflect the signal reflected by the first target to the first radar again.
  • the first radar is arranged on the side of the vehicle corresponding to the static reflector, the first radar is used to detect the static reflector and the second target; the second radar is arranged on the vehicle corresponding to the first target On one side, the second radar is used to detect the first target.
  • the step of calculating the symmetrical mirror area of the first target with respect to the reflecting surface includes:
  • the reflecting surface as a reference surface, calculate the mirror image area in the field of view of the first radar, and the mirror image area and the first target are spatially mirror-symmetrical with respect to the reflecting surface.
  • the calculation of the mirror area includes:
  • the movement speed and movement continuity of the second target and the first target are used to determine whether the movement characteristics of the second target and the first target are symmetrical with respect to the reflecting surface.
  • the judging whether the motion characteristics of the second target and the first target are symmetrical with respect to the reflecting surface, if so, the second target is considered to be the mirror image of the first target, and the second target is ignored; otherwise, the step is regarded as no mirror target, include:
  • first moving speed and the second moving speed are mirror-symmetrical with respect to the reflecting surface. If they are, the second target is considered to be a mirror image of the first target, and the second target is ignored; otherwise, it is regarded as a non-mirroring target.
  • a movement continuity judging step is further included, and the movement continuity judging step includes:
  • the judging whether the first moving speed and the second moving speed are mirror-symmetrical about the reflecting surface includes the following sub-steps:
  • the first moving speed and the second moving speed are considered to be mirror-symmetrical with respect to the reflecting surface; otherwise, the first moving speed is regarded as the first moving speed and the second moving speed.
  • the moving speed and the second moving speed are asymmetric with respect to the reflecting surface.
  • the present invention discloses a method for removing a mirror target based on a vehicle angle radar.
  • the method utilizes the special relationship between the mirror target and the real target to quickly and effectively identify the mirror target.
  • the method is simple and novel. It is easy to implement, which is conducive to improving the vehicle-mounted millimeter-wave radar's ability to perceive targets and the environment in the blind area of the vehicle, improve the accuracy and reliability of radar detection, and provide the driver with accurate and effective reference information, thereby improving the performance of assisted driving and automatic driving systems. safety.
  • the method can be realized only by relying on existing radars arranged in the four corners of the vehicle, there is no need to modify the vehicle hardware system and the algorithm of the existing radar target track detection, the application cost is low, and it is conducive to large-scale promotion.
  • Fig. 1 is a schematic flow chart of a method for removing a mirror target based on a vehicle-mounted angle radar according to the present invention.
  • FIG. 2 is a schematic diagram of the positional relationship among the vehicle, the stationary reflector, the first radar, the second radar, the first target, and the second target in Embodiment 1.
  • FIG. 3 is a schematic diagram of a coordinate system established by a vehicle in Embodiment 1.
  • FIG. 3 is a schematic diagram of a coordinate system established by a vehicle in Embodiment 1.
  • this embodiment provides a method for removing mirrored targets based on a vehicle-mounted corner radar.
  • the method is based on four radars arranged at the four corners of the vehicle, namely the left front corner radar (L1) and the left rear corner radar ( L2), right front corner radar (R1), and right rear corner radar (R2), they all have their own coverage.
  • the operating frequency of the radar in this embodiment includes but is not limited to 24 GHz and 77 GHz.
  • the mirror target removal method specifically includes the following steps:
  • the stationary reflector 2 is detected by the first radar 3.
  • the stationary reflector 2 and the first radar 3 are located on the same side of the vehicle, and the reflection surface of the stationary reflector 2 relative to the first radar 3 is analyzed.
  • static reflector 2 The detection of static reflector 2 is very important for the entire mirror target removal method. Whether there is a static reflector 2 directly determines whether the mirror target is likely to exist. If there is no static reflector 2 around the vehicle 1, the radar system can It is directly determined that no mirror target exists.
  • a stationary object near the vehicle such as the railing of the isolation belt or the wall around the tunnel, it may become a stationary reflector.
  • the vehicle When the vehicle is moving, as long as the objects around the vehicle and the vehicle and radar meet certain requirements. When the geometric relationship between the two, it can also produce a mirror target, which becomes a static reflector. .
  • the detection process of the stationary reflector it is first necessary to detect objects around the vehicle 1 through the on-board radar, and select objects whose distance from the vehicle 1 is less than the first threshold as the target object. Due to the close distance between the target object and the vehicle 1, the system will consider that the target object is likely to reflect the radar's transmitted signal, so the system will regard the target object as a key inspection object.
  • the vehicle-mounted radar can be used to detect the moving speed of the target object, and determine whether the moving speed of the target object is less than the second threshold, that is, whether the target object is in a stationary state or close to a stationary state. Only when the moving speed of the target object is less than the second threshold, will the target object be regarded as the stationary reflector 2; otherwise, the target object will be ignored and the detection and screening of other target objects will continue.
  • the speed of vehicle 1 is generally low, and may even be zero, while stationary objects located near vehicle 1
  • the speed is also relatively low, and may even be zero.
  • the vehicle-mounted radar is directly used to detect the distance between the stationary object and the vehicle 1, and when the distance between the stationary object and the vehicle 1 is less than the first threshold, the stationary object can be used as the target object.
  • the moving speed of the target object is further detected and analyzed to determine whether the moving speed of the target object is less than the second threshold value. Only when the moving speed of the target object is less than the second threshold value can the target object be regarded as a stationary reflector 2.
  • the moving speed of the target object can also be decomposed into X-axis speed and Y-axis speed, and the X-axis speed and Y-axis speed can be compared with the set threshold respectively.
  • Screen static reflector 2 That is, when both the X-axis speed and the Y-axis speed of the target object are less than the set threshold, the target object is regarded as the stationary reflector 2.
  • the vehicle-mounted radar closest to the stationary reflector 2 can be used as the first radar 3.
  • the pair of stationary reflectors 2 can be determined.
  • the reflective surface refers to a surface capable of reflecting the transmitted signal of the first radar 3 to the first target 5, and reflecting the signal reflected by the first target 5 to a certain surface of the first radar 3 again.
  • the first target 5 here refers to a real target that exists around the vehicle 1 and needs to be fed back to the driver in time, that is, a real target.
  • the stationary reflector 2 can reflect the transmitted signal of the first radar 3 to the first target 5, and reflect the signal reflected by the first target 5 to the first radar 3 again, the first target 5 is regarded as a real target, That is, the real target must be located on both sides of the vehicle 1 with the stationary reflector 2.
  • the first radar 3 and the second radar 4 refer to two radars arranged on the side of the vehicle 1 corresponding to the stationary reflector 2 and the first target 5.
  • the first radar 3 refers to a vehicle-mounted radar arranged adjacent to the stationary reflector 2, and the first radar 3 is arranged corresponding to the stationary reflector 2. That is, the first radar is arranged on the side of the vehicle corresponding to the stationary reflector, and the first radar is used to detect the second target.
  • the second radar 4 refers to a vehicle-mounted radar set close to the first target 5.
  • the second radar 4 is set corresponding to the first target 5.
  • the second radar 4 is used to detect the position information of the first target 5, that is, the second The radar is arranged on the side of the vehicle corresponding to the first target. Taking the RCTA scenario as an example, since the first target 5 is located on the right side of the vehicle 1, the first target 5 is detected by the second radar 4, and the first target 5 is not detected by the first radar 3.
  • the stationary reflector 2 is located on the left rear side of the vehicle, so the stationary reflector 2 is detected by the first radar 3.
  • the symmetrical mirror image area of the first target 5 with respect to the reflecting surface it is necessary to first establish a rectangular coordinate system with the position of the vehicle 1 as the origin, as shown in FIG. 3. And according to the rectangular coordinate system, the position coordinates of the first target 5 are calculated. At the same time, the coordinate position of the reflecting surface needs to be calculated. Then, with the reflecting surface as the reference surface, the mirror area in the field of view of the first radar can be calculated, and the mirror area and the first target 5 are mirror-symmetrical with respect to the reflecting surface.
  • the symmetric point of the first target 5 with respect to the reflecting surface is first calculated, and the symmetric point and the first target 5 are mirror symmetric with respect to the reflecting surface; and then the symmetric point is taken as the center.
  • the mirror area can also be calculated by other methods.
  • the calculated mirror area can be a rectangle, a square, or a triangle. As long as the area centered on the symmetry point can be used as the mirror area, there is no specific limitation here.
  • the first radar 3 is arranged on the side adjacent to the stationary reflector 2.
  • the first radar 3 can be used to detect the mirrored area to determine whether there is a second target 6 in the mirrored area. If there is a second target 6, the number of the second target 6 there's a few.
  • the stationary reflector 2 is arranged on the left rear side of the vehicle 1, and the first target 5 is arranged on the right rear side of the vehicle 1.
  • the position of the stationary reflector 2 (X B , Y B )
  • the position of the symmetry point can be calculated (X A ', Y A '), take the position of the symmetry point as the center and the preset length as the radius to calculate the mirror image area.
  • the first radar set at the left rear of the vehicle 1 is used to detect the mirror image area, and three targets P, Q and R are detected.
  • the radar system will default that these three targets have good spatial symmetry with the first target 5, so the radar system will save the three targets P, Q, and R as the second target, so as to In the subsequent steps, one by one screening is carried out.
  • the second target 6 Determine whether the motion characteristics of the second target 6 and the first target 5 are symmetrical with respect to the reflecting surface. If they are, the second target 6 is considered to be the mirror image of the first target 5, that is, the second target 6 is the mirror target, which is ignored at this time The second target 6; otherwise, it is regarded as a non-mirrored target.
  • the three targets P, Q, and R detected in step 104 are successively regarded as the second target 6, and their motion characteristics are examined one by one to detect whether the motion characteristics of each target and the first target 5 are symmetrical with respect to the reflection surface. If the movement characteristics of the target object and the movement characteristics of the first target 5 are symmetric with respect to the reflecting surface, the target object is regarded as a mirror image of the first target, and the ignoring process is performed.
  • the three aspects of the spatial position, moving speed and motion continuity of the second target 6 and the first target 5 are mainly used to determine whether the motion characteristics of the second target 6 and the first target 5 are related to the reflective surface. Weighed. Since the mirrored area itself and the first target are symmetrical with respect to the reflecting surface, the target located in the mirrored area must meet the requirements of spatial symmetry with the first target.
  • the moving speed of the second target 6 and the first target 5 can be directly used to determine whether the motion characteristics of the second target 6 and the first target 5 are symmetrical with respect to the reflecting surface.
  • the second radar 4 and the first radar 3 it is necessary to use the second radar 4 and the first radar 3 to obtain the first moving speed of the first target 5 and the second moving speed of the second target 6 respectively.
  • the second target 6 includes P, Q, and R. These three targets, that is, the first radar 3 needs to be used to obtain the second moving speeds of the three targets P, Q and R respectively. Then verify in turn whether the first moving speed and each second moving speed are mirror-symmetrical about the reflecting surface.
  • the second target 6 is considered to be the mirror image of the first target 5.
  • the absolute value of the speed can be directly used as the difference method to judge. Assuming that the target P is consistent with the moving speed of the first target 5, then the first target The difference between the absolute value of the moving speed of 5 and the absolute value of the moving speed of the target P must be smaller than the preset value.
  • other methods can also be used to verify whether the first moving speed and the second moving speed are mirror-symmetrical with respect to the reflecting surface.
  • the first movement speed and the second movement speed are first decomposed. That is, the first moving speed is decomposed first, and the X-axis moving speed of the first target 5 is calculated. Then the second moving speed is decomposed, and the X-axis moving speed of the second target 6 is calculated. Finally, it is judged whether the X-axis movement speed of the first target 5 and the X-axis movement speed of the second target 6 are mirror-symmetrical with respect to the reflecting surface. If so, the first moving speed and the second moving speed are considered to be mirror-symmetrical with respect to the reflecting surface; otherwise It is considered that the first moving speed and the second moving speed are asymmetrical with respect to the reflecting surface.
  • the first movement speed it is also possible to decompose the first movement speed to obtain the Y-axis movement speed of the first target 5; decompose the second movement speed to calculate the Y-axis movement speed of the second target 6. Then it is determined whether the Y-axis movement speed of the first target 5 and the Y-axis movement speed of the second target 6 are mirror-symmetrical with respect to the reflecting surface, so as to determine whether the first moving speed and the second moving speed are mirror-symmetrical with the reflecting surface.
  • the first moving speed and the second moving speed in the X axis can determine the mirror symmetry of the first moving speed and the second moving speed with respect to the mirror image symmetry of the reflecting surface, or you can use the first moving speed And the second moving speed in the Y axis to determine the sub-velocity, even in order to improve the accuracy, you can also use the first moving speed and the second moving speed in the X-axis sub-velocity and the Y-axis sub-velocity to determine at the same time
  • the mirror symmetry of the first moving speed and the second moving speed with respect to the reflecting surface is not limited here.
  • the step of judging the continuity of motion is further included, that is, the first target 5 and the second target 6 are continuously tracked to verify whether they are continuous in a segment.
  • the first moving speed of the first target 5 can maintain a symmetrical relationship with the second moving speed of the second target 6 with respect to the reflecting surface, so as to avoid accidents and improve the accuracy of screening mirrored targets.
  • the second target 6 can Think of it as a mirror image of the first target 5.
  • the second radar 4 needs to be used to continuously obtain multiple frames of radar detection characteristics about the first target 5, and the first radar 3 is used to continuously obtain multiple frames of radar detection characteristics on the second target 6 at the same time. That is, the movement states of the first target 5 and the second target 6 are detected multiple times by the second radar 4 and the first radar 3. Then calculate the first moving speed of the first target 5 in the radar detection characteristics corresponding to each frame, and calculate the second moving speed of the second target 6 in the radar detection characteristics corresponding to each frame to obtain the first target 5 and the second target 5 The moving speed of target 6 at different moments.
  • first movement speed and the second movement speed obtained at the same time are compared, and it is judged whether the first movement speed and the second movement speed obtained at each time are both mirror-symmetrical about the reflecting surface. If so, the second target is considered 6 It is the mirror image of the first target 5, and the second target 6 is ignored; otherwise, it is regarded as no mirror target.
  • the solution discussed in the present invention is aimed at the situation where the real target appears in the coverage area of one radar, while the stationary reflector and the mirror target appear in the coverage area of another radar.
  • the real target and the mirror target will appear in the rear radars on the left and right sides, namely R2 and L2, where the left rear radar is equivalent to the first radar 3, and the right rear radar is equivalent to the second radar 4.
  • the mirror target removal method is to combine the track points detected by the left rear radar and the right rear radar, and filter the information to filter out false mirror targets. Now take the specific scenario of RCTA as an example to explain the principle of the mirror target removal method.
  • the "target” here is actually the mirror target of the real target, which is false. Since the mirror target and the real target are in a mirror image relationship with respect to the reflecting surface, the real target and the mirror target have spatial geometric symmetry, speed consistency and the same movement continuity.
  • the real target, mirrored target, static reflector, vehicle, vehicle rear left radar and vehicle rear right radar are represented by points A, A', B, O, O 2 and O 1 respectively , The coordinate relationship obtained is shown in Figure 3.
  • the result of the introduction shows that the real target and the mirror target have a left-right symmetric relationship with the center axis BB' of the stationary reflector as the symmetry axis, that is, the real target and the mirror target first have perfect spatial geometric symmetry.
  • the direction of the stationary reflector 2 that is, the direction of the straight line BB′ is parallel to the direction of the vehicle 1.
  • it is transformed into a real target A driving towards the vehicle in the horizontal direction, that is, in the X direction.
  • the positions of the real target A in the own vehicle coordinate system are (X A1 , Y A ) and (X A2 , Y A ), respectively.
  • the positions of the mirror target A'at t 1 and t 2 are (X A'1 ,Y A' ) and (X A'2 ,Y A' ), respectively. From the geometric symmetry of the real target and the mirrored target in space, the following formula can be derived:
  • v xA and v xA' are the same.
  • the directions of these two speeds are opposite, that is, in the X-axis direction, the speeds of the real target and the mirrored target are the same, and both have Speed consistency.
  • the horizontal component of the radial velocity that is, the direction vector of the lateral velocity
  • the motion characteristics of the real target are constantly close to the vehicle, and the horizontal component of the radial velocity is the same.
  • the direction vector of the lateral speed is also positive, so the speed of the mirrored target and the real target are not only the same in magnitude, but the directions are both close to the vehicle.
  • the movement continuity of the real target and the mirror target is mainly manifested in the fact that as long as the mirror target and the real target are detected by the radar on the corresponding side, the characteristics of this spatial symmetry and speed consistency are in time, that is, several consecutive frames of radar detection. The characteristics can be maintained until the real target is not detected or exceeds the coverage area of the radar. So when there is a stationary reflector 2 and the tracks detected in the left and right radars meet the above three characteristics, then they are a pair of real targets and mirrored targets, which can be regarded as the target detected on the same side as the stationary reflector 2. It is the mirror target. But as long as any one of the characteristics is not satisfied, the detected target will not be a mirrored target.
  • a target is detected at a symmetrical position in a certain frame, and its velocity vector is also exactly the same as the target detected on the other side.
  • symmetry is found Or consistency cannot be continuously guaranteed, then the target cannot be a mirror target, so it will not be wrongly judged as a mirror target and lose the real target.
  • the mirror target removal method is also applicable to scenes similar to the RCTA scene, such as the lane change assistance scene, the door opening warning scene, and so on.
  • the above analysis logic can also be used to realize the judgment of the mirroring target, and the details will not be repeated again.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Radar Systems Or Details Thereof (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种基于车载角雷达的镜像目标去除方法，包括：通过第一雷达检测其附近是否存在静止物体，并分析出相对于第一雷达的反射面，利用第二雷达获取第一目标；计算出第一目标关于反射面对称的镜像区域；利用第一雷达在镜像区域内提取第二目标；判断第二目标和第一目标的运动特性是否关于反射面对称，若具有对称性，且对称性维持一段时间，则认为第二目标是第一目标的镜像，忽略第二目标；否则视为无镜像目标。利用了镜像目标和真实目标之间的特殊关系，实现了对镜像目标的有效鉴别，有利于改善车载毫米波雷达对汽车盲区内目标及环境的感知能力，提升辅助驾驶与自动驾驶系统的安全性。

Description

一种基于车载角雷达的镜像目标去除方法 技术领域
本发明涉及雷达目标检测技术领域，特别涉及一种基于车载角雷达的镜像目标去除方法。
背景技术
随着自动驾驶技术的发展，越来越多的车型开始配备毫米波雷达。毫米波雷达通常安装在汽车前后四个角落，构成汽车前后的四个角雷达。汽车通过四个角雷达对车身周边环境尤其是汽车视角盲区内的目标进行感知，对可能发生的潜在危险进行分析，以实现盲区检测、变道辅助、后方交叉预警、后碰撞预警、开门预警等功能，为驾驶员的决策提供重要信息，以提高辅助和自动驾驶的安全性。虽然车载毫米波雷达对目标具有较为敏锐的探测能力，但只靠单一雷达其覆盖范围有限，依然不能够完全应对复杂的实际场景，在实际应用过程中甚至会出现错误的判断，引起不必要的损失。其中一个常见的情况就是在雷达视野范围内，出现了除感兴趣的目标以外的其他“干扰”目标，比如其他运动或者静止目标。雷达发射的电磁波照射在感兴趣的目标，除了有一部分电磁波直接返回并被雷达接收，另外还有一部分电磁波在感兴趣的目标与“干扰”目标之间发生多次传播，导致多径现象产生。而对于多径现象产生的回波同样会被雷达所检测到，雷达根据多径现象产生的回波判断得到的“目标”其实并不是在真实目标的位置上，而是在其他方位上(要么更远的径向距离或者其他角度上)的一个虚假的镜像目标。如若这个镜像目标出现在某侧雷达的某个功能报警区，而又没有采取甄别的措施，那么雷达就可能做出错误的判断，触发不必要的报警。
例如当本车从停车位倒退出来时，后方交叉预警功能需要查看在本车后方倒车轨迹内是否存在横向开来的车辆。此时，如果本车的一边(以左边为例说明)停有一辆车并且其尾部突出一部分，当有目标车从本车的右后边横向驶来，设置在本车右边的角雷达会检测到真实目标车进入后方交叉预警区而报警。而与此同时，由于停放在本车左边的静止车的存在，当左雷达和真实目标之间的相对位置满足一定的几何关系时，左雷达发射的电磁波在静止车和真实目标之间多次反射后会重新被左雷达接收，导致左边雷达也会检测到有一个“目标”在本车的左边朝本车开来，即镜像目标。因而，左雷达也同时会发出报警，这让雷达报警系统错误地认为在本车的两边都有目标行驶过来，从而做出错误的判断。实际上，本车的左雷达检测到的这个“目标”并不存在，它是由真实目标在左雷达的视野范围内产生的镜像目标。镜像目标的存在会影响雷达检测的精度和置信度，甚至为驾驶员提供错误的信息，对驾驶员的决策形成误导，不利于改善辅助和自动驾驶的安全性。
发明内容
本发明为了解决上述技术问题，提供了一种基于车载角雷达的镜像目标去除方法，基于设置在车辆四个角落的雷达，包括如下步骤：
通过第一雷达检测静止反射体，静止反射体与第一雷达位于车辆的同一侧，分析出静止反射体相对于第一雷达的反射面；
利用第二雷达获取第一目标；
计算出第一目标关于反射面对称的镜像区域；
利用第一雷达在镜像区域内提取第二目标；
判断第二目标和第一目标的运动特性是否关于反射面对称，若是，则认为第二目标是第一目标的镜像，忽略第二目标；否则视为无镜像目标。
进一步的，所述检测静止反射体包括如下步骤：
对车辆周围的物体进行检测，筛选出与车辆距离小于第一阈值的物体作为目标物体；
判断目标物体的移动速度是否小于第二阈值，若是，则将目标物体视为静止反射体；否则忽略目标物体。
进一步的，所述反射面是指能够将第一雷达的发射信号反射至第一目标，并将第一目标反射回来的信号再次反射给第一雷达。
进一步的，所述第一雷达设置在车辆与静止反射体相对应一侧，所述第一雷达用于检测静反射体和第二目标；所述第二雷达设置在车辆与第一目标相对应一侧，所述第二雷达用于检测第一目标。
进一步的，所述计算出第一目标关于反射面对称的镜像区域的步骤，包括：
以车辆所在位置建立坐标系；
获取第一目标的位置坐标；
以反射面为参考面，计算在第一雷达的视野范围里的镜像区域，所述镜像区域与第一目标关于反射面在空间上呈现镜像对称。
进一步的，所述镜像区域的计算包括：
计算第一目标关于反射面的对称点，对称点与第一目标关于反射面镜像对称；
以对称点为中心，以预设长度为半径画圆，圆内区域即为镜像区域。
进一步的，利用第二目标和第一目标的移动速度以及运动连续性来判断第二目标和第一目标的运动特性是否关于反射面对称。
进一步的，所述判断第二目标和第一目标的运动特性是否关于反射面对称，若是，则认为第二目标是第一目标的镜像，忽略第二目标；否则视为无镜像目标步骤，包括：
获取第一目标的第一移动速度和第二目标的第二移动速度；
判断第一移动速度和第二移动速度是否关于反射面镜像对称，若是，则认为第二目标是第一目标的镜像，忽略第二目标；否则视为无镜像目标。
进一步的，在判断第一移动速度和第二移动速度关于反射面镜像对称之后，还包括运动连续性判断步骤，所述运动连续性判断步骤包括：
利用第二雷达连续获取多帧关于第一目标的雷达检测特性，同时利用第一雷达连续获取多帧关于第二目标的雷达检测特性；
计算出第一目标在对应的各帧雷达检测特性中的第一移动速度；
计算出第二目标在对应的各帧雷达检测特性中的第二移动速度；
将同一时刻获得的第一移动速度和第二移动速度进行比对，判断是否每一时刻获得的第一移动速度和第二移动速度均关于反射面镜像对称，若是，则认为第二目标是第一目标的镜像，忽略第二目标；否则视为无镜像目标。
进一步的，所述判断第一移动速度和第二移动速度是否关于反射面镜像对称包括如下子步骤：
对第一移动速速进行分解，计算第一目标的X轴向移动速度；
对第二移动速度进行分解，计算第二目标的X轴向移动速度；
判断第一目标的X轴向移动速度与第二目标的X轴向移动速度是否关于反射面镜像对称，若是，则认为第一移动速度和第二移动速度关于反射面镜像对称；否则认为第一移动速度和第二移动速度关于反射面不对称。
本发明所起到的有益技术效果如下：
与现有技术相比较，本发明公开了一种基于车载角雷达的镜像目标去除方法，利用方法利用镜像目标和真实目标之间的特殊关系，能够快速有效的鉴别出镜像目标，方法简单新颖，易于实现，利于改善车载毫米波雷达对汽车盲区内目标及环境的感知能力，提高雷达检测的准确性和可靠性，为驾驶员提供准确而有效的参考信息，从而提升辅助驾驶与自动驾驶系统的安全性。此外，由于该方法仅依赖现有的设置在车辆四个角落的雷达就可以实现，无需对车辆硬件系统及已有雷达目标航迹检测的算法进行修改，应用成本较低，利于大范围推广。
附图说明
图1为本发明基于车载角雷达的镜像目标去除方法的流程示意图。
图2为实施例1中车辆、静止反射体、第一雷达、第二雷达、第一目标及第二目标的位置关系示意图。
图3为实施例1中以车辆建立的坐标系示意图。
附图标记：
1-车辆，2-静止反射体，3-第一雷达，4-第二雷达，5-第一目标，6-第二目标。
附图仅用于示例性说明，不能理解为对本专利的限制；为了更好说明本实施例，附图某些部件会有省略、放大或缩小，并不代表实际产品的尺寸；对于本领域技术人员来说，附图中某些公知结构及其说明可能省略是可以理解的；相同或相似的标号对应相同或相似的部件；附图中描述位置关系的用语仅用于示例性说明，不能理解为对本专利的限制。
具体实施方式
下面结合附图对本发明的较佳实施例进行详细阐述，以使本发明的优点和特征更易被本领域技术人员理解，从而对本发明的保护范围作出更为清楚的界定。
实施例1：
如图1所示，本实施例提供了一种基于车载角雷达的镜像目标去除方法，该方法基于设置在车辆四个角落的四个雷达，即左前角雷达(L1)、左后角雷达(L2)、右前角雷达(R1)、和右后角雷达(R2)，它们均有自己的覆盖范围。本实施例中的雷达工作频率包括但不限于24GHz、77GHz。该镜像目标去除方法具体包括如下步骤：
101、通过第一雷达3检测静止反射体2，静止反射体2与第一雷达3位于车辆的同一侧，分析出静止反射体2相对于第一雷达3的反射面。
静止反射体2的检测对于整个镜像目标去除方法来讲是至关重要的，是否存在静止反射体2直接决定了镜像目标是否有可能存在，若车辆1周围无静止反射体2，则雷达系统可以直接判定无镜像目标存在。
本实施例中，如果车辆附近有静止物体，如隔离带的栏杆或者隧道周围的墙体等，均有可能成为静止反射体，而当车辆运动时，只要车辆周围的物体与车辆及雷达满足一定的几何关系时，也可以产生镜像目标，成为静止反射体。。具体在，在静止反射体的检测过程中，首先需要通过车载雷达对车辆1周围的物体进行检测，筛选出与车辆1距离小于第一阈值的物体作为目标物体。由于目标物体与车辆1的距离较近，系统会认为目标物体极有可能会对雷达的发射信号产生反射作用，因此系统会将目标物体作为重点检查物体。当确定了目标物体后，便可以利用车载雷达检测目标物体的移动速度，判断目标物体的移动速度是否小于第二阈值，即目标物体是否处于静止状态或接近静止状态。只有当目标物体的移动速度小于第二阈值时，才会将目标物体作为静止反射体2，否则忽略目标物体，继续对其他目标物体进行检测筛选。
以后方交叉预警场景(Rear Cross Travel Alert-RCTA)为例进行说明，如图2所示，在RCTA场景中，车辆1的速度一般较低，甚至可能为零，而位于车辆1附近的静止物体的 速度也相对较低，甚至也可能为零。此时直接利用车载雷达检测静止物体与车辆1之间的距离，当静止物体与车辆1之间的距离小于第一阈值时，可以将该静止物体作为目标物体。并对目标物体的移动速度做进一步检测分析，判断目标物体的移动速度是否小于第二阈值，只有目标物体的移动速度小于第二阈值才能将目标物体作为静止反射体2。
当然，在检测静止反射体2的过程中，也可以将目标物体的移动速度分解成X轴速度和Y轴速度，通过将X轴速度和Y轴速度分别与设定的阈值进行比对，来筛选静止反射体2。即当目标物体的X轴速度与Y轴速度均小于设定的阈值时，才将目标物体视为静止反射体2。
一旦确定了静止反射体2后，便可以将与静止反射体2最邻近的车载雷达作为第一雷达3，并根据第一雷达3与静止反射体2的相对位置，确定出静止反射体2对第一雷达3的静止反射面。本实施例中，所说的反射面是指够将第一雷达3的发射信号反射至第一目标5，并将第一目标5反射回来的信号再次反射给第一雷达3的某个表面。这里的第一目标5是指存在于车辆1周围且需要被及时反馈给驾驶员的真实存在的目标，即真实目标。
102、利用第二雷达4获取第一目标5。
由于静止反射体2能够将第一雷达3的发射信号反射至第一目标5，并将第一目标5反射回来的信号再次反射给第一雷达3，因此第一目标5作为真实存在的目标，即真实目标，其必然与静止反射体2位于车辆1的两侧，。而第一雷达3和第二雷达4则是指设置在车辆1与静止反射体2、第一目标5相对应一侧的两个雷达。通常第一雷达3是指与静止反射体2邻近设置的车载雷达，第一雷达3与静止反射体2相对应设置。即第一雷达设置在车辆与静止反射体相对应一侧，所述第一雷达用于检测第二目标。而第二雷达4则是指与第一目标5临近设置的车载雷达，第二雷达4与第一目标5相对应设置，第二雷达4用于检测第一目标5的位置信息，即第二雷达设置在车辆与第一目标相对应一侧。以RCTA场景为例，由于第一目标5位于车辆1的右侧，因而第一目标5是被第二雷达4检测到的，第一目标5并没有被第一雷达3检测到。而静止反射2位于车辆的左后侧，因此静止反射体2是被第一雷达3检测到的。
103、计算出第一目标5关于反射面对称的镜像区域。
为了计算出第一目标5关于反射面对称的镜像区域，需要先以车辆1所在的位置为原点，建立直角坐标系，如图3所示。并根据直角坐标系，计算出第一目标5的位置坐标。同时还需要计算出反射面的坐标位置。然后以反射面为参考面，才能计算出在第一雷达的视野范围里的镜像区域，镜像区域与第一目标5关于反射面是镜像对称的。
本实施例中，在镜像区域的计算过程中，是先计算得到第一目标5关于反射面的对 称点，对称点与第一目标5是关于反射面镜像对称；然后再以对称点为中心，以预设长度为半径画圆，最后将圆内区域视为镜像区域。当然，镜像区域也可以采用其他方法计算，计算得到的镜像区域可以是长方形、正方形或三角形等，只要是以对称点为中心的区域均可作为镜像区域，在此不做具体限制。
104、利用第一雷达3在镜像区域内提取第二目标6。
第一雷达3设置在邻近静止反射体2一侧，利用第一雷达3可以对镜像区域进行检测，确定镜像区域是否存在第二目标6，如果存在第二目标6，则第二目标6的数量有几个。
为了描述方便，仍以后方交叉预警场景(RCTA)为例进行说明，如图2所示。在图2中，静止反射体2设置在车辆1的左后侧，第一目标5设置在车辆1的右后侧。利用设置在车辆1右后方的第二雷达4检测第一目标5的位置为(X A，Y A)，根据静止反射体2的位置(X B，Y B)，能够计算出对称点的位置(X A'，Y A')，以对称点的位置为中心，以预设长度为半径，计算得到镜像区域。然后利用设置在车辆1左后方的第一雷达对镜像区域进行检测，检测到三个目标物P、Q及R。此时，雷达系统会默认为这三个目标均与第一目标5具有良好的空间对称性，因此雷达系统会将P、Q及R这三个目标物均作为第二目标进行保存，以便在后续步骤中进行逐一筛查。
105、判断第二目标6和第一目标5的运动特性是否关于反射面对称，若是，则认为第二目标6是第一目标5的镜像，即第二目标6就是镜像目标，此时忽略第二目标6；否则视为无镜像目标。
将步骤104中检测到三个目标物P、Q及R依次作为第二目标6，逐一代入考查其运动特性，以检测每一个目标物与第一目标5的运动特性是否关于反射面对称。若目标物的运动特性与第一目标5的运动特性关于反射面对称，则将该目标物视为第一目标的镜像，进行忽略处理。本实施例中，主要利用了第二目标6和第一目标5的空间位置、移动速度以及运动连续性这三个方面来判断第二目标6和第一目标5的运动特性是否关于反射面对称。由于镜像区域本身与第一目标关于反射面对称，因此，位于镜像区域内的目标物必然与第一目标满足空间对称的要求。
因此，可以直接以第二目标6和第一目标5的移动速度来判断第二目标6和第一目标5的运动特性是否关于反射面对称。一般需要先利用第二雷达4和第一雷达3分别获取第一目标5的第一移动速度以及第二目标6的第二移动速度，本实施例中，第二目标6包括P、Q及R这三个目标物，也就是说，需要利用第一雷达3分别获取P、Q及R这三个目标物的第二移动速度。然后再依次验证第一移动速度和各第二移动速度是否关于反射面镜像对称，若第一移动速度和第二移动速度关于反射面镜像对称，则认为该第二目标6是第一目标5的 镜像，忽略第二目标6；否则视为无镜像目标。验证第一移动速度和第二移动速度是否关于反射面镜像对称，可以直接采用速度绝对值做差法判断，假设目标物P是与第一目标5的移动速度是保持一致的，那么第一目标5的移动速度绝对值与目标物P的移动速度绝对值之差必然小于预设值。当然，也可以采用其他方法对第一移动速度和第二移动速度是否关于反射面镜像对称进行验证。
本实施例中，为了判断第一移动速度和第二移动速度是否关于反射面镜像对称，先对第一移动速度和第二移动速度进行了分解。即先对第一移动速度进行分解，计算出第一目标5的X轴向移动速度。然后再对第二移动速度进行分解，计算出第二目标6的X轴向移动速度。最后判断第一目标5的X轴向移动速度与第二目标6的X轴向移动速度是否关于反射面镜像对称，若是，则认为第一移动速度和第二移动速度关于反射面镜像对称；否则认为第一移动速度和第二移动速度关于反射面不对称。
当然，也可以对第一移动速度进行分解，获取第一目标5的Y轴向移动速度；对第二移动速度进行分解，计算出第二目标6的Y轴向移动速度。然后再判断第一目标5的Y轴向移动速度与第二目标6的Y轴向移动速度是否关于反射面镜像对称，以确定第一移动速度和第二移动速度是否关于反射面镜像对称。在具体使用过程中，可以利用第一移动速度和第二移动速度在X轴向的分速度来判断第一移动速度与第二移动速度关于反射面的镜像对称性，也可以利用第一移动速度和第二移动速度在Y轴向的分速度来判断，甚至为了提高准确性，也可以同时利用第一移动速度和第二移动速度在X轴向的分速度以及Y轴向的分速度来判断第一移动速度与第二移动速度关于反射面的镜像对称性，在此不做限制。
作为优选的，在判断第一移动速度和第二移动速度关于反射面镜像对称之后，还包括运动连续性判断步骤，即对第一目标5和第二目标6进行连续跟踪，验证是否在一段连续时间内，第一目标5的第一移动速度均能够与第二目标6的第二移动速度保持关于反射面的对称关系，以避免偶然事件的发生，提高对镜像目标的筛选精度。换句话说，如果在T 1到T 2这段时间内，第一目标5的第一移动速度始终与第二目标6的第二移动速度保持关于反射面的对称，那么第二目标6就可以视作第一目标5的镜像。在连续性判断过程中，首先需要利用第二雷达4连续获取多帧关于第一目标5的雷达检测特性，同时利用第一雷达3连续获取多帧关于第二目标6的雷达检测特性。即通过第二雷达4和第一雷达3对第一目标5和第二目标6的运动状态进行多次检测。然后计算出第一目标5在对应各帧雷达检测特性中的第一移动速度，计算出第二目标6在对应各帧雷达检测特性中的第二移动速度，以获取第一目标5和第二目标6在不同时刻的移动速度。最后将同一时刻获得的第一移动速度和第二移动速度进行比对，判断是否每一时刻获得的第一移动速度和第二移动速度均关于反射面镜像对 称，若是，则认为第二目标6是第一目标5的镜像，忽略第二目标6；否则视为无镜像目标。
现对本实施例公开的镜像目标去除方法的设计原理进行如下说明：
本发明所讨论的方案是针对真实目标出现在一个雷达的覆盖范围内，而静止反射体及镜像目标出现在另一个雷达的覆盖范围内的情况。例如在上述的RCTA场景中，真实目标和镜像目标会分别出现在左右两边的后雷达中，即R2和L2，其中左后雷达相当于第一雷达3，右后雷达相当于第二雷达4。镜像目标去除方法就是结合左后雷达与右后雷达检测到的航迹点迹的信息，对这些信息进行筛选，以滤除掉虚假的镜像目标。现以RCTA的具体场景为例，对镜像目标去除方法的原理展开说明。
如图2和图3所示，在RCTA的场景中，在真实目标从本车的右后边朝本车向左边行驶的过程中,由于本车附近(这里以左边为例)有一静止反射体2的存在，当角度合适的时候，左后雷达L2(相当于第一雷达)的发射信号经过静止反射体2的反射会到达真实目标(即为第一目标)，并从真实目标反射回来的信号经静止反射体2再反射又回到左后雷达L2。以距离向的路径分析，左后雷达认为在A’这个位置有一个“目标”从左向右靠近本车。其实在A’这个位置并没有目标存在，这里的“目标”其实是真实目标的镜像目标，是虚假的。由于镜像目标与真实目标是关于反射面呈镜像关系的，真实目标和镜像目标具有空间几何对称性、速度一致性以及相同的运动连续性。以车辆位置为点建立坐标系，则真实目标、镜像目标、静止反射体、车辆、车辆左后雷达及车辆右后雷达分别用点A、A’、B、O、O 2及O 1来表示，得到的坐标关系如图3所示。由图3可知，左后雷达O 2发射的电磁波经静止反射体B到达真实目标A，雷达回波原路返回经静止反射体B再被左后雷达所接收，从而获取到镜像目标。在此路径中，左后雷达O 2到真实目标A的距离为R A＝O 2B+AB；而左后雷达与镜像目标之间的距离为R A'＝O 2B+A'B，而R A＝R A。假设B’是静止反射体B在AA’连线上的垂直焦点，镜像目标相对于左后雷达O 2的方位角是θ 3，根据电磁波传播原理及以上信息，可以推出(其中θ 1是电磁波的入射角，θ 2是反射角)：θ 1＝θ 2＝θ 3，进而可以推出A'B＝AB以及A'B'＝AB'。推出的结果说明真实目标与镜像目标是以静止反射体的中轴线BB’为对称轴呈现左右对称的关系，即真实目标与镜像目标之间首先具有完美的空间几何对称性。
关于真实目标和镜像目标的速度一致性，为描述简单起见，可以进一步假设静止反射体2的方向，也就是直线BB’的方向与车辆1的方向平行。这样的话，就转化成真实目标A在水平方向上，即X方向上朝本车行驶过来。那么假设在时间t 1和t 2时刻，真实目标A在本车坐标系下的位置分别为(X A1,Y A)和(X A2,Y A)。同时镜像目标A’在t 1和t 2时刻的位置分别是(X A'1,Y A')和(X A'2,Y A')。由真实目标与镜像目标在空间上的几何对称性可以推出 如下公式：
|X A1-X B'|＝|X A'1-X B'|         (1)
|X A2-X B'|＝|X A'2-X B'|     (2)
|X A1-X A2|＝|X A'1-X A'2|       (3)
由公式(1)、公式(2)及公式(3)，可推出
v xA＝|X A2-X A1|/(t 2-t 1)＝
-v xA'＝|X A'1-X A'2|/(t 2-t 1)，
v xA与v xA'的大小是相等的，在车身坐标系下，这两个速度的方向是相反的，即在X轴方向上，真实目标与镜像目标的速度大小是相同的，两者具有速度一致性。而且，真实目标的运动特性中不断靠近本车，其径向速度的水平分量，即横向速度的方向矢量为正；同样，镜像目标的运动特性中不断靠近本车，其径向速度的水平分量，即横向速度的方向矢量也为正，因而镜像目标与真实目标的速度不仅大小相等，且方向均为靠近车辆。
而真实目标和镜像目标的运动连续性主要表现在，只要镜像目标和真实目标被相应侧的雷达检测到，这种空间对称性和速度一致性的特性在时间上，即连续几帧的雷达检测特性中，能够一直得到保持，直到真实目标不被检测到或者超出雷达的覆盖区域。所以当有静止反射体2存在，且左右雷达中检测到的航迹都满足上面三个特性，那么它们就是一对真实目标与镜像目标，即可认为与静止反射体2同一侧检测到的目标为镜像目标。但只要其中任何一个特性得不到满足，那么这个检测到的目标就不会是镜像目标。比如说，有可能恰好在某帧的时候，在对称的位置上检测到一个目标，且它的速度矢量也恰好与另一侧检测到的目标一致，然而通过连续跟踪几帧后，发现对称性或一致性不能得到持续保证，那么该目标就不可能是镜像目标，因此也就不会被错误地判断为镜像目标而丢失真实目标。
以上仅针对RCTA的场景为例进行了说明，但实际上该镜像目标去除方法同样适用与RCTA场景类似的场景，如变道辅助场景、开门预警场景等。在与RCTA类似的场景中，也可以利用上述的分析逻辑实现对镜像目标的判断，再次不做重复赘述。
显然，本发明的上述实施例仅仅是为清楚地说明本发明所作的举例，而并非是对本发明的实施方式的限定。对于所属领域的普通技术人员来说，在上述说明的基础上还可以做出其它不同形式的变化或变动。这里无需也无法对所有的实施方式予以穷举。凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明权利要求的保护范围之内。

Claims (10)

  1. 一种基于车载角雷达的镜像目标去除方法，其特征在于，基于设置在车辆四个角落的雷达，包括如下步骤：
    通过第一雷达检测静止反射体，静止反射体与第一雷达位于车辆的同一侧，分析出静止反射体相对于第一雷达的反射面；
    利用第二雷达获取第一目标；
    计算出第一目标关于反射面对称的镜像区域；
    利用第一雷达在镜像区域内提取第二目标；
    判断第二目标和第一目标的运动特性是否关于反射面对称，若是，则认为第二目标是第一目标的镜像，忽略第二目标；否则视为无镜像目标。
  2. 如权利要求1所述一种基于车载角雷达的镜像目标去除方法，其特征在于，所述检测静止反射体包括如下步骤：
    对车辆周围的物体进行检测，筛选出与车辆距离小于第一阈值的物体作为目标物体；
    判断目标物体的移动速度是否小于第二阈值，若是，则将目标物体视为静止反射体；否则忽略目标物体。
  3. 如权利要求1所述一种基于车载角雷达的镜像目标去除方法，其特征在于，所述反射面是指能够将第一雷达的发射信号反射至第一目标，并将第一目标反射回来的信号再次反射给第一雷达。
  4. 如权利要求1所述一种基于车载角雷达的镜像目标去除方法，其特征在于，
    所述第一雷达设置在车辆与静止反射体相对应一侧，所述第一雷达用于检测静反射体和第二目标；所述第二雷达设置在车辆与第一目标相对应一侧，所述第二雷达用于检测第一目标。
  5. 如权利要求1所述一种基于车载角雷达的镜像目标去除方法，其特征在于，所述计算出第一目标关于反射面对称的镜像区域的步骤，包括：
    以车辆所在位置建立坐标系；
    获取第一目标的位置坐标；
    以反射面为参考面，计算在第一雷达的视野范围里的镜像区域，所述镜像区域与第一目标关于反射面镜像对称。
  6. 如权利要求5所述一种基于车载角雷达的镜像目标去除方法，其特征在于，所述镜像区域的计算包括：
    计算第一目标关于反射面的对称点，对称点与第一目标关于反射面镜像对称；
    以对称点为中心，以预设长度为半径画圆，圆内区域即为镜像区域。
  7. 如权利要求1所述一种基于车载角雷达的镜像目标去除方法，其特征在于，
    利用第二目标和第一目标移动速度以及运动连续性这两个方面来判断第二目标和第一目标的运动特性是否关于反射面对称。
  8. 如权利要求7所述一种基于车载角雷达的镜像目标去除方法，其特征在于，所述判断第二目标和第一目标的运动特性是否关于反射面对称，若是，则认为第二目标是第一目标的镜像，忽略第二目标；否则视为无镜像目标步骤，包括：
    获取第一目标的第一移动速度和第二目标的第二移动速度；
    判断第一移动速度和第二移动速度是否关于反射面镜像对称，若是，则认为第二目标是第一目标的镜像，忽略第二目标；否则视为无镜像目标。
  9. 如权利要求8所述一种基于车载角雷达的镜像目标去除方法，其特征在于，
    在判断第一移动速度和第二移动速度关于反射面镜像对称之后，还包括运动连续性判断步骤，所述运动连续性判断步骤包括：
    利用第二雷达连续获取多帧关于第一目标的雷达检测特性，同时利用第一雷达连续获取多帧关于第二目标的雷达检测特性；
    计算出第一目标在对应的各帧雷达检测特性中的第一移动速度；
    计算出第二目标在对应的各帧雷达检测特性中的第二移动速度；
    将同一时刻获得的第一移动速度和第二移动速度进行比对，判断是否每一时刻获得的第一移动速度和第二移动速度均关于反射面镜像对称，若是，则认为第二目标是第一目标的镜像，忽略第二目标；否则视为无镜像目标。
  10. 如权利要求8所述一种基于车载角雷达的镜像目标去除方法，其特征在于，所述判断第一移动速度和第二移动速度是否关于反射面镜像对称包括如下子步骤：
    对第一移动速速进行分解，计算第一目标的X轴向移动速度；
    对第二移动速度进行分解，计算第二目标的X轴向移动速度；
    判断第一目标的X轴向移动速度与第二目标的X轴向移动速度是否关于反射面镜像对称，若是，则认为第一移动速度和第二移动速度关于反射面镜像对称；否则认为第一移动速度和第二移动速度关于反射面不对称。
PCT/CN2019/121202 2019-11-22 2019-11-27 一种基于车载角雷达的镜像目标去除方法 WO2021097880A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911154918.0 2019-11-22
CN201911154918.0A CN110940980B (zh) 2019-11-22 2019-11-22 一种基于车载角雷达的镜像目标去除方法

Publications (1)

Publication Number Publication Date
WO2021097880A1 true WO2021097880A1 (zh) 2021-05-27

Family

ID=69907974

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/121202 WO2021097880A1 (zh) 2019-11-22 2019-11-27 一种基于车载角雷达的镜像目标去除方法

Country Status (2)

Country Link
CN (1) CN110940980B (zh)
WO (1) WO2021097880A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102332509B1 (ko) * 2020-05-22 2021-11-29 현대모비스 주식회사 후방 교차 충돌 경고 방법 및 장치
CN112835026B (zh) * 2020-12-31 2024-02-20 福瑞泰克智能系统有限公司 雷达镜像目标检测方法、装置、雷达设备和车辆
CN113009467B (zh) * 2021-03-09 2022-12-30 森思泰克河北科技有限公司 一种雷达盲区目标检测跟踪方法、装置及终端设备
CN113504508B (zh) * 2021-04-13 2023-11-17 惠州市德赛西威智能交通技术研究院有限公司 一种改善雷达低频包络及rcta镜像目标检测的方法
CN114443636B (zh) * 2022-01-25 2024-09-17 同济大学 一种基于毫米波雷达的隧道轨迹数据镜像去除方法
CN114779180A (zh) * 2022-06-20 2022-07-22 成都瑞达物联科技有限公司 一种面向车路协同雷达的多径干扰镜像目标滤除方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150198697A1 (en) * 2014-01-15 2015-07-16 Panasonic Corporation Radar apparatus
CN104793192A (zh) * 2014-01-21 2015-07-22 罗伯特·博世有限公司 用于角度估计的方法以及用于机动车的雷达传感器
CN106004659A (zh) * 2016-08-03 2016-10-12 安徽工程大学 车辆周围环境感知系统及其控制方法
CN109532828A (zh) * 2017-09-19 2019-03-29 丰田自动车株式会社 车辆周边监视装置
CN109870680A (zh) * 2018-10-26 2019-06-11 北京润科通用技术有限公司 一种目标分类方法及装置
CN110203204A (zh) * 2019-05-17 2019-09-06 深圳森云智能科技有限公司 一种车辆周边环境感知方法

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8976059B2 (en) * 2012-12-21 2015-03-10 Raytheon Canada Limited Identification and removal of a false detection in a radar system
DE102013209530A1 (de) * 2013-05-23 2014-11-27 Robert Bosch Gmbh Bestimmung eines elevations-dejustagewinkels eines radarsensors eines kraftfahrzeugs
CN103364795B (zh) * 2013-07-17 2015-07-29 中国科学院上海光学精密机械研究所 合成孔径激光成像雷达的光学成像系统与光学成像方法
US9810782B2 (en) * 2015-03-20 2017-11-07 Delphi Technologies, Inc. Vehicle radar system with image reflection detection
CN106204568A (zh) * 2016-07-06 2016-12-07 天津大学 极化sar图像人造目标提取方法
US20180095163A1 (en) * 2016-10-03 2018-04-05 John Lovberg Phase-modulated continuous wave radar system (with prbs codes)
JP7053982B2 (ja) * 2017-05-25 2022-04-13 ミツミ電機株式会社 ゴースト除去方法及びレーダ装置
US10585188B2 (en) * 2017-07-28 2020-03-10 Valeo Radar Systems, Inc. Broadside detection system and techniques for use in a vehicular radar
CN107656255B (zh) * 2017-10-25 2019-10-25 中国人民解放军国防科技大学 基于多径回波的超宽带雷达动目标二维定位方法
CN108318864B (zh) * 2018-02-06 2021-10-22 成都纳雷科技有限公司 一种用于雷达目标检测中消除多径目标的方法及装置
CN110320518B (zh) * 2019-05-31 2023-01-24 惠州市德赛西威汽车电子股份有限公司 一种车载bsd毫米波雷达安装位置自动标定方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150198697A1 (en) * 2014-01-15 2015-07-16 Panasonic Corporation Radar apparatus
CN104793192A (zh) * 2014-01-21 2015-07-22 罗伯特·博世有限公司 用于角度估计的方法以及用于机动车的雷达传感器
CN106004659A (zh) * 2016-08-03 2016-10-12 安徽工程大学 车辆周围环境感知系统及其控制方法
CN109532828A (zh) * 2017-09-19 2019-03-29 丰田自动车株式会社 车辆周边监视装置
CN109870680A (zh) * 2018-10-26 2019-06-11 北京润科通用技术有限公司 一种目标分类方法及装置
CN110203204A (zh) * 2019-05-17 2019-09-06 深圳森云智能科技有限公司 一种车辆周边环境感知方法

Also Published As

Publication number Publication date
CN110940980B (zh) 2022-12-27
CN110940980A (zh) 2020-03-31

Similar Documents

Publication Publication Date Title
WO2021097880A1 (zh) 一种基于车载角雷达的镜像目标去除方法
US10605896B2 (en) Radar-installation-angle calculating device, radar apparatus, and radar-installation-angle calculating method
US11768286B2 (en) Method of determining the yaw rate of a target vehicle
WO2018040853A1 (zh) 设置微波雷达和超声波传感器的汽车盲区探测系统及方法
JP5003674B2 (ja) レーダ装置および移動体
US10371810B2 (en) Radar device
US20090292468A1 (en) Collision avoidance method and system using stereo vision and radar sensor fusion
US20110291874A1 (en) Vehicle radar system and method for detecting objects
RU2671444C1 (ru) Устройство содействия при вождении
CN106168667B (zh) 具有图像反射检测的车载雷达系统
CN109367529B (zh) 毫米波雷达组合安装结构及虚拟隧道构建与障碍判断方法
WO2022134510A1 (zh) 一种车载bsd毫米波雷达低速下障碍物识别方法
Wang et al. Vehicle detection and width estimation in rain by fusing radar and vision
CN111699404B (zh) 行驶辅助目标获取方法与装置、雷达、行驶系统与车辆
CN104106102A (zh) 警报装置
WO2013038477A1 (ja) 警報装置
US11891081B2 (en) Warning apparatus
WO2020039840A1 (ja) レーダ処理装置
WO2020189775A1 (ja) 警報装置
JP2001216596A (ja) 路上検出装置及び自動車両
Blanc et al. Track to track fusion method applied to road obstacle detection
JP2002181936A (ja) 障害物位置計測方法および障害物位置計測装置
JP2002122669A (ja) 物体位置検出方法
WO2021046962A1 (zh) 一种屏蔽门与车体间的障碍物检测系统及检测方法
CN116299473A (zh) 一种基于mimo毫米波雷达的横穿目标探测方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19953492

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19953492

Country of ref document: EP

Kind code of ref document: A1