WO2023061355A1 - 速度检测方法、装置、设备及可读存储介质 - Google Patents

速度检测方法、装置、设备及可读存储介质 Download PDF

Info

Publication number
WO2023061355A1
WO2023061355A1 PCT/CN2022/124576 CN2022124576W WO2023061355A1 WO 2023061355 A1 WO2023061355 A1 WO 2023061355A1 CN 2022124576 W CN2022124576 W CN 2022124576W WO 2023061355 A1 WO2023061355 A1 WO 2023061355A1
Authority
WO
WIPO (PCT)
Prior art keywords
velocity
detection target
detection
moment
distance
Prior art date
Application number
PCT/CN2022/124576
Other languages
English (en)
French (fr)
Inventor
丁磊
Original Assignee
华人运通(上海)自动驾驶科技有限公司
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 华人运通(上海)自动驾驶科技有限公司 filed Critical 华人运通(上海)自动驾驶科技有限公司
Publication of WO2023061355A1 publication Critical patent/WO2023061355A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO 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
    • G01S13/00Systems 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
    • G01S13/02Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/50Systems of measurement based on relative movement of target
    • G01S13/58Velocity or trajectory determination systems; Sense-of-movement determination systems
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02TCLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00Road transport of goods or passengers
    • Y02T10/10Internal combustion engine [ICE] based vehicles
    • Y02T10/40Engine management systems

Definitions

  • the present application relates to the field of measurement technology, in particular to a speed detection method, device, equipment and readable storage medium.
  • the radar is usually used to measure the distance and radial velocity of multiple reflection points of the detection target to determine its lateral velocity and longitudinal velocity.
  • the multiple reflection points are generated by the detection target’s reflection of the electromagnetic wave of the radar.
  • the detection target may not be able to generate multiple reflection points; or, the included angle between the azimuth angles of the adjacent reflection points generated by the detection target is small, which leads to the existence of the determination of the lateral velocity and the longitudinal velocity. low accuracy problem.
  • the embodiment of the present application provides a speed detection method, device, equipment and readable storage medium to solve the problems existing in related technologies.
  • the technical solution is as follows:
  • the embodiment of the present application provides a speed detection method, including:
  • the first corresponding relationship is used to characterize the relationship between the lateral velocity, longitudinal velocity and velocity angle of the detection target at the kth moment, where the velocity angle is the lateral velocity and the detection target at the kth moment The angle between the moving speeds;
  • the second corresponding relationship is used to characterize the relationship between the azimuth angle, lateral velocity, longitudinal velocity and radial velocity of the detection target at the kth moment;
  • the lateral velocity and the longitudinal velocity are determined.
  • the embodiment of the present application provides a speed detection device, including:
  • the first establishment module is used to establish the first corresponding relationship, the first corresponding relationship is used to characterize the relationship between the lateral speed, longitudinal speed and speed angle of the detection target at the kth moment, wherein the speed angle is the lateral speed and Detect the angle between the moving speeds of the target at the kth moment;
  • the second establishing module is used to establish a second corresponding relationship, and the second corresponding relationship is used to characterize the relationship between the azimuth angle, lateral velocity, longitudinal velocity and radial velocity of the detection target at the kth moment;
  • the determining module is configured to determine the lateral velocity and the longitudinal velocity based on the first correspondence and the second correspondence.
  • an embodiment of the present application provides an electronic device, the electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions that can be executed by the at least one processor , so that at least one processor can execute the above speed detection method.
  • an embodiment of the present application provides a computer-readable storage medium, which stores computer instructions, and when the computer instructions are run on a computer, the method in any one of the above-mentioned aspects is executed.
  • the advantages or beneficial effects of the above technical solution at least include: using the first corresponding relationship and the second corresponding relationship of the detected target at the k-th moment to determine the lateral velocity and longitudinal velocity of the detected target at the k-th moment, so that the detected target only needs to
  • the speed detection can be realized by generating one reflection point, without the need to detect multiple reflection points of the target, which can reduce the influence of the distance factor and the azimuth angle resolution of the radar on the detection accuracy, thereby improving the speed detection accuracy.
  • FIG. 1 is a schematic diagram of radar detection for a detection target in the related art
  • FIG. 2A is a schematic flow chart of a speed detection method according to an embodiment of the present application.
  • FIG. 2B is a schematic diagram of a detection target according to an embodiment of the present application.
  • Fig. 3 is a schematic flow chart of determining a first correspondence according to an embodiment of the present application.
  • FIG. 4A is a schematic flow diagram of obtaining a first position according to an embodiment of the present application.
  • FIG. 4B is a schematic diagram of obtaining a first preselected detection frame according to an embodiment of the present application.
  • FIG. 4C is a schematic diagram of acquiring a semantically segmented image according to an embodiment of the present application.
  • FIG. 5 is a schematic flow chart of detecting the azimuth angle of a target according to an embodiment of the present application
  • FIG. 6 is a schematic flowchart of a speed detection method according to another embodiment of the present application.
  • FIG. 7 is a schematic diagram of various application scenarios according to an embodiment of the present application.
  • FIG. 8 is a structural block diagram of a speed detection device according to an embodiment of the present application.
  • FIG. 9 is a block diagram of an electronic device used to implement the speed detection method of the embodiment of the present application.
  • the radar for the detection of the lateral velocity and longitudinal velocity of the detection target, the radar is usually used to emit electromagnetic waves to the detection target, so that multiple reflection points of the detection target reflect the electromagnetic waves, so that the radar can detect Radial velocity and azimuth for multiple reflection points. Furthermore, multiple correspondences between the radial velocity, the transverse velocity, and the longitudinal velocity are constructed through the azimuth, and the transverse velocity and the longitudinal velocity can be determined by using the multiple correspondences to realize detection. Among them, multiple correspondences between radial velocity, lateral velocity, and longitudinal velocity are constructed by azimuth angle as shown in formula (1):
  • V ri is the motion velocity V of the detection target and the radial velocity of the P i reflection point
  • ⁇ i is the azimuth angle of the P i reflection point of the detection target
  • v x is the longitudinal velocity of the detection target
  • v y is the detection target The lateral velocity of the target.
  • FIG. 1 shows the situation that the detection target produces a first reflection point P1 and a second reflection point P2 .
  • this detection method is not applicable when the detection target is far away from the radar and the azimuth angle interval between adjacent reflection points is small. For example, when the distance between the detection target and the radar exceeds 80m, the detection target cannot generate multiple reflection points; or, when the azimuth interval of adjacent reflection points is small, the radar cannot distinguish them. Therefore, there is a problem that the accuracy of speed detection is low.
  • the speed detection method may include:
  • Step S201 establishing a first corresponding relationship
  • the first corresponding relationship is used to characterize the relationship between the lateral velocity, longitudinal velocity and velocity angle of the detected target at the kth moment, wherein the velocity angle is the lateral velocity and the motion of the detected target Angle between velocities.
  • the transverse velocity is the transverse component of the movement velocity
  • the longitudinal velocity is the longitudinal component of the movement velocity.
  • the first correspondence can be represented by the following formula (2):
  • ⁇ k , v x,k , and v y,k are the velocity angle, longitudinal velocity, and lateral velocity of the detection target at the kth moment in turn.
  • Step S202 establishing a second corresponding relationship, the second corresponding relationship is used to characterize the relationship between the azimuth angle, lateral velocity, longitudinal velocity and radial velocity of the detection target at the kth moment.
  • Radial velocity can be obtained by radar or radar recorder, for example, millimeter wave radar, lidar, lidar recorder (Laser Radar Recorder, LRR), etc.
  • the second corresponding relationship can be expressed by the following formula (3):
  • v r,k and ⁇ k are the radial velocity and azimuth angle of the detection target at the k-th moment in sequence, and v r,k represents the radial component of the motion velocity V k of the detection target at the k-th moment.
  • Step S203 based on the first corresponding relationship and the second corresponding relationship, determine the lateral velocity and the longitudinal velocity.
  • the above scheme uses the first corresponding relationship and the second corresponding relationship of the detected target at the k-th moment to determine the lateral velocity and longitudinal velocity of the detected target at the k-th moment, so that the detected target only needs to generate one reflection point to realize speed detection , without the need to detect multiple reflection points of the target, which can reduce the influence of the distance factor and the azimuth angle resolution of the radar on the detection accuracy, thereby improving the accuracy of speed detection.
  • step S201 may include:
  • Step S301 Obtain the first position of the detection target at time k-1 and the second position at time k.
  • the first position and the second position may be obtained by using a position sensor
  • the position sensor may be a camera, a laser sensor, or the like.
  • Step S302 determining the displacement angle between the first position and the second position
  • Step S303 using the displacement angle as the velocity angle.
  • the displacement angle ⁇ k of the detected target can be determined by the following formula (4):
  • the displacement angle ⁇ k is used as the velocity angle through the following formula (6):
  • the lateral distance change y k -y k-1 between the second position and the first position, the longitudinal distance change x k -x k-1 and the distance between the kth moment and the k-1th moment are usually used.
  • the time interval ⁇ T is used to determine the lateral velocity of the detection target at the kth moment.
  • the lateral velocity v y,k can be determined by the following formula (7):
  • the time error can be eliminated, which is beneficial to reduce the speed error and improve the speed detection. the accuracy.
  • the velocity angle is determined by obtaining more than two position coordinates.
  • the interval is greater than the acquisition time interval of the camera, which can extend the measurement time window, thereby effectively eliminating time errors and improving the accuracy of speed detection; on the other hand, smoothing the average of multiple speed angles also makes speed detection more accurate.
  • obtaining the first position of the detected target at the k-1th moment may include:
  • Step S401 acquiring the first image of the detection target at time k-1.
  • the first image may be acquired by a camera, and the resolution of the camera to detect the position change of the target is greater than the resolution of the radar to detect the position change of the target.
  • Step S402 input the first image into the target detection model to obtain the first preselected detection frame;
  • the target detection model is obtained by training a deep learning network model based on a plurality of sample images;
  • the sample images include images of the detection target;
  • Step S403 determining the first mass point information from the first preselected detection frame
  • Step S404 performing coordinate transformation on the first mass point information to obtain a first position.
  • the first image is input into the target detection model, and a detection frame whose Intersection over Union (IoU) ratio (Intersection over Union, IoU) with the label frame 410 is determined to be equal to or greater than 0.8 is the first preselected detection frame .
  • the first preselected detection frame 420 may be a rectangular frame, and the pixel coordinates of the ith vertex Ai of the first preselected detection frame 420 are (u 1i , v 1i ), where 1 ⁇ i ⁇ 4.
  • the pixel coordinates (u 1 , v 1 ) of the first mass point B1 are determined from the pixel coordinates of the vertices of the first preselected detection frame 320, wherein,
  • M is the transformation matrix determined by the internal and external parameters of the camera.
  • the camera can be used to acquire the second image of the detection target at the k-th moment, and the second position coordinates (x k , y k ) of the detection target at the k-th moment can be determined using a determination method similar to the above-mentioned first position coordinates.
  • the calculation speed is fast, which can not only improve the speed of determining position information, but also remove redundant pixels in the first image, so that the first preselected
  • the detection frame only retains the pixels of the detection target to the greatest extent, which is conducive to improving the accuracy of the first particle information, thereby improving the accuracy of obtaining the first position.
  • determining the first mass point information from the first preselected detection frame may include:
  • the geometric center of the first preselected detection frame 420 may be used as the pixel coordinates of the first particle B1 .
  • the first preselected detection frame is semantically segmented, redundant pixels in the first preselected detection frame are further removed, pixels of the detection target are retained, and a semantically segmented image 430 with the detection target is obtained; based on semantic segmentation For the circumscribed circle outline or the inscribed circle outline of the image 430 (not shown in the figure), determine the pixel coordinates corresponding to the center of the circumscribed circle outline or the inscribed circle outline as the pixel coordinates of the first mass point.
  • the azimuth of the detection target can be determined by the following steps:
  • Step S501 acquiring the second position and radial distance of the detection target at the kth moment
  • Step S502 Determine the azimuth of the detection target according to the second position and the radial distance.
  • the second position is determined by the second image acquired by the camera at the kth moment.
  • the radial distance is determined by the radar sending electromagnetic waves to a reflection point (for example, the first reflection point P 1 ) of the detection target and receiving the reflected electromagnetic waves.
  • Determining the azimuth of the detection target at the kth moment may include:
  • the azimuth of the detected target at the kth moment is determined.
  • the azimuth ⁇ k is determined by the following formula (9):
  • y k is the lateral distance between the detection target and the camera
  • r k is the radial distance of the detection target at the kth moment.
  • the following formula (10) can be used to determine the distance between the radial velocity v r,k and the lateral velocity v y,k , the longitudinal velocity v x,k and the azimuth angle ⁇ k of the detection target at the kth moment Relationship:
  • the lateral distance between the detection target and the camera obtained by the camera and the radial distance obtained by the radar are used to determine the azimuth of the detection target, and the accuracy of the azimuth is higher than that of the azimuth measured by the radar.
  • Using the azimuth angle instead of the azimuth angle obtained by radar measurement is beneficial to improve the accuracy of speed detection.
  • the speed detection method may also include:
  • Step S601 based on the first corresponding relationship and the second corresponding relationship, establishing a measurement model of the detection target;
  • Step S602 based on a preset constant velocity motion model (Constant Velocity, CV), establish a process model for detecting the target;
  • Constant Velocity, CV Constant Velocity
  • Step S603 based on the measurement model and the process model, use the Unscented Kalman Filter (UKF) to estimate the optimal lateral velocity and optimal longitudinal velocity of the detection target at the k-th moment.
  • UPF Unscented Kalman Filter
  • the distance of the detection target at the kth moment is estimated
  • the optimal lateral speed and the optimal longitudinal speed can effectively improve the accuracy of speed detection.
  • the measurement model may include:
  • x k-1 and y k-1 are the longitudinal distance and lateral distance of the detection target at the k-1 moment respectively
  • x k , y k , r k , v r , v x , v y , ⁇ k , ⁇ k are respectively the longitudinal distance, transverse distance, radial distance, radial velocity, longitudinal velocity, transverse velocity, displacement angle/velocity angle, and azimuth angle of the detection target at the kth moment.
  • the process model includes:
  • x k , y k , v x,k , v y,k are respectively the longitudinal distance, lateral distance, longitudinal velocity, and lateral velocity of the detection target at the kth moment
  • x k-1 , y k-1 , v x,k-1 and v y,k-1 are respectively the longitudinal distance, lateral distance, longitudinal velocity, and lateral velocity of the detection target at the k-1st moment
  • ⁇ T is the distance between the k-th moment and the k-1th moment. time interval between.
  • Fig. 7 is a schematic diagram of various application scenarios according to an embodiment of the present application.
  • the speed detection method of the embodiment of the present application can be applied to an automatic driving vehicle.
  • it may be applicable to scenarios such as lane-changing and cutting-in, lane-changing and cutting-out of the target vehicle 710, and crossroad passing.
  • the target vehicle door 710 can accurately detect the lateral velocity and longitudinal velocity of the detection target (including detection of vehicles, pedestrians, etc.), which helps the target vehicle 710 to better perform path planning or Obstacle avoidance, etc.
  • Fig. 8 is a structural block diagram of a speed detection device according to an embodiment of the present application. As shown in Figure 8, the speed detection device 800 may include:
  • the first establishing module 810 is used to establish a first corresponding relationship, and the first corresponding relationship is used to characterize the relationship between the lateral speed, longitudinal speed and speed angle of the detection target at the kth moment, wherein the speed angle is the lateral speed and the angle between the motion velocity of the detection target at the kth moment;
  • the second establishing module 820 is used to establish a second corresponding relationship, and the second corresponding relationship is used to characterize the relationship between the azimuth angle, lateral velocity, longitudinal velocity and radial velocity of the detected target at the kth moment;
  • the determination module 830 is configured to determine the lateral velocity and the longitudinal velocity based on the first correspondence and the second correspondence.
  • the first establishment module 810 may include:
  • the first acquiring submodule is used to acquire the first position of the detection target at the k-1th moment and the second position at the kth moment;
  • the first determining submodule is used to determine the displacement angle between the first position and the second position
  • the first acquisition submodule may include:
  • An acquisition unit configured to acquire the first image of the detection target at the k-1th moment
  • the recognition unit is used to input the first image into the target detection model to obtain the first preselected detection frame;
  • the target detection model is obtained based on a plurality of sample images trained by a deep learning network model;
  • the sample image includes an image of the detection target;
  • a determining unit configured to determine the first mass point information from the first preselected detection frame
  • the conversion unit is configured to perform coordinate conversion on the first mass point information to obtain the first position.
  • the determination unit can be used for:
  • the second establishment module 820 may include:
  • the second acquisition submodule is used to acquire the second position and radial distance of the detection target at the kth moment
  • the second determining submodule is used to determine the azimuth of the detection target according to the second position and the radial distance.
  • the speed detection device may also include:
  • a measurement model establishment module configured to establish a measurement model of the detection target based on the first correspondence and the second correspondence
  • a process model establishment module used to establish a process model of the detection target based on a preset uniform motion model
  • the estimation module is used for estimating the optimal lateral velocity and the optimal longitudinal velocity of the detection target at time k by using the unscented Kalman filter based on the measurement model and the process model.
  • the measurement model may include:
  • x k-1 and y k-1 are the longitudinal distance and lateral distance of the detection target at the k-1 moment respectively
  • x k , y k , r k,R , v r, k , v x,k , v y, k , ⁇ k , and ⁇ k are the longitudinal distance, transverse distance, radial distance, radial velocity, longitudinal velocity, transverse velocity, displacement angle/velocity angle, and azimuth angle of the detected target at the kth moment, respectively.
  • the process model includes:
  • x k , y k , v x,k , v y,k are respectively the longitudinal distance, lateral distance, longitudinal velocity, and lateral velocity of the detection target at the kth moment
  • x k-1 , y k-1 , v x,k-1 and v y,k-1 are respectively the longitudinal distance, lateral distance, longitudinal velocity, and lateral velocity of the detection target at the k-1st moment
  • ⁇ T is the distance between the k-th moment and the k-1th moment. time interval between.
  • FIG. 9 is a block diagram of an electronic device used to implement the speed detection method of the embodiment of the present application.
  • the electronic device includes: a memory 910 and a processor 920 , and instructions executable on the processor 920 are stored in the memory 910 .
  • the processor 920 executes the instruction, the speed detection method in the above-mentioned embodiments is realized.
  • the number of memory 910 and processor 920 may be one or more.
  • the electronic device is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers.
  • Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smart phones, wearable devices, and other similar computing devices.
  • mobile devices such as personal digital processing, cellular telephones, smart phones, wearable devices, and other similar computing devices.
  • the components shown herein, their connections and relationships, and their functions, are by way of example only, and are not intended to limit implementations of the applications described and/or claimed herein.
  • the electronic device may also include a communication interface 930 for communicating with external devices for interactive data transmission.
  • the various devices are interconnected using different buses and may be mounted on a common motherboard or otherwise as desired.
  • the processor 920 may process instructions executed within the electronic device, including instructions stored in or on the memory to display graphical information of a GUI on an external input/output device such as a display device coupled to an interface.
  • multiple processors and/or multiple buses may be used with multiple memories and multiple memories, if desired.
  • multiple electronic devices may be connected, with each device providing some of the necessary operations (eg, as a server array, a set of blade servers, or a multi-processor system).
  • the bus can be divided into address bus, data bus, control bus and so on. For ease of representation, only one thick line is used in FIG. 9 , but it does not mean that there is only one bus or one type of bus.
  • the memory 910, processor 920, and communication interface 930 are integrated on one chip, then the memory 910, processor 920, and communication interface 930 can communicate with each other through the internal interface.
  • processor can be a central processing unit (Central Processing Unit, CPU), and can also be other general-purpose processors, digital signal processors (Digital Signal Processing, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), Field Programmable Gate Array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general purpose processor may be a microprocessor or any conventional processor or the like. It should be noted that the processor may be a processor supporting Advanced RISC Machines (ARM) architecture.
  • the embodiment of the present application provides a computer-readable storage medium (such as the above-mentioned memory 910 ), which stores computer instructions, and the program implements the method provided in the embodiment of the present application when executed by a processor.
  • a computer-readable storage medium such as the above-mentioned memory 910
  • the memory 910 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application program required by at least one function; use of the created data, etc.
  • the memory 910 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices.
  • the memory 910 may optionally include a memory that is remotely located relative to the processor 920, and these remote memories may be connected to electronic devices for implementing the speed detection method through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
  • first and second are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features.
  • the features defined as “first” and “second” may explicitly or implicitly include at least one of these features.
  • “plurality” means two or more, unless otherwise specifically defined.
  • each part of the present application may be realized by hardware, software, firmware or a combination thereof.
  • various steps or methods may be implemented by software or firmware stored in memory and executed by a suitable instruction execution system. All or part of the steps of the method in the above embodiments can be completed by instructing related hardware through a program.
  • the program can be stored in a computer-readable storage medium. When the program is executed, it includes one of the steps of the method embodiment or its combination.
  • each functional unit in each embodiment of the present application may be integrated into one processing module, each unit may exist separately physically, or two or more units may be integrated into one module.
  • the above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. If the above-mentioned integrated modules are implemented in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium.
  • the storage medium may be a read-only memory, a magnetic disk or an optical disk, and the like.

Landscapes

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

Abstract

本申请提出一种速度检测方法、装置、设备及可读存储介质,其中,该速度检测方法包括:建立第一对应关系,第一对应关系用于表征检测目标在第k时刻的横向速度、纵向速度以及速度夹角之间的关系,其中,速度夹角为横向速度和检测目标在第k时刻的运动速度之间的夹角;建立第二对应关系,第二对应关系用于表征检测目标在第k时刻的方位角、横向速度、纵向速度以及径向速度之间的关系;基于第一对应关系和第二对应关系,确定出横向速度和纵向速度。本申请的技术方案能够有效提高速度检测的准确度。

Description

速度检测方法、装置、设备及可读存储介质
本申请要求于2021年10月11日提交中国专利局、申请号为202111180088.6、发明名称为“速度检测方法、装置、设备及可读存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及测量技术领域,尤其涉及一种速度检测方法、装置、设备及可读存储介质。
背景技术
目前,通常利用雷达测量检测目标的多个反射点的距离和径向速度来确定其横向速度和纵向速度,其中,多个反射点为检测目标对雷达的电磁波进行反射而产生。然而,受距离因素影响,检测目标可能无法产生多个反射点;或者,检测目标所产生的相邻反射点的方位角之间的夹角较小,这就导致横向速度和纵向速度的确定存在准确度低的问题。
发明内容
本申请实施例提供一种速度检测方法、装置、设备及可读存储介质,以解决相关技术存在的问题,技术方案如下:
第一方面,本申请实施例提供了一种速度检测方法,包括:
建立第一对应关系,第一对应关系用于表征检测目标在第k时刻的横向速度、纵向速度以及速度夹角之间的关系,其中,速度夹角为横向速度和检测目标在第k时刻的运动速度之间的夹角;
建立第二对应关系,第二对应关系用于表征检测目标在第k时刻的方位角、横向速度、纵向速度以及径向速度之间的关系;
基于第一对应关系和第二对应关系,确定出横向速度和纵向速度。
第二方面,本申请实施例提供了一种速度检测装置,包括:
第一建立模块,用于建立第一对应关系,第一对应关系用于表征检测目标在第k时刻的横向速度、纵向速度以及速度夹角之间的关系,其中,速度夹角为横向速度和检测目标在第k时刻的运动速度之间的夹角;
第二建立模块,用于建立第二对应关系,第二对应关系用于表征检测目标在第k时刻的方位角、横向速度、纵向速度以及径向速度之间的关系;
确定模块,用于基于第一对应关系和第二对应关系,确定出横向速度和纵向速度。
第三方面,本申请实施例提供了一种电子设备,该电子设备包括:至少一个处理器;以及与至少一个处理器通信连接的存储器;其中,存储器存储有可被至少一个处理器执行的指令,以使至少一个处理器能够执行上述速度检测方法。
第四方面,本申请实施例提供了一种计算机可读存储介质,计算机可读存储介质存储计算机指令,当计算机指令在计算机上运行时,上述各方面任一种实施方式中的方法被执行。
上述技术方案中的优点或有益效果至少包括:利用检测目标在第k时刻的第一对应关系和第二对应关系,确定出检测目标在第k时刻的横向速度和纵向速度,使得检测目标只需产生一个反射点即可实现速度检测,而无需检测目标产生多个反射点,可以减少距离因素和雷达的方位角度分辨率对检测准确度的影响,从而提升速度检测的准确度。
上述概述仅仅是为了说明书的目的,并不意图以任何方式进行限制。除上述描述的示意性的方面、实施方式和特征之外,通过参考附图和以下的详细描述,本申请进一步的方面、实施方式和特征将会是容易明白的。
附图说明
在附图中,除非另外规定,否则贯穿多个附图相同的附图标记表示相同或相似的部件或元素。这些附图不一定是按照比例绘制的。应该理解,这些附图仅描绘了根据本申请公开的一些实施方式,而不应将其视为是对本申请范围的限制。
图1为相关技术中雷达针对检测目标的检测示意图;
图2A为根据本申请一实施例的速度检测方法的流程示意图;
图2B为根据本申请实施例的针对检测目标的检测示意图;
图3为根据本申请实施例的第一对应关系的确定流程示意图;
图4A为根据本申请实施例的获取第一位置的流程示意图;
图4B为根据本申请实施例的获取第一预选检测框的示意图;
图4C为根据本申请实施例的获取语义分割图像的示意图;
图5为根据本申请实施例的检测目标的方位角的一种流程示意图;
图6为根据本申请另一实施例的速度检测方法的流程示意图;
图7为根据本申请实施例的多种应用场景示意图;
图8为根据本申请实施例的速度检测装置的结构框图;
图9为用来实现本申请实施例的速度检测方法的电子设备的框图。
具体实施方式
在下文中,仅简单地描述了某些示例性实施例。正如本领域技术人员可认识到的那样,在不脱离本申请的精神或范围的情况下,可通过各种不同方式修改所描述的实施例。因此,附图和描述被认为本质上是示例性的而非限制性的。
如图1所示,在相关技术中,针对检测目标的横向速度和纵向速度的检测,通常利用雷达向检测目标发射电磁波,使得检测目标的多个反射点对电磁波进行反射,从而雷达可以检测到多个反射点的径向速度和方位角。再者,通过方位角构建径向速度与横向速度及纵向速度之间的多个对应关系,利用多个对应关系即可确定出横向速度和纵向速度,实现检测。其中,通过方位角构建径向速度与横向速度及纵向速度之间的多个对应关系如公式(1)所示:
Figure PCTCN2022124576-appb-000001
其中,V ri为检测目标的运动速度V在第P i反射点的径向速度,θ i为检测 目标的第P i反射点的方位角,v x为检测目标的纵向速度,v y为检测目标的横向速度。图1中示出检测目标产生第一反射点P 1和第二反射点P 2的情况。
但是,这种检测方式在检测目标距离雷达较远且相邻反射点之间的方位角间隔较小时并不适用,例如当检测目标与雷达之间的距离超过80m时,则检测目标无法产生多个反射点;或者,当相邻反射点的方位角间隔较小时,雷达无法分辨。因此,存在速度检测的准确度低的问题。
为解决上述技术问题,本申请提供一种速度检测方法,如图2所示,该速度检测方法可以包括:
步骤S201、建立第一对应关系,第一对应关系用于表征检测目标在第k时刻的横向速度、纵向速度以及速度夹角之间的关系,其中,速度夹角为横向速度和检测目标的运动速度之间的夹角。
其中,横向速度为运动速度的横向分量,纵向速度为运动速度的纵向分量。示例性地,如图2B所示,第一对应关系可以通过如下公式(2)表示:
Figure PCTCN2022124576-appb-000002
其中,α k、v x,k、v y,k依次为所述检测目标在第k时刻的速度夹角、纵向速度、横向速度。
步骤S202、建立第二对应关系,第二对应关系用于表征检测目标在第k时刻的方位角、横向速度、纵向速度以及径向速度之间的关系。
径向速度可以由雷达或雷达记录器获取,例如,毫米波雷达、激光雷达、激光雷达记录器(Laser Radar Recorder,LRR)等。第二对应关系可以通过如下公式(3)表示:
v r,k=v x,k*cos(θ k)+v y,k*sin(θ k)     (3)
其中,v r,k、θ k依次为检测目标在第k时刻的径向速度、方位角,v r,k表示检测目标在第k时刻的运动速度V k的径向分量。
步骤S203、基于第一对应关系和第二对应关系,确定出横向速度和纵向速度。
上述方案,利用检测目标在第k时刻的第一对应关系和第二对应关系,确定出检测目标在第k时刻的横向速度和纵向速度,使得检测目标只需产生一个反射点即可实现速度检测,而无需检测目标产生多个反射点,可以减少距离因素和雷达的方位角度分辨率对检测准确度的影响,从而提升速度检测的准确度。
在一种实施方式中,如图3所示,步骤S201可以包括:
步骤S301、获取检测目标在第k-1时刻的第一位置,以及在第k时刻的第二位置。
示例性地,第一位置和第二位置可以利用位置传感器获取,位置传感器可以是摄像头、激光传感器等。
步骤S302、确定第一位置与第二位置之间的位移夹角;
步骤S303、将位移夹角作为速度夹角。
在一个示例中,请参考图2B,摄像头和雷达所在位置相同(例如,激光雷达记录器与摄像头所在位置相同),可以通过如下公式(4)确定出检测目标的位移夹角α k
Figure PCTCN2022124576-appb-000003
在另一个示例中,第k时刻与第k-1时刻之间的时间间隔ΔT大于摄像头的采集时间间隔,则在检测目标从第一位置运动至第二位置的过程中,可以获取n个位置坐标(x m,y m),其中,m=1,2,…,n,n为整数,且n≥1,可以通过如下公式(5)确定位移夹角α k
Figure PCTCN2022124576-appb-000004
再者,请参考图2B,通过如下公式(6)将位移夹角α k作为速度夹角:
Figure PCTCN2022124576-appb-000005
在相关技术中,通常利用第二位置与第一位置之间的横向距离变化y k-y k-1、纵向距离变化x k-x k-1及第k时刻与第k-1时刻之间的时间间 隔ΔT来确定检测目标在第k时刻的横向速度。例如,横向速度v y,k可以通过如下公式(7)确定:
Figure PCTCN2022124576-appb-000006
但是,由于ΔT可能不准确,使得横向速度的确定容易受时间误差影响,导致横向速度的误差较大。例如,y k-y k-1=0.6m,ΔT=30ms,则v y,k=20m/s;如果ΔT产生2ms的定时误差,则获取到的ΔT=28ms,最终计算出v y,k=21.4m/s,产生1.4m/s的速度误差。
在本实施方式中,通过将检测目标从第一位置运动至第二位置的位移夹角作为检测目标在第k时刻的速度夹角,可以消除时间误差,有利于减小速度误差,提高速度检测的准确度。再者,在检测目标从第一位置运动至第二位置的过程中,通过获取两个以上的位置坐标来确定速度夹角,一方面,由于第k时刻与第k-1时刻之间的时间间隔大于摄像头的采集时间间隔,能够延长测量的时间窗口,从而有效消除时间误差,提高速度检测的准确度;另一方面,通过对多个速度夹角进行均值平滑,也使得速度检测更加准确。
在一种实施方式中,如图4A所示,获取检测目标在第k-1时刻的第一位置可以包括:
步骤S401、获取检测目标在第k-1时刻的第一图像。
其中,示例性地,第一图像可以通过摄像头采集得到,摄像头对检测目标的位置变化的分辨率大于雷达对检测目标的位置变化的分辨率。
步骤S402、将第一图像输入目标检测模型,得到第一预选检测框;目标检测模型基于多个样本图像训练深度学习网络模型而得到;样本图像中包括检测目标的图像;
步骤S403、从第一预选检测框中确定出第一质点信息;
步骤S404、将第一质点信息进行坐标转换,得到第一位置。
在一个示例中,如图4B所示,将第一图像输入目标检测模型,确定出与标注框410的交并比(Intersection over Union,IoU)等于或大于0.8的检测框为第一预选检测框。第一预选检测框420可以为矩形框,第一预选检测框420的第i个顶点Ai的像素坐标为(u 1i,v 1i),其中,1≤i≤4。 由第一预选检测框320的各顶点的像素坐标确定出第一质点B1的像素坐标(u 1,v 1),其中,
Figure PCTCN2022124576-appb-000007
基于第一质点B1的像素坐标(u 1,v 1)与世界坐标系之间的映射关系,确定出第一质点在第k-1时刻的第一位置坐标(x k-1,y k-1),其中,第一质点B1的像素坐标(u 1,v 1)与世界坐标系之间的映射关系如公式(8)所示:
Figure PCTCN2022124576-appb-000008
其中,M为摄像头的内外参数所确定的转换矩阵。
相应的,可以采用摄像头在第k时刻获取检测目标的第二图像,并采用类似上述第一位置坐标的确定方法确定出检测目标在第k时刻的第二位置坐标(x k,y k)。
在本实施方式中,通过将第一图像输入到目标检测模型进行目标检测,其计算速度快,不仅能提高位置信息得到确定速度,而且能去除第一图像中的冗余像素,使得第一预选检测框仅最大程度保留检测目标的像素,有利于提升第一质点信息的准确度,从而提高获取第一位置的准确度。
在一种实施方式中,从第一预选检测框中确定出第一质点信息,可以包括:
将第一预选检测框的中心作为第一质点信息;或者,
将第一预选检测框进行语义分割,得到具有检测目标的语义分割图像,并从语义分割图像中确定出第一质点信息。
示例性地,如图4B所示,在第一预选检测框420为矩形框的情况下,可以将第一预选检测框420的几何中心作为第一质点B1的像素坐标。
或者,如图4C所示,将第一预选检测框进行语义分割,进一步去除第一预选检测框中的冗余像素,保留检测目标的像素,得到具有检测目标的语义分割图像430;基于语义分割图像430的外接圆轮廓或内接圆轮廓(图中未示出),确定外接圆轮廓或内接圆轮廓的中心所对应的像素坐标为第一 质点的像素坐标。
基于此,有利于提高第一质点信息确定的准确度,从而提高速度检索的准确度。
在一种实施方式中,如图5所示,检测目标的方位角可以通过如下步骤确定出:
步骤S501、获取检测目标在第k时刻的第二位置和径向距离;
步骤S502、根据第二位置以及径向距离,确定检测目标的方位角。
在一个示例中,请一并参考图2B,第二位置由摄像头在第k时刻获取的第二图像确定出。径向距离由雷达向检测目标的一个反射点(例如第一反射点P 1)发送电磁波和接收反射后的电磁波确定出。
确定检测目标在第k时刻的方位角可以包括:
从第二位置坐标中确定检测目标与摄像头之间的横向距离;
基于横向距离和雷达测量得到的径向距离,确定检测目标在第k时刻的方位角。例如,通过如下公式(9)确定出方位角θ k
Figure PCTCN2022124576-appb-000009
其中,y k为检测目标与摄像头之间的横向距离,r k为检测目标在第k时刻的径向距离。
在另一个示例中,可以通过如下公式(10)确定出检测目标在第k时刻的径向速度v r,k与横向速度v y,k、纵向速度v x,k及方位角θ k之间的关系:
v r,k=v x,k*cos(θ k)+v y,k*sin(θ k)     (10)
在本实施方式中,采用摄像头获取检测目标与摄像头之间的横向距离以及雷达获取的径向距离确定检测目标的方位角,该方位角的准确度高于雷达测得的方位角的准确度,利用该方位角替代雷达测量得到的方位角,有利于提升速度检测的准确度。
在一种实施方式中,如图6所示,速度检测方法还可以包括:
步骤S601、基于第一对应关系和第二对应关系,建立检测目标的量测模型;
步骤S602、基于预设的匀速运动模型(Constant Velocity,CV),建立检测目标的过程模型;
步骤S603、基于量测模型和过程模型,采用无迹卡尔曼滤波(Unscented Kalman Filter,UKF)估计检测目标在第k时刻的最优横向速度和最优纵向速度。
基于此,通过建立检测目标的量测模型和过程模型,并采用无迹卡尔曼滤波将摄像头采集的位置信息和雷达采集的径向距离、径向速度进行融合,估计检测目标在第k时刻的最优横向速度和最优纵向速度,能够有效提高速度检测的精度。
在一种实施方式中,量测模型可以包括:
Figure PCTCN2022124576-appb-000010
其中,x k-1、y k-1分别为检测目标在第k-1时刻的纵向距离和横向距离,x k、y k、r k、v r、v x、v y、α k、θ k分别为检测目标在第k时刻的纵向距离、横向距离、径向距离、径向速度、纵向速度、横向速度、位移夹角/速度夹角、方位角。
在一种实施方式中,过程模型包括:
Figure PCTCN2022124576-appb-000011
其中,x k、y k、v x,k、v y,k分别为所述检测目标在第k时刻的纵向距离、横向距离、纵向速度、横向速度,x k-1、y k-1、v x,k-1、v y,k-1分别为所述检测目标在第k-1时刻的纵向距离、横向距离、纵向速度、横向速度,ΔT为第k时刻与第k-1时刻之间的时间间隔。
图7为根据本申请实施例的多种应用场景示意图。如图7所示,本申 请实施例的速度检测方法可适用于自动驾驶车辆。例如,可适用于目标车辆710的变道切入、变道切出以及十字路口通行等场景。在这些应用场景中,目标车门710能够准确地检测出的检测目标(包括检测车辆、行人等)的横向速度和纵向速度,有助于目标车辆710在自动驾驶场景下更好地进行路径规划或障碍物避让等。
图8为根据本申请实施例的速度检测装置的结构框图。如图8所示,该速度检测装置800可以包括:
第一建立模块810,用于建立第一对应关系,第一对应关系用于表征检测目标在第k时刻的横向速度、纵向速度以及速度夹角之间的关系,其中,速度夹角为横向速度和检测目标在第k时刻的运动速度之间的夹角;
第二建立模块820,用于建立第二对应关系,第二对应关系用于表征检测目标在第k时刻的方位角、横向速度、纵向速度以及径向速度之间的关系;
确定模块830,用于基于第一对应关系和第二对应关系,确定出横向速度和纵向速度。
在一种实施方式中,第一建立模块810可以包括:
第一获取子模块,用于获取检测目标在第k-1时刻的第一位置,以及在第k时刻的第二位置;
第一确定子模块,用于确定第一位置与第二位置之间的位移夹角;
设置子模块,用于将位移夹角作为速度夹角。
在一种实施方式中,第一获取子模块可以包括:
获取单元,用于获取检测目标在第k-1时刻的第一图像;
识别单元,用于将第一图像输入目标检测模型,得到第一预选检测框;目标检测模型基于多个样本图像训练深度学习网络模型而得到;样本图像中包括检测目标的图像;
确定单元,用于从第一预选检测框中确定出第一质点信息;
转换单元,用于将第一质点信息进行坐标转换,得到第一位置。
在一种实施方式中,确定单元可以用于:
将第一预选检测框的中心作为第一质点信息;或者,
将第一预选检测框进行语义分割,得到具有检测目标的语义分割图像,并从语义分割图像中确定出第一质点信息。
在一种实施方式中,第二建立模块820可以包括:
第二获取子模块,用于获取检测目标在在第k时刻的第二位置和径向距离;
第二确定子模块,用于根据第二位置以及径向距离,确定检测目标的方位角。
在一种实施方式中,该速度检测装置还可以包括:
量测模型建立模块,用于基于第一对应关系和第二对应关系,建立检测目标的量测模型;
过程模型建立模块,用于基于预设的匀速运动模型,建立检测目标的过程模型;
估计模块,用于基于量测模型和过程模型,采用无迹卡尔曼滤波估计检测目标在第k时刻的最优横向速度和最优纵向速度。
在一种实施方式中,量测模型可以包括:
Figure PCTCN2022124576-appb-000012
其中,x k-1、y k-1分别为检测目标在第k-1时刻的纵向距离和横向距离,x k、y k、r k,R、v r,k、v x,k、v y,k、α k、θ k分别为检测目标在第k时刻的纵向距离、横向距离、径向距离、径向速度、纵向速度、横向速度、位移夹角/速度夹角、方位角。
在一种实施方式中,过程模型包括:
Figure PCTCN2022124576-appb-000013
其中,x k、y k、v x,k、v y,k分别为所述检测目标在第k时刻的纵向距离、横向距离、纵向速度、横向速度,x k-1、y k-1、v x,k-1、v y,k-1分别为所述检测目标在第k-1时刻的纵向距离、横向距离、纵向速度、横向速度,ΔT为第k时刻与第k-1时刻之间的时间间隔。
本申请实施例各装置中的各模块的功能可以参见上述方法中的对应描述,在此不再赘述。
图9为用来实现本申请实施例的速度检测方法的电子设备的框图。如图9所示,该电子设备包括:存储器910和处理器920,存储器910内存储有可在处理器920上运行的指令。处理器920执行该指令时实现上述实施例中的速度检测方法。存储器910和处理器920的数量可以为一个或多个。该电子设备旨在表示各种形式的数字计算机,诸如,膝上型计算机、台式计算机、工作台、个人数字助理、服务器、刀片式服务器、大型计算机、和其它适合的计算机。电子设备还可以表示各种形式的移动装置,诸如,个人数字处理、蜂窝电话、智能电话、可穿戴设备和其它类似的计算装置。本文所示的部件、它们的连接和关系、以及它们的功能仅仅作为示例,并且不意在限制本文中描述的和/或者要求的本申请的实现。
该电子设备还可以包括通信接口930,用于与外界设备进行通信,进行数据交互传输。各个设备利用不同的总线互相连接,并且可以被安装在公共主板上或者根据需要以其它方式安装。处理器920可以对在电子设备内执行的指令进行处理,包括存储在存储器中或者存储器上以在外部输入/输出装置(诸如,耦合至接口的显示设备)上显示GUI的图形信息的指令。在其它实施方式中,若需要,可以将多个处理器和/或多条总线与多个存储器和多个存储器一起使用。同样,可以连接多个电子设备,各个设备提供部分必要的操作(例如,作为服务器阵列、一组刀片式服务器、或者多处理器系统)。该总线可以分为地址总线、数据总线、控制总线等。为便于表示,图9中仅用一条粗线表示,但并不表示仅有一根总线或一种类型的总线。
可选的,在具体实现上,如果存储器910、处理器920及通信接口930集成在一块芯片上,则存储器910、处理器920及通信接口930可以通过 内部接口完成相互间的通信。
应理解的是,上述处理器可以是中央处理器(Central Processing Unit,CPU),还可以是其他通用处理器、数字信号处理器(Digital Signal Processing,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现场可编程门阵列(Field Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者是任何常规的处理器等。值得说明的是,处理器可以是支持进阶精简指令集机器(Advanced RISC Machines,ARM)架构的处理器。
本申请实施例提供了一种计算机可读存储介质(如上述的存储器910),其存储有计算机指令,该程序被处理器执行时实现本申请实施例中提供的方法。
可选的,存储器910可以包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需要的应用程序;存储数据区可存储根据用于实现速度检测方法的电子设备的使用所创建的数据等。此外,存储器910可以包括高速随机存取存储器,还可以包括非瞬时存储器,例如至少一个磁盘存储器件、闪存器件、或其他非瞬时固态存储器件。在一些实施例中,存储器910可选包括相对于处理器920远程设置的存储器,这些远程存储器可以通过网络连接至用于实现速度检测方法的电子设备。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
在本说明书的描述中,参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包括于本申请的至少一个实施例或示例中。而且,描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外,在不相互矛盾的情况下,本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
此外,术语“第一”、“第二”仅用于描述目的,而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此,限定有“第一”、 “第二”的特征可以明示或隐含地包括至少一个该特征。在本申请的描述中,“多个”的含义是两个或两个以上,除非另有明确具体的限定。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为,表示包括一个或多个(两个或两个以上)用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分。并且本申请的优选实施方式的范围包括另外的实现,其中可以不按所示出或讨论的顺序,包括根据所涉及的功能按基本同时的方式或按相反的顺序,来执行功能。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤,例如,可以被认为是用于实现逻辑功能的可执行指令的定序列表,可以具体实现在任何计算机可读介质中,以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用,或结合这些指令执行系统、装置或设备而使用。
应理解的是,本申请的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中,多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。上述实施例方法的全部或部分步骤是可以通过程序来指令相关的硬件完成,该程序可以存储于一种计算机可读存储介质中,该程序在执行时,包括方法实施例的步骤之一或其组合。
此外,在本申请各个实施例中的各功能单元可以集成在一个处理模块中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个模块中。上述集成的模块既可以采用硬件的形式实现,也可以采用软件功能模块的形式实现。上述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时,也可以存储在一个计算机可读存储介质中。该存储介质可以是只读存储器,磁盘或光盘等。
以上所述,仅为本申请的具体实施方式,但本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到其各种变化或替换,这些都应涵盖在本申请的保护范围之内。因此,本申请的保护范围应以权利要求的保护范围为准。

Claims (18)

  1. 一种速度检测方法,其特征在于,包括:
    建立第一对应关系,所述第一对应关系用于表征检测目标在第k时刻的横向速度、纵向速度以及速度夹角之间的关系,其中,速度夹角为所述横向速度和所述检测目标在第k时刻的运动速度之间的夹角;
    建立第二对应关系,所述第二对应关系用于表征所述检测目标在所述第k时刻的方位角、所述横向速度、纵向速度以及径向速度之间的关系;
    基于所述第一对应关系和所述第二对应关系,确定出所述横向速度和所述纵向速度。
  2. 根据权利要求1所述的方法,其特征在于,所述建立第一对应关系,包括:
    获取所述检测目标在第k-1时刻的第一位置,以及在所述第k时刻的第二位置;
    确定所述第一位置与所述第二位置之间的位移夹角;
    将所述位移夹角作为所述速度夹角。
  3. 根据权利要求2所述的方法,其特征在于,获取所述检测目标在第k-1时刻的第一位置,包括:
    获取所述检测目标在所述第k-1时刻的第一图像;
    将所述第一图像输入目标检测模型,得到第一预选检测框;所述目标检测模型基于多个样本图像训练深度学习网络模型而得到;所述样本图像中包括所述检测目标的图像;
    从所述第一预选检测框中确定出第一质点信息;
    将所述第一质点信息进行坐标转换,得到第一位置。
  4. 根据权利要求3所述的方法,其特征在于,从所述第一预选检测框中确定出第一质点信息,包括:
    将所述第一预选检测框的中心作为所述第一质点信息;或者,
    将所述第一预选检测框进行语义分割,得到具有所述检测目标的语义分割图像,并从所述语义分割图像中确定出所述第一质点信息。
  5. 根据权利要求1所述的方法,其特征在于,所述建立第二对应关系, 包括:
    获取所述检测目标在所述第k时刻的第二位置和径向距离;
    根据所述第二位置以及所述径向距离,确定所述检测目标的所述方位角。
  6. 根据权利要求1至5任一项所述的方法,其特征在于,还包括:
    基于所述第一对应关系和所述第二对应关系,建立所述检测目标的量测模型;
    基于预设的匀速运动模型,建立所述检测目标的过程模型;
    基于所述量测模型和所述过程模型,采用无迹卡尔曼滤波估计所述检测目标在所述第k时刻的最优横向速度和最优纵向速度。
  7. 根据权利要求6所述的方法,其特征在于,所述量测模型包括:
    Figure PCTCN2022124576-appb-100001
    其中,x k-1、y k-1分别为所述检测目标在第k-1时刻的纵向距离和横向距离,x k、y k、r k,R、v r,k、v x,k、v y,k、α k、θ k分别为所述检测目标在第k时刻的纵向距离、横向距离、径向距离、径向速度、纵向速度、横向速度、位移夹角/速度夹角、方位角。
  8. 根据权利要求6所述的方法,其特征在于,所述过程模型包括:
    Figure PCTCN2022124576-appb-100002
    其中,x k、y k、v x,k、v y,k分别为所述检测目标在第k时刻的纵向距离、横向距离、纵向速度、横向速度,x k-1、y k-1、v x,k-1、v y,k-1分别为所述检测目标在第k-1时刻的纵向距离、横向距离、纵向速度、横向速度,ΔT为第k时刻与第k-1时刻之间的时间间隔。
  9. 一种速度检测装置,其特征在于,包括:
    第一建立模块,用于建立第一对应关系,所述第一对应关系用于表征 检测目标在第k时刻的横向速度、纵向速度以及速度夹角之间的关系,其中,速度夹角为所述横向速度和所述检测目标在第k时刻的运动速度之间的夹角;
    第二建立模块,用于建立第二对应关系,所述第二对应关系用于表征所述检测目标在所述第k时刻的方位角、所述横向速度、纵向速度以及径向速度之间的关系;
    确定模块,用于基于所述第一对应关系和所述第二对应关系,确定出所述横向速度和所述纵向速度。
  10. 根据权利要求9所述的装置,其特征在于,所述第一建立模块包括:
    第一获取子模块,用于获取所述检测目标在第k-1时刻的第一位置,以及在所述第k时刻的第二位置;
    第一确定子模块,用于确定所述第一位置与所述第二位置之间的位移夹角;
    设置子模块,用于将所述位移夹角作为所述速度夹角。
  11. 根据权利要求10所述的装置,其特征在于,所述第一获取子模块包括:
    获取单元,用于获取所述检测目标在所述第k-1时刻的第一图像;
    识别单元,用于将所述第一图像输入目标检测模型,得到第一预选检测框;所述目标检测模型基于多个样本图像训练深度学习网络模型而得到;所述样本图像中包括所述检测目标的图像;
    确定单元,用于从所述第一预选检测框中确定出第一质点信息;
    转换单元,用于将所述第一质点信息进行坐标转换,得到第一位置。
  12. 根据权利要求11所述的装置,其特征在于,所述确定单元用于:
    将所述第一预选检测框的中心作为所述第一质点信息;或者,
    将所述第一预选检测框进行语义分割,得到具有所述检测目标的语义分割图像,并从所述语义分割图像中确定出所述第一质点信息。
  13. 根据权利要求9所述的装置,其特征在于,所述第二建立模块包括:
    第二获取子模块,用于获取所述检测目标在在所述第k时刻的第二位置和径向距离;
    第二确定子模块,用于根据所述第二位置以及所述径向距离,确定所述检测目标的所述方位角。
  14. 根据权利要求9至13任一项所述的装置,其特征在于,还包括:
    量测模型建立模块,用于基于所述第一对应关系和所述第二对应关系,建立所述检测目标的量测模型;
    过程模型建立模块,用于基于预设的匀速运动模型,建立所述检测目标的过程模型;
    估计模块,用于基于所述量测模型和所述过程模型,采用无迹卡尔曼滤波估计所述检测目标在所述第k时刻的最优横向速度和最优纵向速度。
  15. 根据权利要求14所述的装置,其特征在于,所述量测模型包括:
    Figure PCTCN2022124576-appb-100003
    其中,x k-1、y k-1分别为所述检测目标在第k-1时刻的纵向距离和横向距离,x k、y k、r k,R、v r,k、v x,k、v y,k、α k、θ k分别为所述检测目标在第k时刻的纵向距离、横向距离、径向距离、径向速度、纵向速度、横向速度、位移夹角/速度夹角、方位角。
  16. 根据权利要求14所述的装置,其特征在于,所述过程模型包括:
    Figure PCTCN2022124576-appb-100004
    其中,x k、y k、v x,k、v y,k分别为所述检测目标在第k时刻的纵向距离、横向距离、纵向速度、横向速度,x k-1、y k-1、v x,k-1、v y,k-1分别为所述检测目标在第k-1时刻的纵向距离、横向距离、纵向速度、横向速度,ΔT为第k时刻与第k-1时刻之间的时间间隔。
  17. 一种电子设备,其特征在于,包括:
    至少一个处理器;以及
    与所述至少一个处理器通信连接的存储器;其中,
    所述存储器存储有可被所述至少一个处理器执行的指令,所述指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行权利要求1-8中任一项所述的方法。
  18. 一种计算机可读存储介质,所述计算机可读存储介质内存储有计算机指令,所述计算机指令被处理器执行时实现如权利要求1-8中任一项所述的方法。
PCT/CN2022/124576 2021-10-11 2022-10-11 速度检测方法、装置、设备及可读存储介质 WO2023061355A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111180088.6A CN115963485A (zh) 2021-10-11 2021-10-11 速度检测方法、装置、设备及可读存储介质
CN202111180088.6 2021-10-11

Publications (1)

Publication Number Publication Date
WO2023061355A1 true WO2023061355A1 (zh) 2023-04-20

Family

ID=85888244

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/124576 WO2023061355A1 (zh) 2021-10-11 2022-10-11 速度检测方法、装置、设备及可读存储介质

Country Status (2)

Country Link
CN (1) CN115963485A (zh)
WO (1) WO2023061355A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103901425A (zh) * 2014-04-23 2014-07-02 武汉大学 一种测量运动物体横向速度的雷达及方法
CN106560728A (zh) * 2015-10-06 2017-04-12 通用汽车环球科技运作有限责任公司 用于目标速度估算的雷达视觉融合
CN110531357A (zh) * 2018-05-23 2019-12-03 Aptiv技术有限公司 估计移动目标在水平面中速度大小的方法和雷达检测系统
CN111366928A (zh) * 2020-02-21 2020-07-03 北京小马慧行科技有限公司 车辆速度的确定方法及装置、存储介质、处理器
CN113176561A (zh) * 2020-01-27 2021-07-27 福特全球技术公司 目标速度检测

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103901425A (zh) * 2014-04-23 2014-07-02 武汉大学 一种测量运动物体横向速度的雷达及方法
CN106560728A (zh) * 2015-10-06 2017-04-12 通用汽车环球科技运作有限责任公司 用于目标速度估算的雷达视觉融合
CN110531357A (zh) * 2018-05-23 2019-12-03 Aptiv技术有限公司 估计移动目标在水平面中速度大小的方法和雷达检测系统
CN113176561A (zh) * 2020-01-27 2021-07-27 福特全球技术公司 目标速度检测
CN111366928A (zh) * 2020-02-21 2020-07-03 北京小马慧行科技有限公司 车辆速度的确定方法及装置、存储介质、处理器

Also Published As

Publication number Publication date
CN115963485A (zh) 2023-04-14

Similar Documents

Publication Publication Date Title
US11002840B2 (en) Multi-sensor calibration method, multi-sensor calibration device, computer device, medium and vehicle
US20190205665A1 (en) Method, apparatus, and device for determining lane line on road
EP3910543A2 (en) Method for training object detection model, object detection method and related apparatus
KR102543952B1 (ko) 차선의 확정방법, 포지셔닝 정밀도의 평가방법, 장치, 기기 및 컴퓨터 프로그램
EP3910533B1 (en) Method, apparatus, electronic device, and storage medium for monitoring an image acquisition device
CN110927708A (zh) 智能路侧单元的标定方法、装置及设备
CN111126268B (zh) 关键点检测模型训练方法、装置、电子设备及存储介质
CN109959937B (zh) 长廊环境基于激光雷达的定位方法、存储介质及电子设备
WO2023155387A1 (zh) 多传感器目标检测方法、装置、电子设备以及存储介质
TW201934967A (zh) 用於校準光強度的系統和方法
US11953602B2 (en) Detecting three-dimensional structure models at runtime in vehicles
WO2022141116A1 (zh) 三维点云分割方法和装置、可移动平台
WO2021185104A1 (zh) 一种车道线信息确定方法及装置
WO2022062355A1 (zh) 一种融合定位方法及装置
JP2021135286A (ja) 座標変換方法、装置及びデータ処理装置
CN112689842B (zh) 一种目标检测方法以及装置
WO2023061355A1 (zh) 速度检测方法、装置、设备及可读存储介质
CN114662600B (zh) 一种车道线的检测方法、装置和存储介质
WO2023179032A1 (zh) 图像处理方法及装置、电子设备、存储介质、计算机程序、计算机程序产品
WO2023000540A1 (zh) 一种高炉料面形状测量方法、终端设备及存储介质
US20160314592A1 (en) Method and apparatus for registration of interior and exterior three dimensional scans based on a semantic feature of a structure
WO2021098320A1 (zh) 一种道路约束确定方法及装置
CN114964204A (zh) 地图构建方法、地图使用方法、装置、设备和存储介质
CN110399892B (zh) 环境特征提取方法和装置
CN112241675A (zh) 物体检测模型训练方法及装置

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: 22880292

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE