WO2023130620A1 - 基于毫米波雷达方差统计的目标车辆运动状态判断方法 - Google Patents

基于毫米波雷达方差统计的目标车辆运动状态判断方法 Download PDF

Info

Publication number
WO2023130620A1
WO2023130620A1 PCT/CN2022/088408 CN2022088408W WO2023130620A1 WO 2023130620 A1 WO2023130620 A1 WO 2023130620A1 CN 2022088408 W CN2022088408 W CN 2022088408W WO 2023130620 A1 WO2023130620 A1 WO 2023130620A1
Authority
WO
WIPO (PCT)
Prior art keywords
track
target vehicle
threshold
state
curve
Prior art date
Application number
PCT/CN2022/088408
Other languages
English (en)
French (fr)
Inventor
宋玛君
黄小月
王奇
吴军
Original Assignee
南京楚航科技有限公司
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 南京楚航科技有限公司 filed Critical 南京楚航科技有限公司
Priority to DE112022000008.2T priority Critical patent/DE112022000008T5/de
Publication of WO2023130620A1 publication Critical patent/WO2023130620A1/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
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/50—Systems of measurement based on relative movement of target
    • G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
    • 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
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/06—Systems determining position data of a target
    • G01S13/42—Simultaneous measurement of distance and other co-ordinates
    • 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
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/50—Systems of measurement based on relative movement of target
    • G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
    • G01S13/589—Velocity or trajectory determination systems; Sense-of-movement determination systems measuring the velocity vector
    • 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
    • G01S13/66—Radar-tracking systems; Analogous systems
    • G01S13/72—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar
    • G01S13/723—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar by using numerical data
    • 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
    • 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

Definitions

  • the present application relates to the technical field of data processing based on millimeter-wave radar, for example, it relates to a method for judging the motion state of a target vehicle based on variance statistics of millimeter-wave radar.
  • the technology of sensing the lane change recognition of the target vehicle is mainly manifested in the camera sensor, based on the image recognition lane line and the lateral distance of the vehicle target, and judging the lateral distance between the target vehicle and the lane line to determine whether the target vehicle is changing lanes .
  • millimeter-wave radar With the development of millimeter-wave radar technology, new requirements are put forward for the detection of the target vehicle ahead. It is hoped that the lateral vibration of the target vehicle can be reduced under the premise that the vehicle equipped with millimeter-wave radar can stably track the vehicle target. Since the millimeter-wave radar does not have an advantage in the angle of detection of the target vehicle compared to the accuracy of the distance, the accuracy of the millimeter-wave radar in detecting the lateral distance of the vehicle target vehicle is not very high.
  • This application provides a method for judging the motion state of a target vehicle based on millimeter-wave radar variance statistics, which is suitable for the scene where a vehicle equipped with millimeter-wave radar judges whether the target vehicle in front is changing lanes, and aims at the situation where the target vehicle is driving in a straight line Next, smoothing is performed on the lateral shake measurement data of the target vehicle.
  • the present application provides a method for judging the motion state of a target vehicle based on variance statistics of millimeter-wave radar, including:
  • sum_dot_num is greater than cycle_thr, it is judged whether the vehicle where the millimeter-wave radar is located is performing a curved motion, and based on the judgment result that the vehicle where the millimeter-wave radar is located is performing a curved motion, the track of the target vehicle is navigated in this frame
  • the track state in the track data is judged as unknown unknown, and the straight-line statistical variables strightCount, sum_y and sum_dot_num of the track of the target vehicle are set to zero; based on the judgment result that the vehicle where the millimeter-wave radar is located is not performing curved motion, Calculate the varianceY of the lateral position y of the target vehicle in the sliding window, calculate the mean value all_mean_y of the historical lateral position y of the target vehicle according to sum_y and sum_dot_num, calculate the lateral position y of the target vehicle in the sliding window based on the mean value all_mean_y The variance all_mean_varianceY;
  • the track state of the track of the target vehicle in the track data of this frame is judged and updated, including:
  • the track of the target vehicle in the current frame of track data satisfies the condition that the track is a linear motion
  • the condition that the trajectory is linear motion is: varianceY ⁇ varianceY_threshold, and the absolute value abs_vy of the lateral velocity vy of the target vehicle in the track data of this frame satisfies abs_vy ⁇ vy_threshold; wherein, varianceY_threshold is the variance threshold of the trajectory linear motion state , vy_threshold is the lateral velocity threshold of the trajectory linear motion state;
  • the trajectory state of the target vehicle's trajectory in the last frame of trajectory data is a curve
  • the transition condition of the trajectory state from a curve to a straight line is: varianceY ⁇ varianceY_threshold, and the absolute value abs_vy of the lateral velocity vy of the target vehicle in the track data of this frame satisfies abs_vy ⁇ vy_threshold;
  • sum_y is the lateral position y of the target vehicle in the track data of this frame;
  • the track state of the track of the target vehicle in the frame of track data is updated to a curve, and the straightCount of the track is set to 0; in response to determining that the track of the target vehicle in the current frame of track data does not meet the condition that the track is a curved motion, update the track state of the track of the target vehicle in the current frame of track data to a straight line, and the track The straightCount is incremented by 1.
  • FIG. 1 is a schematic flow chart of a method for judging the motion state of a target vehicle based on millimeter-wave radar variance statistics in an embodiment of the present application;
  • Fig. 2 is a schematic diagram of the coordinate system formed by the vehicle where the millimeter-wave radar is located in the embodiment of the present application;
  • Fig. 3 is a schematic flowchart of judging and updating the track state of the target vehicle track in the track data of the current frame in the embodiment of the present application.
  • the embodiment of the present application provides a method for judging the motion state of a target vehicle based on the variance statistics of the millimeter-wave radar, including:
  • the historical position of the target vehicle in the multi-frame track data is stored in the form of a sliding window, and the window number of the sliding window is cycle_thr.
  • the window number cycle_thr of the sliding window may be 5.
  • Each frame of track data includes multiple tracks, and the target vehicle corresponds to one track in each frame of track data.
  • the lateral position of the target vehicle’s track is the lateral position of the target vehicle, and the speed of the target vehicle’s track is the target the speed of the vehicle.
  • the sum sum_y of the lateral position y of the target vehicle in the multi-frame track data is an integer greater than 1), and the number of times sum_dot_num of the lateral position y of the calculated track is accumulated, where sum_y and sum_dot_num are set to 0 at the initial time of the track.
  • the horizontal position y of the track refers to the coordinate value on the Y axis of the updated position of the target vehicle in this frame of track data in the coordinate system shown in Figure 2.
  • the box area in the coordinate system shown in Figure 2 represents the millimeter wave
  • the vehicle where the radar is located takes the center of the rear axle of the vehicle where the millimeter-wave radar is located as the origin, the axis of the vehicle where the millimeter-wave radar is located is the x-axis direction, and the lateral direction of the vehicle where the millimeter-wave radar is located is the Y-axis direction.
  • each frame of track data may also include tracks of other vehicles.
  • the method adopted is consistent with the method for judging the motion state of the target vehicle.
  • the number of tracks refers to the times of recording the lateral position y of the track in the sliding window.
  • sum_dot_num is greater than cycle_thr, then judge whether the vehicle where the millimeter wave radar is located is moving in a curve, if the vehicle where the millimeter wave radar is located is moving in a curve, then set the track state of the track of the target vehicle in the track data of this frame to unknown unknown, and set The linear statistical variables strightCount, sum_y and sum_dot_num of the track are set to zero.
  • the vehicle where the millimeter-wave radar is located is not moving in a curved line, calculate the varianceY of the lateral position y of the track in the sliding window, then calculate the mean all_mean_y of the historical lateral position y of the track according to the track sum_y and sum_dot_num, and finally calculate the sliding window
  • the track lateral position y within the window is based on the variance all_mean_varianceY of the mean all_mean_y.
  • the following method can be used to determine whether the vehicle where the millimeter wave radar is located is moving in a curve: Obtain the speed and yaw rate yawRate of the vehicle where the millimeter wave radar is located, such as the speed of the vehicle where the millimeter wave radar is located > the curve speed threshold, and the yaw rate yawRate If the absolute value of > curve yaw rate threshold, it is determined that the vehicle where the millimeter-wave radar is located is moving in a curve.
  • the curve velocity threshold can be set to 0.1m/s
  • the curve yaw rate threshold can be set to 0.1-0.15rad/s.
  • the track state of the target vehicle in the track data of this frame is judged and updated. See Figure 3. There are three types of track states: straight line, curve, and unknown unknown. Since the target vehicle in the track data of this frame The basis for judging the trajectory state of the track is related to the track state of the target vehicle’s track in the last frame of track data, so it is necessary to check the track state of the target vehicle’s track in the last frame of track data. include:
  • the track state of the track of the target vehicle in the last frame of track data is unknown unknown, it is judged whether the track of the target vehicle in the track data of this frame satisfies the condition that the track is a straight-line motion, and the condition for the track to be a straight-line motion is: varianceY ⁇ varianceY_threshold, and the absolute value abs_vy of the lateral velocity vy of the target vehicle in the track data of this frame satisfies abs_vy ⁇ vy_threshold.
  • varianceY_threshold is the variance threshold of the straight line state
  • vy_threshold is the lateral velocity threshold of the straight line state.
  • trajectory state of the target vehicle judges whether the trajectory of the target vehicle in the current frame of trajectory data satisfies the transition condition of the trajectory state from a curve to a straight line, and the trajectory state changes from a curve to a straight line
  • the transition condition is: varianceY ⁇ varianceY_threshold, and the absolute value of the lateral velocity vy of the target vehicle in the track data of this frame ⁇ vy_threshold. This judgment condition is the same as the above-mentioned condition that the track is a straight line motion.
  • the track of the target vehicle in the track data of this frame satisfies the transition condition from a curve to a straight line, then the track of the target vehicle in the track data of this frame is The track state is updated to be straight.
  • the straightCount of the target vehicle’s track in the current frame’s track data will be increased by 1, and the statistics of the historical track will be restarted.
  • the track state of the track of the target vehicle in the last frame of track data is a straight line, then judge whether the track of the target vehicle in the track data of this frame satisfies the condition that the track is a curved motion, and the condition for the track to be a curved motion is: all_mean_varianceY >all_varianceY_threshold, meanwhile, the absolute value of the lateral velocity vy of the target vehicle in this frame of track data>curve_vy_threshold.
  • all_varianceY_threshold is the variance threshold of the curve state
  • curve_vy_threshold is the lateral velocity threshold of the curve state.
  • the track of the target vehicle in the track data of this frame meets the condition that the track is a curved motion, its state will be updated to a curve.
  • StraightCount is set to 0. If it is determined that the track of the target vehicle in the track data of this frame does not meet the condition that the track is a curved motion, update the state to a straight line and add 1 to the straightCount of the track.
  • the track state of the track of the target vehicle in the current frame track data is a curve, it means that the target vehicle is changing lanes.
  • the lateral speed threshold vy_threshold of the straight line state, the lateral speed threshold Curve_vy_threshold of the curved state, the variance threshold varianceY_threshold of the straight line state, and the variance threshold all_mean_varianceY_threshold of the curved state of the embodiment of the present application can be set to a constant value, but the target vehicle track and the millimeter wave
  • the longitudinal distance of the vehicle where the radar is located and the speed of the target vehicle track will have a certain impact on the recognition accuracy. In order to reduce this impact, it can also be dynamically adjusted.
  • the lateral velocity threshold vy_threshold in the straight line state can be set to a constant value, which can be 0.5m/s.
  • the initial value of the lateral velocity threshold Curve_vy_threshold in the curve state can be set to 0.5m/s
  • the initial value of the variance threshold varianceY_threshold in the straight line state can be set to 0.01 square meters
  • the initial value of the variance threshold all_mean_varianceY_threshold in the curve state can be set as 0.7 square meters.
  • the variance threshold varianceY_threshold of the straight line state and the variance threshold all_mean_varianceY_threshold of the curve state are dynamically adjusted. Dynamically adjust the lateral velocity threshold Curve_vy_threshold of the curve state according to the absolute value abs_vx of the longitudinal velocity vx of the target vehicle in the track data of this frame.
  • the dynamic adjustment method is as follows:
  • the variance threshold varianceY_threshold of the straight state is adjusted to 0.03 square meters, and the variance threshold all_mean_varianceY_threshold of the curved state is adjusted to 1.8 square meters;
  • the variance threshold varianceY_threshold of the straight state is adjusted to 0.02 square meters, and the variance threshold all_mean_varianceY_threshold of the curve state is adjusted to 1.6 square meters;
  • the variance threshold all_mean_varianceY_threshold of the straight line state is adjusted to 1.2 square meters;
  • the lateral velocity threshold Curve_vy_threshold of the curve state is adjusted to 0.8m/s. It should be noted that the way of dynamic adjustment is not limited to the above way, and the way of dynamic adjustment such as linear change can also be adopted.
  • the application provides an electronic device, including:
  • processors one or more processors
  • memory configured to store one or more programs
  • the one or more processors When the one or more programs are executed by the one or more processors, the one or more processors are made to implement the method for judging the target motion state based on variance statistics of the millimeter-wave radar as described above.
  • the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the target motion state judgment based on millimeter-wave radar variance statistics as described above is realized method.
  • the process described above with reference to the flow chart can be implemented by electronic equipment, and the electronic equipment includes a processor (such as a central processing unit, a graphics processing unit, etc.), which can execute multiple appropriate action and handling.
  • a processor such as a central processing unit, a graphics processing unit, etc.
  • the processes described above with reference to the flowcharts may be implemented as computer software programs.
  • the embodiments of the present application include a computer program product, which includes a computer program carried on a non-transitory computer readable medium, where the computer program includes program code configured to execute the method shown in the flowchart.
  • the above-mentioned computer-readable storage medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.
  • the computer readable storage medium may be a non-transitory computer readable storage medium.
  • the present application identifies the target vehicle trajectory in real time based on variance statistics.
  • the lateral velocity of the target vehicle is small and the variance of the lateral position of the target vehicle (target vehicle track) in the sliding window is small, it means that the target vehicle is moving in a straight line.
  • the horizontal position of the target vehicle track in the sliding window is based on the historical horizontal position
  • the variance is calculated by filtering the mean value.
  • the target vehicle When the variance is greater than the threshold and the target vehicle track filtering speed is greater than the speed threshold, the target vehicle is identified as moving in a curve, that is, the target vehicle has a lane-changing behavior.
  • this method is more accurate in identifying the track state of the target vehicle in a straight line or changing lanes, and lays the foundation for the subsequent smoothing of the track of the target vehicle.

Landscapes

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

Abstract

本申请公开了一种基于毫米波雷达方差统计的目标车辆运动状态判断方法。该方法包括以滑窗的方式存储目标车辆在多帧航迹数据中的历史位置，根据所述历史位置累加计算所述目标车辆在所述多帧航迹数据中的横向位置y之和，并累加计算目标车辆的横向位置y的次数；判断毫米雷达波所在车辆是否处于曲线运动，在未进行曲线运动时，计算滑窗内目标车辆的横向位置y的方差，根据横向位置y之和以及标记横向位置y的次数计算目标车辆的历史横向位置y的均值，计算滑窗内目标车辆的横向位置y基于y的均值的方差；根据横向位置y的方差以及y基于y的均值的方差判断目标车辆航迹在本帧航迹数据中的轨迹状态并更新。

Description

基于毫米波雷达方差统计的目标车辆运动状态判断方法
本申请要求在2022年1月5日提交中国专利局、申请号为202210003432.2受理的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本申请涉及基于毫米波雷达数据处理技术领域，例如涉及一种基于毫米波雷达方差统计的目标车辆运动状态判断方法。
背景技术
目前在感知目标车辆变道识别的技术上，主要表现在摄像头传感器上，基于图像识别车道线及车辆目标的横向距离，判断目标车辆与车道线之间的横向距离去判断目标车辆是否在变道。
随着毫米波雷达的技术发展，对检测到的前方目标车辆提出了新的需求，希望能在安装有毫米波雷达的车辆稳定跟踪车辆目标的前提下，减少目标车辆的横向抖动性。由于毫米波雷达在对目标车辆的检测中，相比于距离的精准性，角度上并不占据优势，因此毫米波雷达在检测车辆目标车辆的横向距离上，精准度不是很高。
发明内容
本申请提供一种基于毫米波雷达方差统计的目标车辆运动状态判断方法，适用于安装有毫米波雷达的车辆判断前方目标车辆是否存在变道行为的场景，旨在针对目标车辆处于直线行驶的情况下，对目标车辆的横向抖动测量数据进行平滑处理。
本申请提供了一种基于毫米波雷达方差统计的目标车辆运动状态判断方法，包括：
以滑窗的方式存储所述目标车辆的航迹在多帧航迹数据中的历史位置，滑窗的窗口数为cycle_thr，每帧航迹数据中包括所述目标车辆的航迹；
根据所述历史位置累加计算所述目标车辆在所述多帧航迹数据中的横向位置y之和sum_y，并累加计算所述目标车辆的横向位置y的次数sum_dot_num，其中，所述sum_y和sum_dot_num在航迹初始时置为0；
判断滑窗内存储所述目标车辆的横向位置y的点迹数是否大于cycle_thr，基于所述点迹数小于或等于cycle_thr的判断结果，将所述目标车辆的航迹在本帧航迹数据中的轨迹状态判断为未知unknow，判断过程结束；基于所述点迹数大于cycle_thr的判断结果，判断sum_dot_num是否大于cycle_thr；基于所述sum_dot_num小于或等于cycle_thr的判断结果，判断过程结束；
基于sum_dot_num大于cycle_thr的判断结果，判断所述毫米波雷达所在车辆是否在进行曲线运动，基于所述毫米波雷达所在车辆在进行曲线运动的判断 结果，将所述目标车辆的航迹在本帧航迹数据中的轨迹状态判断为未知unknow，并将所述目标车辆的航迹的直线统计变量strightCount、sum_y和sum_dot_num置为零；基于所述毫米波雷达所在车辆未在进行曲线运动的判断结果，计算滑窗内所述目标车辆的横向位置y的方差varianceY，根据sum_y和sum_dot_num计算所述目标车辆的的历史横向位置y的均值all_mean_y，计算滑窗内所述目标车辆的横向位置y基于均值all_mean_y的方差all_mean_varianceY；
根据varianceY和all_mean_varianceY判断所述目标车辆的航迹在本帧航迹数据中的轨迹状态并更新，判断过程结束；
其中，所述根据varianceY和all_mean_varianceY判断所述目标车辆的航迹在本帧航迹数据中的轨迹状态并更新，包括：
响应于确定上一帧航迹数据中所述目标车辆的航迹的轨迹状态为未知unknow，判断所述本帧航迹数据中所述目标车辆的航迹是否满足轨迹为直线运动的条件，其中，所述轨迹为直线运动的条件为：varianceY＜varianceY_threshold，且所述本帧航迹数据中目标车辆的横向速度vy的绝对值abs_vy满足abs_vy≤vy_threshold；其中，varianceY_threshold为轨迹直线运动状态的方差阈值，vy_threshold为轨迹直线运动状态的横向速度阈值；
响应于确定所述上一帧航迹数据中目标车辆的航迹的轨迹状态为曲线，判断所述本帧航迹数据中目标车辆的航迹是否满足轨迹状态从曲线到直线的转变条件，所述轨迹状态从曲线到直线的转变条件为：varianceY＜varianceY_threshold，且所述本帧航迹数据中目标车辆的横向速度vy的绝对值abs_vy满足abs_vy≤vy_threshold；
响应于确定所述本帧航迹数据中目标车辆的航迹满足轨迹为直线运动的条件或所述本帧航迹数据中目标车辆的航迹满足轨迹从曲线到直线的转变条件，将所述本帧航迹数据中目标车辆的航迹的轨迹状态更新为直线，航迹的straightCount加1，并重新开始统计所述目标车辆历史航迹的y值，所述目标车辆的航迹的sum_dot_num＝1，sum_y为本帧航迹数据中目标车辆的横向位置y；
响应于确定所述上一帧航迹数据中目标车辆的航迹的轨迹状态为直线，判断所述本帧航迹数据中目标车辆的航迹是否满足轨迹为曲线运动的条件，所述轨迹为曲线运动的条件为：all_mean_varianceY＞all_varianceY_threshold，且所述本帧航迹数据中目标车辆的横向速度vy的绝对值abs_vy满足abs_vy＞curve_vy_threshold；其中，all_varianceY_threshold为轨迹曲线运动状态的方差阈值，curve_vy_threshold为轨迹曲线运动状态的横向速度阈值；
响应于确定所述本帧航迹数据中目标车辆的航迹满足轨迹为曲线运动的条件，将所述本帧航迹数据中目标车辆的航迹的轨迹状态更新为曲线，航迹的straightCount置0；响应于确定所述本帧航迹数据中目标车辆的航迹不满足轨迹为曲线运动的条件，将所述本帧航迹数据中目标车辆的航迹的轨迹状态更新为直线，航迹的straightCount加1。
附图说明
图1是本申请实施例基于毫米波雷达方差统计的目标车辆运动状态判断方法的流程示意图；
图2是本申请实施例以毫米波雷达所在车辆形成的坐标系的示意图；
图3是本申请实施例中判断本帧航迹数据中目标车辆航迹的轨迹状态并更新的流程示意图。
具体实施方式
如图1所示，本申请实施例提供了一种基于毫米波雷达方差统计的目标车辆运动状态判断方法，包括：
以滑窗的方式存储目标车辆在多帧航迹数据中的历史位置，滑窗的窗口数为cycle_thr。其中，滑窗的窗口数cycle_thr可以为5个。
每帧航迹数据中包括多个航迹，目标车辆在每帧航迹数据中对应一个航迹，目标车辆航迹的横向位置即为目标车辆的横向位置，目标车辆航迹的速度即为目标车辆的速度。
根据历史位置累加计算目标车辆在多帧航迹数据中的横向位置y之和sum_y(即对包含有目标车辆航迹的N帧航迹数据进行累加，计算N帧航迹数据中的N个目标车辆航迹的横向位置y之和sum_y，N为大于1的整数)，并累加计算航迹的横向位置y的次数sum_dot_num，其中，sum_y和sum_dot_num在航迹初始时置为0。航迹的横向位置y是指在图2所示的坐标系中，这帧航迹数据中目标车辆更新的位置在Y轴上的坐标值，图2所示的坐标系中方框区域表示毫米波雷达所在车辆，以毫米波雷达所在车辆的后轴中心为原点，毫米波雷达所在车辆的轴向为x轴方向，毫米波雷达所在车辆的横向为Y轴方向。
在一实施例中，每帧航迹数据中除了包括目标车辆的航迹之外，还可以包括其他车辆的航迹。在对其他车辆进行运动状态判断时，采用的方法与判断目标车辆运动状态的方法一致。
判断滑窗内存储目标车辆的横向位置y的点迹数是否大于cycle_thr，若点迹数不大于cycle_thr，则将本帧航迹数据中目标车辆的航迹的轨迹状态判断为未知unknow，若点迹数大于cycle_thr，则判断sum_dot_num是否大于cycle_thr，若sum_dot_num不大于cycle_thr，则结束。其中，点迹数指的是滑窗内记录航迹横向位置y的次数。
若sum_dot_num大于cycle_thr，则判断毫米波雷达所在车辆是否在曲线运动，若毫米波雷达所在车辆在曲线运动，则将本帧航迹数据中目标车辆的航迹的轨迹状态设置为未知unknow，并将航迹的直线统计变量strightCount、sum_y和sum_dot_num置为零。如果判断结果为毫米波雷达所在车辆不在做曲线运动，则计算滑窗内航迹横向位置y的方差varianceY，然后根据航迹sum_y和sum_dot_num计算航迹的历史横向位置y的均值all_mean_y，最后计算滑窗内航迹横向位置y基于均值all_mean_y的方差all_mean_varianceY。
例如，可通过以下方式来判断毫米波雷达所在车辆是否在曲线运动：获取 毫米波雷达所在车辆的速度和横摆角速度yawRate，如毫米波雷达所在车辆的速度＞曲线速度阈值，且横摆角速度yawRate的绝对值＞曲线横摆角速度阈值，则判定毫米波雷达所在车辆在曲线运动。其中，曲线速度阈值可以设定为0.1m/s，曲线横摆角速度阈值可以设定为0.1-0.15rad/s。
根据varianceY和all_mean_varianceY判断本帧航迹数据中目标车辆的航迹的轨迹状态并更新，参见图3，航迹的轨迹状态共有直线、曲线和未知unknow三种，由于本帧航迹数据中目标车辆的航迹的轨迹状态的判断依据与上一帧航迹数据中目标车辆的航迹的轨迹状态相关，所以需要先查看上一帧航迹数据中目标车辆的航迹的轨迹状态。包括：
若上一帧航迹数据中目标车辆的航迹的轨迹状态为未知unknow，则判断本帧航迹数据中目标车辆的航迹是否满足轨迹为直线运动的条件，轨迹为直线运动的条件为：varianceY＜varianceY_threshold，且本帧航迹数据中目标车辆的横向速度vy的绝对值abs_vy满足abs_vy≤vy_threshold。其中，varianceY_threshold为直线状态的方差阈值，vy_threshold为直线状态的横向速度阈值。若本帧航迹数据中目标车辆的航迹满足轨迹为直线运动的条件，则将本帧航迹数据中目标车辆的航迹的轨迹状态更新为直线。
若上一帧航迹数据中目标车辆的航迹的轨迹状态为曲线，则判断本帧航迹数据中目标车辆的航迹是否满足轨迹状态从曲线到直线的转变条件，轨迹状态从曲线到直线的转变条件为：varianceY＜varianceY_threshold，且本帧航迹数据中目标车辆的横向速度vy的绝对值≤vy_threshold。该判断条件与上述轨迹为直线运动的条件相同，若本帧航迹数据中目标车辆的航迹满足轨迹为从曲线到直线的转变条件，则将本帧航迹数据中目标车辆的航迹的轨迹状态更新为直线。当满足上述两种条件将本帧航迹数据中目标车辆的航迹的轨迹状态更新为直线后，还将本帧航迹数据中目标车辆的航迹的straightCount加1，并重新开始统计历史航迹的y值，航迹的sum_dot_num＝1，sum_y为本帧航迹数据中目标车辆的横向位置y。
若上一帧航迹数据中目标车辆的航迹的轨迹状态为直线，则判断本帧航迹数据中目标车辆的航迹是否满足轨迹为曲线运动的条件，轨迹为曲线运动的条件为：all_mean_varianceY＞all_varianceY_threshold，同时，本帧航迹数据中目标车辆的横向速度vy的绝对值＞curve_vy_threshold。其中，all_varianceY_threshold为曲线状态的方差阈值，curve_vy_threshold为曲线状态的横向速度阈值，若本帧航迹数据中目标车辆的航迹满足轨迹为曲线运动的条件，则将其状态更新为曲线，航迹的straightCount置0，如果确定本帧航迹数据中目标车辆的航迹不满足轨迹为曲线运动的条件，将状态更新为直线，航迹的straightCount加1。当本帧航迹数据中目标车辆的航迹的轨迹状态为曲线时，即表示该目标车辆在变道。
本申请实施例的直线状态的横向速度阈值vy_threshold、曲线状态的横向速度阈值Curve_vy_threshold、直线状态的方差阈值varianceY_threshold和曲线状态的方差阈值all_mean_varianceY_threshold可以设定为一个定值，但目标车辆 航迹与毫米波雷达所在车辆的纵向距离，以及目标车辆航迹的速度会对识别精度造成一定的影响，为了降低这一影响，还可以对其进行动态调整，但考虑到直线状态的判断相对较严格，所以仅将直线状态的横向速度阈值vy_threshold设定为一个定值，可以为0.5m/s。而曲线状态的横向速度阈值Curve_vy_threshold的初始值可设定为0.5m/s，直线状态的方差阈值varianceY_threshold的初始值可设定为0.01平方米，曲线状态的方差阈值all_mean_varianceY_threshold的初始值可设定为0.7平方米。然后根据本帧航迹数据中目标车辆的纵向位置x的绝对值abs_x对直线状态的方差阈值varianceY_threshold和曲线状态的方差阈值all_mean_varianceY_threshold进行动态调整。根据本帧航迹数据中目标车辆的纵向速度vx的绝对值abs_vx对曲线状态的横向速度阈值Curve_vy_threshold进行动态调整。动态调整的方式如下：
当abs_x＞150m时，直线状态的方差阈值varianceY_threshold调整为0.03平方米，曲线状态的方差阈值all_mean_varianceY_threshold调整为1.8平方米；
当150m＞abs_x＞90m时，直线状态的方差阈值varianceY_threshold调整为0.02平方米，曲线状态的方差阈值all_mean_varianceY_threshold调整为1.6平方米；
当90m＞abs_x＞40m时，直线状态的方差阈值all_mean_varianceY_threshold调整为1.2平方米；
当abs_vx＞10m/s时，曲线状态的横向速度阈值Curve_vy_threshold调整为0.8m/s。需要说明的是，动态调整的方式不限于上述方式，也可采用如线性变化的方式进行动态调整。
本申请提供了一种电子设备，包括：
一个或多个处理器；
存储器，设置为存储一个或多个程序；
当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现如前所述的基于毫米波雷达方差统计的目标运动状态判断方法。
本申请提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现如前所述的基于毫米波雷达方差统计的目标运动状态判断方法。
根据本申请的实施例，上文参考流程图描述的过程可以由电子设备实现，电子设备包括处理器(例如中央处理器、图形处理器等)，其可以根据存储在存储器中的程序而执行多种适当的动作和处理。
根据本申请的实施例，上文参考流程图描述的过程可以被实现为计算机软件程序。例如，本申请的实施例包括一种计算机程序产品，其包括承载在非暂态计算机可读介质上的计算机程序，该计算机程序包含设置为执行流程图所示的方法的程序代码。
上述计算机可读存储介质可以是电子设备中所包含的；也可以是单独存在，而未装配入电子设备中。计算机可读存储介质可以是非暂态计算机可读存储介质。
本申请通过实时基于方差统计的方式对目标车辆轨迹识别，当目标车辆的横向速度较小，并且滑窗内目标车辆(目标车辆航迹)的横向位置方差较小，表示目标车辆为直线运动，识别目标车辆为直线运动后，从这帧航迹数据开始重新累计目标车辆航迹的历史滤波数，累加目标车辆航迹的横向位置之和；滑窗内目标车辆航迹的横向位置基于历史横向滤波均值计算方差，当该方差大于阈值，并且目标车辆航迹滤波速度大于速度阈值，识别目标车辆为曲线运动，即目标车辆发生变道行为。通过实时采集雷达数据及路况视频，分析数据，这种方法识别目标车辆轨迹状态直线或变道更加准确，并且为后续目标车辆航迹的平滑处理打下了基础。

Claims (7)

  1. 一种基于毫米波雷达方差统计的目标车辆运动状态的判断方法，包括：
    以滑窗的方式存储所述目标车辆在多帧航迹数据中的历史位置，滑窗的窗口数为cycle_thr，每帧航迹数据中包括所述目标车辆的航迹；
    根据所述历史位置累加计算所述目标车辆在所述多帧航迹数据中的横向位置y之和sum_y，并累加计算所述所述目标车辆的横向位置y的次数sum_dot_num，其中，所述sum_y和sum_dot_num在所述目标车辆初始时置为0；
    判断滑窗内存储所述目标车辆的横向位置y的点迹数是否大于cycle_thr，基于所述点迹数小于或等于cycle_thr的判断结果，将所述目标车辆的航迹在本帧航迹数据中的轨迹状态判断为未知unknow，判断过程结束；基于所述点迹数大于cycle_thr的判断结果，判断sum_dot_num是否大于cycle_thr；基于所述sum_dot_num小于或等于cycle_thr的判断结果，判断过程结束；
    基于sum_dot_num大于cycle_thr的判断结果，判断所述毫米波雷达所在车辆是否在进行曲线运动，基于所述毫米波雷达所在车辆在进行曲线运动的判断结果，将所述目标车辆的航迹在所述本帧航迹数据中的轨迹状态判断为未知unknow，并将所述目标车辆的航迹的直线统计变量strightCount、sum_y和sum_dot_num置为零；基于所述毫米波雷达所在车辆未在进行曲线运动的判断结果，计算滑窗内所述目标车辆的横向位置y的方差varianceY，根据sum_y和sum_dot_num计算所述目标车辆的的历史横向位置y的均值all_mean_y，计算滑窗内所述目标车辆的横向位置y基于均值all_mean_y的方差all_mean_varianceY；
    根据varianceY和all_mean_varianceY判断所述目标车辆的航迹在所述本帧航迹数据中的轨迹状态并更新，判断过程结束；
    其中，所述根据varianceY和all_mean_varianceY判断所述目标车辆的航迹在本帧航迹数据中的轨迹状态并更新，包括：
    响应于确定上一帧航迹数据中所述目标车辆的航迹的轨迹状态为未知unknow，判断所述本帧航迹数据中所述目标车辆的航迹是否满足轨迹为直线运动的条件，所述轨迹为直线运动的条件为：varianceY＜varianceY_threshold，且所述本帧航迹数据中目标车辆的横向速度vy的绝对值abs_vy满足abs_vy≤vy_threshold；其中，varianceY_threshold为轨迹直线运动状态的方差阈值，vy_threshold为轨迹直线运动状态的横向速度阈值；
    响应于确定所述上一帧航迹数据中目标车辆的航迹的轨迹状态为曲线，判断所述本帧航迹数据中目标车辆的航迹是否满足轨迹状态从曲线到直线的转变条件，所述轨迹状态从曲线到直线的转变条件为：varianceY＜varianceY_threshold，且所述本帧航迹数据中目标车辆的横向速度vy的绝对值abs_vy满足abs_vy≤vy_threshold；
    响应于确定所述本帧航迹数据中目标车辆的航迹满足轨迹为直线运动的条件或所述本帧航迹数据中目标车辆的航迹满足轨迹从曲线到直线的转变条件，将所述本帧航迹数据中目标车辆的航迹的轨迹状态更新为直线，航迹的straightCount加1，并重新开始统计所述目标车辆历史航迹的y值，所述目标车辆的航迹的sum_dot_num＝1，sum_y为本帧航迹数据中目标车辆的横向位置y；
    响应于确定所述上一帧航迹数据中目标车辆的航迹的轨迹状态为直线，判断所述本帧航迹数据中目标车辆的航迹是否满足轨迹为曲线运动的条件，所述轨迹为曲线运动的条件为：all_mean_varianceY＞all_varianceY_threshold，且所述本帧航迹数据中目标车辆的横向速度vy的绝对值abs_vy满足abs_vy＞curve_vy_threshold；其中，all_varianceY_threshold为轨迹曲线运动状态的方差阈值，curve_vy_threshold为轨迹曲线运动状态的横向速度阈值；
    响应于确定所述本帧航迹数据中目标车辆的航迹满足轨迹为曲线运动的条件，将所述本帧航迹数据中目标车辆的航迹的轨迹状态更新为曲线，航迹的straightCount置0；响应于确定所述本帧航迹数据中目标车辆的航迹不满足轨迹为曲线运动的条件，将所述本帧航迹数据中目标车辆的航迹的轨迹状态更新为直线，航迹的straightCount加1。
  2. 根据权利要求1所述的方法，其中，所述轨迹直线运动状态的横向速度阈值vy_threshold设定为0.5m/s，所述轨迹曲线运动状态的横向速度阈值Curve_vy_threshold的初始值为0.5m/s，所述轨迹直线运动状态的方差阈值varianceY_threshold的初始值为0.01平方米，所述轨迹曲线运动状态的方差阈值all_mean_varianceY_threshold的初始值为0.7平方米；
    所述方法还包括：
    根据所述本帧航迹数据中目标车辆的纵向位置x的绝对值abs_x对所述轨迹直线状态的方差阈值varianceY_threshold和所述轨迹曲线状态的方差阈值all_mean_varianceY_threshold进行动态调整；通过所述本帧航迹数据中目标车辆的纵向速度vx的绝对值abs_vx对所述轨迹曲线状态的横向速度阈值Curve_vy_threshold进行动态调整。
  3. 根据权利要求2所述的方法，其中，所述根据所述本帧航迹数据中目标车辆的纵向位置x的绝对值abs_x对所述轨迹直线状态的方差阈值varianceY_threshold和所述轨迹曲线状态的方差阈值all_mean_varianceY_threshold进行动态调整，包括：
    响应于确定abs_x＞150m，所述轨迹直线状态的方差阈值varianceY_threshold调整为0.03平方米，所述轨迹曲线状态的方差阈值all_mean_varianceY_threshold调整为1.8平方米；
    响应于确定150m＞abs_x＞90m，所述轨迹直线状态的方差阈值varianceY_threshold调整为0.02平方米，所述轨迹曲线状态的方差阈值all_mean_varianceY_threshold调整为1.6平方米；
    响应于确定90m＞abs_x＞40m，所述轨迹直线状态的方差阈值all_mean_varianceY_threshold调整为1.2平方米；
    所述通过所述本帧航迹数据中目标车辆的纵向速度vx的绝对值abs_vx对所述轨迹曲线状态的横向速度阈值Curve_vy_threshold进行动态调整，包括：
    响应于确定abs_vx＞10m/s，所述轨迹曲线状态的横向速度阈值Curve_vy_threshold调整为0.8m/s。
  4. 根据权利要求1所述的方法，其中，所述基于sum_dot_num大于cycle_thr 的判断结果，判断所述毫米波雷达所在车辆是否在进行曲线运动，包括：
    获取所述毫米波雷达所在车辆的速度和横摆角速度yawRate，响应于确定所述毫米波雷达所在车辆的速度大于曲线速度阈值，且所述毫米波雷达所在车辆的横摆角速度yawRate的绝对值大于曲线横摆角速度阈值，确定所述毫米波雷达所在车辆在进行曲线运动。
  5. 根据权利要求4所述的方法，其中，所述曲线速度阈值为0.1m/s。
  6. 根据权利要求4所述的方法，其中，所述曲线横摆角速度阈值为0.1-0.15rad/s。
  7. 根据权利要求1所述的方法，其中，滑窗的窗口数cycle_thr包括5个。
PCT/CN2022/088408 2022-01-05 2022-04-22 基于毫米波雷达方差统计的目标车辆运动状态判断方法 WO2023130620A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
DE112022000008.2T DE112022000008T5 (de) 2022-01-05 2022-04-22 Verfahren zur Bestimmung eines Bewegungszustands eines Zielfahrzeugs mit Varianzmaßzahlen beruhend auf Millimeterwellenradar

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210003432.2A CN114019497B (zh) 2022-01-05 2022-01-05 一种基于毫米波雷达方差统计的目标变道识别方法
CN202210003432.2 2022-01-05

Publications (1)

Publication Number Publication Date
WO2023130620A1 true WO2023130620A1 (zh) 2023-07-13

Family

ID=80069672

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/088408 WO2023130620A1 (zh) 2022-01-05 2022-04-22 基于毫米波雷达方差统计的目标车辆运动状态判断方法

Country Status (3)

Country Link
CN (1) CN114019497B (zh)
DE (1) DE112022000008T5 (zh)
WO (1) WO2023130620A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114019497B (zh) * 2022-01-05 2022-03-18 南京楚航科技有限公司 一种基于毫米波雷达方差统计的目标变道识别方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109581353A (zh) * 2018-11-27 2019-04-05 北京信息科技大学 一种基于汽车雷达的多目标跟踪方法及系统
CN109858460A (zh) * 2019-02-20 2019-06-07 重庆邮电大学 一种基于三维激光雷达的车道线检测方法
CN110596694A (zh) * 2019-09-20 2019-12-20 吉林大学 一种复杂环境雷达多目标跟踪和道路行驶环境预测方法
CN112198503A (zh) * 2020-10-16 2021-01-08 无锡威孚高科技集团股份有限公司 一种目标航迹预测优化方法、装置及雷达系统
CN113791414A (zh) * 2021-08-25 2021-12-14 南京市德赛西威汽车电子有限公司 一种基于毫米波车载雷达视图的场景识别方法
CN114019497A (zh) * 2022-01-05 2022-02-08 南京楚航科技有限公司 一种基于毫米波雷达方差统计的目标变道识别方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020112827A2 (en) * 2018-11-26 2020-06-04 Mobileye Vision Technologies Ltd. Lane mapping and navigation
CN111192478B (zh) * 2020-04-10 2020-07-21 南京楚航科技有限公司 一种基于毫米波雷达的道路车位状态识别方法
CN112085077B (zh) * 2020-08-28 2023-10-31 东软集团股份有限公司 车辆变道的确定方法、装置、存储介质和电子设备
CN112249008B (zh) * 2020-09-30 2021-10-26 南京航空航天大学 针对复杂动态环境的无人驾驶汽车预警方法
CN113353080A (zh) * 2021-06-01 2021-09-07 上海网车科技有限公司 一种基于可行驶区域的高速公路自动变道方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109581353A (zh) * 2018-11-27 2019-04-05 北京信息科技大学 一种基于汽车雷达的多目标跟踪方法及系统
CN109858460A (zh) * 2019-02-20 2019-06-07 重庆邮电大学 一种基于三维激光雷达的车道线检测方法
CN110596694A (zh) * 2019-09-20 2019-12-20 吉林大学 一种复杂环境雷达多目标跟踪和道路行驶环境预测方法
CN112198503A (zh) * 2020-10-16 2021-01-08 无锡威孚高科技集团股份有限公司 一种目标航迹预测优化方法、装置及雷达系统
CN113791414A (zh) * 2021-08-25 2021-12-14 南京市德赛西威汽车电子有限公司 一种基于毫米波车载雷达视图的场景识别方法
CN114019497A (zh) * 2022-01-05 2022-02-08 南京楚航科技有限公司 一种基于毫米波雷达方差统计的目标变道识别方法

Also Published As

Publication number Publication date
CN114019497B (zh) 2022-03-18
DE112022000008T5 (de) 2023-11-09
CN114019497A (zh) 2022-02-08

Similar Documents

Publication Publication Date Title
KR102399963B1 (ko) 측방향 경사가 있는 도로 평면 출력
US7091838B2 (en) Lane deviation alarm system
US6489887B2 (en) Lane-keep assisting system for vehicle
CN107305632B (zh) 基于单目计算机视觉技术的目标对象距离测量方法与系统
US8175334B2 (en) Vehicle environment recognition apparatus and preceding-vehicle follow-up control system
CN104282020A (zh) 一种基于目标运动轨迹的车辆速度检测方法
CN106295459A (zh) 基于机器视觉和级联分类器的车辆检测和预警方法
CN110194162B (zh) 用于基于可用空间信号来检测靠近切入车辆的系统和方法
EP2950291A1 (en) Road environment recognition system
JP2015505759A (ja) 車両の迫っている横転を検知するための方法
WO2023130620A1 (zh) 基于毫米波雷达方差统计的目标车辆运动状态判断方法
JP2009252198A (ja) 走行環境推測装置、方法及びプログラム並びに車線逸脱警報装置及び操舵アシスト装置
CN105300390B (zh) 障碍物体运动轨迹的确定方法及装置
US6956959B2 (en) Apparatus for recognizing environment
CN112036422A (zh) 一种基于多传感器信息融合的航迹管理方法、系统及计算机可读介质
CN115223131A (zh) 一种自适应巡航的跟随目标车辆检测方法、装置及汽车
US20230245323A1 (en) Object tracking device, object tracking method, and storage medium
CN115123291B (zh) 一种基于障碍物识别的行为预测方法及装置
CN112991734B (zh) 一种视觉车位的车位状态检测系统
JPH11259639A (ja) 走行路認識装置
CN114662600A (zh) 一种车道线的检测方法、装置和存储介质
CN113879312A (zh) 基于多传感器融合的前向目标选择方法、装置和存储介质
CN112465865A (zh) 一种基于背景建模和IoU匹配的多目标跟踪方法
JP4847303B2 (ja) 障害物検出方法、障害物検出プログラムおよび障害物検出装置
CN112309134B (zh) 车辆速度检测方法和装置

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 112022000008

Country of ref document: DE

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

Ref document number: 22918084

Country of ref document: EP

Kind code of ref document: A1