WO2021174507A1 - 参数标定方法、装置、系统和存储介质 - Google Patents

参数标定方法、装置、系统和存储介质 Download PDF

Info

Publication number
WO2021174507A1
WO2021174507A1 PCT/CN2020/078068 CN2020078068W WO2021174507A1 WO 2021174507 A1 WO2021174507 A1 WO 2021174507A1 CN 2020078068 W CN2020078068 W CN 2020078068W WO 2021174507 A1 WO2021174507 A1 WO 2021174507A1
Authority
WO
WIPO (PCT)
Prior art keywords
radar
measurement unit
inertial measurement
gravity vector
coordinate system
Prior art date
Application number
PCT/CN2020/078068
Other languages
English (en)
French (fr)
Inventor
朱晏辰
李延召
Original Assignee
深圳市大疆创新科技有限公司
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 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to CN202080004275.6A priority Critical patent/CN113767264A/zh
Priority to PCT/CN2020/078068 priority patent/WO2021174507A1/zh
Publication of WO2021174507A1 publication Critical patent/WO2021174507A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C25/00Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass
    • 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
    • G01S7/00Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/48Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
    • G01S7/497Means for monitoring or calibrating

Definitions

  • the present disclosure relates to the field of computer technology, and in particular to a parameter calibration method, a parameter calibration device, a parameter calibration system and a readable storage medium.
  • the parameter calibration between the existing radar and the inertial measurement unit usually requires the integration calculation of the data measured by the IMU, but due to the drift problem of the IMU device measurement, the information obtained by the integration is very fast It will diverge, which will have a greater impact on the calibration accuracy.
  • the present disclosure provides a parameter calibration method for radar and inertial measurement unit, wherein the above-mentioned radar and the above-mentioned inertial measurement unit are fixedly connected, and the above-mentioned method includes: obtaining N sets of gravity vector data, wherein the above-mentioned N sets of gravity vector data Each set of gravity vector data is obtained after changing the attitude of the aforementioned radar and the aforementioned inertial measurement unit, and the aforementioned radar and the aforementioned inertial measurement unit are in a stationary state. Each aforementioned set of gravity vector data includes the first in the radar coordinate system. A gravity vector and a second gravity vector in the inertial measurement unit coordinate system, where N is an integer greater than 2; and the rotation transformation between the radar coordinate system and the inertial measurement unit coordinate system is determined according to the N sets of gravity vector data parameter.
  • the present disclosure also provides a parameter calibration device, including: a processor; a memory, used to store one or more programs, wherein, when the one or more programs are executed by the processor, the processor realizes the above The described parameter calibration method.
  • the present disclosure also provides a parameter calibration system, which includes a radar, an inertial measurement unit and a parameter calibration device.
  • the above-mentioned inertial measurement unit is fixedly connected with the above-mentioned radar;
  • the above-mentioned parameter calibration device includes: a processor and a memory, and the memory is used to store one or more programs, wherein when the above-mentioned one or more programs are executed by the above-mentioned processor, the above-mentioned processing
  • the device implements the above-mentioned parameter calibration method.
  • the present disclosure also provides a readable storage medium with executable instructions stored thereon, and when the instructions are executed by a processor, the processor executes the above-mentioned parameter calibration method.
  • the radar and the inertial measurement unit are kept stationary, and the radar and the inertial measurement unit in their respective coordinate systems are obtained by obtaining the gravity vector of the radar and the inertial measurement unit to determine the radar according to the gravity vector data.
  • Rotation transformation parameter between the coordinate system and the inertial measurement unit coordinate system is obtained by obtaining the gravity vector of the radar and the inertial measurement unit to determine the radar according to the gravity vector data.
  • the gravity vectors of the radar and the inertial measurement unit in their respective coordinate systems are obtained in a stationary state, and the rotation transformation parameters are determined through the gravity vector data, the point cloud registration calculation and the integration of the IMU data are not involved in the parameter calibration process It avoids the cumbersome point cloud matching process and the IMU integral calculation process, solves the technical problem of using related technologies to calibrate parameters, and the calibration accuracy is low, resulting in low accuracy of the sensor measurement results. The technical effect of stable and high calibration accuracy.
  • Fig. 1 schematically shows an application scenario of a parameter calibration method and device for radar and inertial measurement unit according to an embodiment of the present disclosure.
  • Fig. 2 schematically shows a flowchart of a parameter calibration method for radar and inertial measurement unit according to an embodiment of the present disclosure.
  • Fig. 3 schematically shows a flowchart of obtaining N sets of gravity vector data according to an embodiment of the present disclosure.
  • FIG. 4 schematically shows a schematic diagram of collecting a point cloud data set of a first calibration reference object and a second calibration reference object by lidar according to an embodiment of the present disclosure.
  • Fig. 5 schematically shows a flow chart of determining the first gravity vector in the radar coordinate system according to K groups of point cloud data sets according to an embodiment of the present disclosure.
  • Fig. 6 schematically shows a flow chart of obtaining a second gravity vector measured by an inertial measurement unit during data collection by a radar according to an embodiment of the present disclosure.
  • Fig. 7 schematically shows a flowchart of a parameter calibration method according to another embodiment of the present disclosure.
  • Fig. 8 schematically shows a block diagram of a parameter calibration device according to an embodiment of the present disclosure.
  • Fig. 9 schematically shows a block diagram of a parameter calibration system according to an embodiment of the present disclosure.
  • the point cloud data set of the object after performing imaging processing based on these data, can obtain a three-dimensional image.
  • the point cloud data set can be a point set composed of points containing three-dimensional coordinates, which can be used to characterize the shape of the target.
  • the types of radar may include, for example, LiDAR.
  • IMU Inertial measurement unit
  • IMU Inertial measurement unit
  • an inertial sensing unit can be equipped with a three-axis gyroscope and a three-directional accelerometer to measure the angular velocity and acceleration of the object in the three-dimensional space, and to calculate the posture of the object.
  • the measurement data of the two are obtained in their respective body coordinate systems, and the sensor external parameters can refer to the three-dimensional relationship between the respective body coordinate systems of the radar and the inertial measurement unit. Spatial transformation relationship.
  • the calibration of parameters between the radar and the inertial measurement unit usually requires the use of radar point cloud matching calculation, but the point cloud registration method is more likely to affect the final calibration accuracy.
  • the information obtained by the integration will quickly diverge, which will have a greater impact on the calibration accuracy.
  • the embodiments of the present disclosure provide a parameter calibration method for radar and inertial measurement unit, which realizes that the point cloud registration calculation and the integration of IMU data are not involved in the parameter calibration process of radar and inertial measurement unit. , To avoid the tedious point cloud matching process and the IMU integral calculation process.
  • FIG. 1 schematically shows an application scenario of a parameter calibration method and device for radar and inertial measurement unit according to an embodiment of the present disclosure. It should be noted that FIG. 1 is only an example of a scenario where the embodiment of the present disclosure can be applied to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiment of the present disclosure cannot be used for other devices. , System, environment or scene.
  • the radar 101 and the inertial measurement unit 102 can be fixedly connected, and the measurement data of the two are obtained in their respective body coordinate systems.
  • the inertial measurement unit 102 is also in a stationary state.
  • the inertial measurement unit 102 is also in a moving state.
  • the fixed connection manner of the radar 101 and the inertial measurement unit 102 is not limited.
  • the radar 101 and the inertial measurement unit 102 may be rigidly connected through a connecting object, or the inertial measurement unit 102 may also be integrated in the radar 101.
  • the radar 101 can obtain point cloud data in the field of view while keeping the radar 101 and the inertial measurement unit 102 stationary.
  • the radar 101 can obtain point cloud data of the calibration object 103 and the calibration object 104 in the field of view.
  • the calibration object 103 and the calibration object 104 may be some prior information, such as plumb board, vertical wall, plumb line, etc.
  • the number of the calibration object 103 and the calibration object 104 may be multiple.
  • the expression of the gravity vector in the radar coordinate system can be effectively measured.
  • the accelerometer of the inertial measurement unit 102 can synchronously obtain the expression of the gravity vector in the coordinate system of the inertial measurement unit.
  • the point cloud registration calculation and IMU integration problems are not involved in the parameter calibration process, and the gravity vector is obtained with certain prior information, and a more accurate rotation transformation can be obtained based on the gravity vector in different sensor coordinate systems Parameter, the calibration process is relatively simple.
  • Fig. 2 schematically shows a flowchart of a parameter calibration method for radar and inertial measurement unit according to an embodiment of the present disclosure.
  • the parameter calibration method for radar and inertial measurement unit includes operations S210 to S220.
  • N sets of gravity vector data are obtained, where each set of gravity vector data in the N sets of gravity vector data is obtained after changing the attitude of the radar and the inertial measurement unit, and making the radar and the inertial measurement unit in a stationary state.
  • each set of gravity vector data includes the first gravity vector in the radar coordinate system and the second gravity vector in the inertial measurement unit coordinate system, and N is an integer greater than 2.
  • the radar and the inertial measurement unit can be in a static state in the current attitude, and then the calibration object can be scanned by the radar, and the calculation is based on the data obtained from the scan Get the first gravity vector in the radar coordinate system.
  • the inertial measurement unit can be used to measure the second gravity vector in the inertial measurement unit coordinate system.
  • the N sets of gravity vector data are obtained in the stationary state of the radar and the inertial measurement unit (ie, IMU), the drift problem of the gravity vector measured by the inertial measurement unit during the movement can be avoided.
  • the number of times of changing the posture can be determined according to the calibration accuracy of the finally determined rotation transformation parameter.
  • the number of gestures can be 3 times, 5 times, and so on.
  • each time the attitude of the radar and the inertial measurement unit is changed one or more sets of gravity vector data can be obtained.
  • the type of radar is not limited.
  • it may be laser radar, millimeter wave radar, ultrasonic radar, and so on.
  • the inertial measurement unit may include an accelerometer.
  • the accelerometer may measure the data of the gravity vector in the coordinate system of the inertial measurement unit.
  • the size of N can be determined according to the calibration accuracy of the finally determined rotation transformation parameter. Specifically, the size of N may be 3, 10, 20, 30 or 50 and other integers.
  • a rotation transformation parameter between the radar coordinate system and the inertial measurement unit coordinate system is determined according to the N sets of gravity vector data.
  • the point cloud data information collected by radar scanning and the gravitational acceleration information collected by the IMU in a stationary state can be used to jointly calculate the rotation transformation parameters of the two sensors by setting a certain a priori calibration object.
  • the translation transformation parameter between the radar coordinate system and the inertial measurement unit coordinate system can also be determined, and the rotation transformation parameter and the translation transformation parameter can be determined as the coordinate transformation between the radar coordinate system and the inertial measurement unit coordinate system. parameter. This can solve the problem of inaccurate measurement data due to inconsistent coordinate systems during multi-sensor fusion between the radar and the inertial measurement unit.
  • the radar and the inertial measurement unit may be fixedly connected, and the application scenarios of the radar and the inertial measurement unit are not limited.
  • the radar and the inertial measurement unit may be fixedly connected and then set on the drone.
  • the radar and the inertial measurement unit may be fixedly connected and then installed on the unmanned vehicle.
  • the radar and the inertial measurement unit are kept stationary, and the radar and the inertial measurement unit in their respective coordinate systems are obtained by obtaining the gravity vector of the radar and the inertial measurement unit to determine the radar according to the gravity vector data.
  • Rotation transformation parameter between the coordinate system and the inertial measurement unit coordinate system is obtained by obtaining the gravity vector of the radar and the inertial measurement unit to determine the radar according to the gravity vector data.
  • the gravity vectors of the radar and the inertial measurement unit in their respective coordinate systems are obtained in a stationary state, and the rotation transformation parameters are determined through the gravity vector data, the point cloud registration calculation and the integration of the IMU data are not involved in the parameter calibration process It avoids the cumbersome point cloud matching process and the IMU integral calculation process, solves the technical problem of using related technologies to calibrate parameters, and the calibration accuracy is low, resulting in low accuracy of the sensor measurement results. The technical effect of stable and high calibration accuracy.
  • FIG. 2 The method shown in FIG. 2 will be further described below with reference to FIGS. 3 to 5 in combination with specific embodiments.
  • Fig. 3 schematically shows a flowchart of obtaining N sets of gravity vector data according to an embodiment of the present disclosure.
  • obtaining N sets of gravity vector data includes operations S310 to S340.
  • the method includes operation S310 to operation S340.
  • K groups of point cloud data sets are acquired, where the K groups of point cloud data sets are obtained by the radar for each calibration reference object of the K calibration reference objects respectively, where K is greater than 1. Integer.
  • K calibration reference objects can be placed non-parallel in advance, and the surface normal vector of each calibration reference object is perpendicular to the direction of gravity.
  • the K calibration reference objects may be vertical calibration objects.
  • the types of calibration reference objects include at least plumb board and/or building wall and so on.
  • the K calibration reference objects may belong to the same type of reference object or different types of reference objects.
  • the K calibration reference objects may all be plumb boards, or part of the K calibration reference objects may be plumb boards and part of the building wall.
  • the size of K is not limited. For example, it may be 2, 3, 5, or other integers greater than 1.
  • the K calibration reference objects include a first calibration reference object and a second calibration reference object.
  • the radar can collect data on the first calibration reference object for a predetermined period of time to obtain a set of point cloud data sets for the first calibration reference object. Then, the radar can perform data collection on the second calibration reference object for a predetermined period of time to obtain a set of point cloud data sets for the second calibration reference object.
  • the predetermined duration is not limited, for example, it may be 2 seconds, 3 seconds, or 5 seconds, etc., which may be determined according to the type of radar and the collection effect.
  • the lidar may be a non-repetitive scanning lidar.
  • the non-repetitive scanning lidar it can have better calibration accuracy due to its measurement feature that it can obtain complete coverage of the field of view data under static conditions.
  • the scanning path of the lidar may be curved.
  • the scanning density of the field of view of the lidar gradually increases within a certain period of time.
  • a point cloud coverage of nearly 100% in the field of view of the lidar can be reached in a short period of time. Therefore, by calibrating The way that the reference object is placed in the field of view of the lidar can quickly obtain the point cloud data of the calibration reference object, which improves the calibration efficiency.
  • the first gravity vector in the radar coordinate system is determined according to the K sets of point cloud data sets.
  • the data of the gravity vector in the coordinate system of the inertial measurement unit can be obtained by the accelerometer of the inertial measurement unit, that is, the second gravity vector can be obtained by measurement.
  • the first gravity vector and the second gravity vector obtained every time the attitude of the radar and the inertial measurement unit are changed can be used as a set of gravity vector data.
  • FIG. 4 schematically shows a schematic diagram of collecting a point cloud data set of a first calibration reference object and a second calibration reference object by lidar according to an embodiment of the present disclosure.
  • the lidar 411 and the inertial measurement unit 412 are integrated on the same device 410.
  • the first calibration reference object 420 and the second calibration reference object 430 can be scanned in the field of view of the lidar 411, respectively.
  • the first calibration reference object 420 and the second calibration reference object 430 can be placed non-parallel to ensure that the surface normal vector is perpendicular to the gravity vector, and it is required to remain stationary during the data collection process.
  • the first calibration reference 420 and the second calibration reference object 430 is placed in the field of view of the lidar 411 to quickly obtain the point cloud data of the calibration reference object.
  • the process of one static data collection is shown in Figure 4, and the duration of one static data collection can be t, for example, 3 seconds.
  • X points on the first calibration reference object 420 and Y points on the second calibration reference object 430 can be acquired.
  • the inertial measurement unit 412 can obtain a accelerometer measurement values within t, and the second gravity vector can be determined based on the a accelerometer measurement values.
  • Fig. 5 schematically shows a flowchart of determining a first gravity vector in a radar coordinate system according to K groups of point cloud data sets according to an embodiment of the present disclosure.
  • determining the first gravity vector in the radar coordinate system according to the K groups of point cloud data sets includes operations S510 to S520.
  • the first gravity vector in the radar coordinate system is calculated according to the K plane normal vectors.
  • the K calibration reference objects include a first calibration reference object and a second calibration reference object.
  • the plane normal vector V1 of the first calibration reference object can be calculated by fitting the point cloud data set of the first calibration reference object.
  • the plane normal vector V2 of the second calibration reference object is calculated by fitting the point cloud data set of the second calibration reference object.
  • the expression of the first gravity vector in the radar coordinate system is calculated, which can be expressed as L v.
  • the calibration reference object of the present disclosure is not limited to the first calibration reference object and the second calibration reference object, and may also include multiple calibration reference objects such as the third calibration reference object.
  • the K calibration reference objects also including the third calibration reference object as an example.
  • the plane normal vector V3 of the third calibration reference object can be calculated by fitting the point cloud data set of the third calibration reference object. By cross-multiplying any two of the plane normal vector V1, the plane normal vector V2, and the plane normal vector V3, multiple calculation results are obtained, and the expression of the first gravity vector in the radar coordinate system is determined according to the multiple calculation results. For example, the average value of multiple calculation results may be calculated, and the average value may be determined as the first gravity vector.
  • Fig. 6 schematically shows a flow chart of obtaining a second gravity vector measured by an inertial measurement unit during data collection by a radar according to an embodiment of the present disclosure.
  • obtaining the second gravity vector measured by the inertial measurement unit during data collection by the radar includes operations S610 to S620.
  • the inertial measurement unit can measure multiple gravity measurement vectors during the radar's data collection process, which can be represented by I v i in the inertial measurement The gravity measurement vector in the unit coordinate system.
  • a second gravity vector is determined according to a plurality of gravity measurement vectors.
  • a gravity measurement vector with the smallest variance can be determined from a plurality of gravity measurement vectors as the second gravity vector.
  • the second gravity vector can be expressed as I v.
  • the radar coordinate system and inertia can be determined according to the N sets of gravity vector data Rotation transformation parameters between the measurement unit coordinate systems.
  • N sets of gravity vector data can be used to obtain the rotation transformation parameters between the radar coordinate system and the inertial measurement unit coordinate system through the least squares algorithm.
  • the solving equation can be in the following form.
  • the optimal rotation matrix q namely the radar coordinate system and the inertial measurement unit coordinate can be obtained. Rotation transformation parameters between systems.
  • Fig. 7 schematically shows a flowchart of a parameter calibration method according to another embodiment of the present disclosure.
  • the parameter calibration method includes operation S710 to operation S770.
  • a calibration reference object with a priori information is selected to build a calibration scene.
  • two lifted plumb plates can be used as calibration reference objects to ensure that the surface normal vector is perpendicular to the gravity vector, and the two calibration plates are not placed in parallel. At the same time, it is required to remain still during the data collection process.
  • the wall of the building is vertical, and the wall can also be used as a reference for calibration.
  • the lidar collects the point cloud data of the calibration reference object and the inertial measurement unit measures the acceleration of gravity.
  • operation S760 it is determined whether the number of data groups meets the number requirement. If it is satisfied, perform S770, otherwise, perform operation S720.
  • N sets of gravity vector data are used to jointly solve to obtain rotation transformation parameters.
  • the point cloud registration is not involved in the parameter calibration process
  • the problem of calculation and integration of IMU data avoids the tedious point cloud matching process and the IMU integration calculation process, and solves the technical problem of using related technologies for parameter calibration, and the calibration accuracy is low, which leads to the low accuracy of sensor measurement results.
  • the technical effect of simplifying the calibration process and obtaining stable and higher calibration accuracy is achieved.
  • Fig. 8 schematically shows a block diagram of a parameter calibration device according to an embodiment of the present disclosure.
  • the parameter calibration device 800 includes a processor 810 and a memory 820.
  • the memory 820 is configured to store one or more programs, where when the one or more programs are executed by the processor 810, the processor is caused to implement the parameter calibration method as described above.
  • the processor 810 may include, for example, a general-purpose microprocessor, an instruction set processor and/or a related chipset and/or a special-purpose microprocessor (for example, an application specific integrated circuit (ASIC)), and so on.
  • the processor 810 may also include on-board memory for caching purposes.
  • the processor 810 may be a single processing unit or multiple processing units for executing different actions of a method flow according to an embodiment of the present disclosure.
  • Fig. 9 schematically shows a block diagram of a parameter calibration system according to an embodiment of the present disclosure.
  • the parameter calibration system 900 includes a radar 910, an inertial measurement unit 920 and a parameter calibration device 930.
  • the inertial measurement unit 920 and the radar 910 are fixedly connected.
  • the parameter calibration device 930 is the same as the parameter calibration device 800 shown in FIG. 8.
  • a readable storage medium having executable instructions stored thereon, and when the instructions are executed by a processor, the processor executes the parameter calibration method as described above.
  • the readable storage medium may be included in the device/device/system described in the above embodiments; or it may exist alone without being assembled into the device/device/system.
  • the aforementioned readable storage medium carries one or more programs, and when the aforementioned one or more programs are executed, the method according to the embodiments of the present disclosure is implemented.
  • the readable storage medium may be a nonvolatile readable storage medium, for example, it may include but not limited to: portable computer disk, hard disk, random access memory (RAM), read only memory (ROM), Erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.
  • a readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device.
  • each block in the flowchart or block diagram may represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more for realizing the specified logic function.
  • Executable instructions may also occur in a different order from the order marked in the drawings. For example, two blocks shown one after another can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved.
  • each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by It is realized by a combination of dedicated hardware and computer instructions.

Landscapes

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

Abstract

一种参数标定方法、一种参数标定装置、一种参数标定系统和一种可读存储介质。该方法包括:用于雷达和惯性测量单元的参数标定方法,其中,雷达和惯性测量单元固定连接,方法包括:获取N组重力向量数据,其中,N组重力向量数据中的每组重力向量数据是在改变雷达和惯性测量单元的姿态之后,并使雷达和惯性测量单元处于静止状态下得到的,每组重力向量数据包括在雷达坐标系下的第一重力向量和在惯性测量单元坐标系下的第二重力向量,N为大于2的整数;以及根据N组重力向量数据确定雷达坐标系和惯性测量单元坐标系之间的旋转变换参数。

Description

参数标定方法、装置、系统和存储介质 技术领域
本公开涉及计算机技术领域,尤其涉及一种参数标定方法、一种参数标定装置、一种参数标定系统和一种可读存储介质。
背景技术
随着传感器技术的快速发展,越来越多的传感器被应用于例如无人机、无人车等自动化控制技术中。在多传感器融合应用方案中,雷达与惯性测量单元的联合使用已成为多传感器融合中不可缺少的方案之一。而实现雷达与惯性测量单元之间的参数标定,对多传感器融合的应用至关重要,影响到传感器测量结果的准确性。
目前,现有的雷达和惯性测量单元(Inertial measurement unit,简称IMU)之间的参数标定通常需要对IMU测量的数据进行积分计算,但由于IMU器件测量存在漂移问题,其积分得到的信息很快会发散,给标定精度带来较大影响。
公开内容
本公开提供了一种用于雷达和惯性测量单元的参数标定方法,其中,上述雷达和上述惯性测量单元固定连接,上述方法包括:获取N组重力向量数据,其中,上述N组重力向量数据中的每组重力向量数据是在改变上述雷达和上述惯性测量单元的姿态之后,并使上述雷达和上述惯性测量单元处于静止状态下得到的,上述每组重力向量数据包括在雷达坐标系下的第一重力向量和在惯性测量单元坐标系下的第二重力向量,上述N为大于2的整数;以及根据上述N组重力向量数据确定上述雷达坐标系和上述惯性测量单元坐标系之间的旋转变换参数。
本公开还提供了一种参数标定装置,包括:处理器;存储器,用于存储一个或多个程序,其中,当上述一个或多个程序被上述处理器执行时,使得上述处理器实现如上所述的参数标定方法。
本公开还提供了一种参数标定系统,包括:雷达、惯性测量单元和参数标定装置。上述惯性测量单元与上述雷达固定连接;上述参数标定装置包括:处理器和存储器,存储器用于存储一个或多个程序,其中,当上述一个或多个程序被上述处理器执行时,使得上述处理器实现如上所述的参数标定方法。
本公开还提供了一种可读存储介质,其上存储有可执行指令,该指令被处理器执行时使处理器执行如上所述的参数标定方法。
通过本公开的实施例,在改变雷达和惯性测量单元的姿态之后,将雷达和惯性测量单元保持静止状态,通过获取雷达和惯性测量单元在各自坐标系下的重力向量,根据重力向量数据确定雷达坐标系和惯性测量单元坐标系之间的旋转变换参数。由于是在静止状态下获得雷达和惯性测量单元在各自坐标系下的重力向量,并通过重力向量数据确定旋转变换参数,使得在参数标定过程中不涉及点云配准计算以及IMU数据的积分问题,避免了繁琐的点云匹配过程以及IMU积分计算过程,解决了采用相关技术进行参数标定时,标定精度较低,导致传感器测量结果的准确性较低的技术问题,达到了简化标定流程,获得稳定且较高的标定精度的技术效果。
附图说明
附图是用来提供对本公开的进一步理解,并且构成说明书的一部分,与下面的具体实施方式一起用于解释本公开,但并不构成对本公开的限制。在附图中:
图1示意性示出了根据本公开实施例的用于雷达和惯性测量单元的参数标定方法及装置的应用场景。
图2示意性示出了根据本公开实施例的用于雷达和惯性测量单元的参数标定方法的流程图。
图3示意性示出了根据本公开实施例的获取N组重力向量数据的流程图。
图4示意性示出了根据本公开实施例的通过激光雷达采集第一标定参照物和第二标定参照物的点云数据集的示意图。
图5示意性示出了根据本公开实施例的根据K组点云数据集确定在雷 达坐标系下的第一重力向量的流程图。
图6示意性示出了根据本公开实施例的获取惯性测量单元在雷达进行数据采集的过程中测量得到的第二重力向量的流程图。
图7示意性示出了根据本公开另一实施例的参数标定方法的流程图。
图8示意性示出了根据本公开实施例的参数标定装置的框图。
图9示意性示出了根据本公开实施例的参数标定系统的框图。
具体实施方式
下面将结合实施例和实施例中的附图,对本公开技术方案进行清楚、完整的描述。显然,所描述的实施例仅仅是本公开一部分实施例,而不是全部的实施例。基于本公开中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本公开保护的范围。在下面的详细描述中,为便于解释,阐述了许多具体的细节以提供对本公开实施例的全面理解。然而,明显地,一个或多个实施例在没有这些具体细节的情况下也可以被实施。此外,在以下说明中,省略了对公知结构和技术的描述,以避免不必要地混淆本公开的概念。
附图中示出了一些方框图和/或流程图。应理解,方框图和/或流程图中的一些方框或其组合可以由计算机程序指令来实现。这些计算机程序指令可以提供给通用计算机、专用计算机或其他可编程数据处理装置的处理器,从而这些指令在由该处理器执行时可以创建用于实现这些方框图和/或流程图中所说明的功能/操作的装置。本公开的技术可以硬件和/或软件(包括固件、微代码等)的形式来实现。另外,本公开的技术可以采取存储有指令的计算机可读存储介质上的计算机程序产品的形式,该计算机程序产品可供指令执行系统使用或者结合指令执行系统使用。
在本文中,需要理解的是,所涉及的术语可以是用于实现本公开一部分的技术手段或者其它总结性技术术语,例如,术语可以包括:
雷达(Radar):通过发射探测信号扫描目标物,如通过发射脉冲激光不断的扫描目标物,接收目标物反射回来的光信号,可以得到目标物上部分或全部目标点的数据,即可以得到目标物的点云数据集,根据这些数据进行成像处理后,可以得到三维立体图像。其中,点云数据集可以是由包 含三维坐标的点所组成的点集,可以用来表征目标物的形状。雷达的种类例如可以包括激光雷达(LiDAR)。
惯性测量单元(inertial measurement unit,简称IMU):测量物体三轴姿态角(或角速率)以及加速度的装置。一般地,一个惯性传感单元内可以装有三轴的陀螺仪和三个方向的加速度计,来测量物体在三维空间中的角速度和加速度,并以此解算出物体的姿态。
在相关技术中,对于多传感器融合应用,需要解决传感器外参标定问题。具体地,对于雷达和惯性测量单元的融合应用,二者的测量数据是分别在各自的体坐标系下得到的,传感器外参可以是指雷达和惯性测量单元各自的体坐标系之间的三维空间变换关系。而在目前,雷达和惯性测量单元之间的参数标定通常需要使用雷达点云匹配计算,但点云的配准方式较容易影响最终标定精度。同时,还需要对IMU测量的数据进行积分计算,但由于IMU器件测量存在漂移问题,其积分得到的信息很快会发散,给标定精度带来较大影响。
基于此,本公开的实施例提供了一种用于雷达和惯性测量单元的参数标定方法,实现了在雷达和惯性测量单元的参数标定过程中不涉及点云配准计算以及IMU数据的积分问题,避免了繁琐的点云匹配过程以及IMU积分计算过程。
下面参考图1阐述本公开实施例的用于雷达和惯性测量单元的参数标定方法及装置的应用场景。
图1示意性示出了根据本公开实施例的用于雷达和惯性测量单元的参数标定方法及装置的应用场景。需要注意的是,图1所示仅为可以应用本公开实施例的场景的示例,以帮助本领域技术人员理解本公开的技术内容,但并不意味着本公开实施例不可以用于其他设备、系统、环境或场景。
如图1所示,雷达101和惯性测量单元102可以固定连接,二者的测量数据是分别在各自的体坐标系下得到的。在雷达101保持静止的情况下,惯性测量单元102也处于静止状态。在雷达101运动的情况下,惯性测量单元102也处于运动状态。雷达101和惯性测量单元102的固定连接方式不做限定,例如,可以是通过连接物将雷达101和惯性测量单元102进行刚性连接,或者,也可以将惯性测量单元102集成在雷达101内部。
根据本公开的实施例,在保持雷达101和惯性测量单元102静止的情况下,雷达101可以获得视场内的点云数据。例如,雷达101可以获得视场内标定物103和标定物104的点云数据。其中,标定物103和标定物104可以是一些先验信息,如铅锤板,垂直的墙面、铅垂线等,标定物103标定物103和标定物104的数量可以是多个。根据视场内标定物103和标定物104的点云数据,可有效测量重力向量在雷达坐标系下的表达。与此同时,在固定连接的惯性测量单元102中,惯性测量单元102的加速度计可同步获得重力向量在惯性测量单元坐标系下的表达。
通过改变雷达101和惯性测量单元102的姿态,可以获得重力向量分别在雷达坐标系和惯性测量单元坐标系下的多组数据,最终通过多组数据联合求解两个坐标系之间的旋转变换,可以得到雷达坐标系和惯性测量单元坐标系之间的旋转变换参数。
通过本公开的实施例,在参数标定过程中不涉及点云配准计算以及IMU积分问题,借助一定的先验信息获得重力向量,基于不同传感器坐标系下的重力向量能够获得较为精确的旋转变换参数,标定过程较为简单。
图2示意性示出了根据本公开实施例的用于雷达和惯性测量单元的参数标定方法的流程图。
如图2所示,该用于雷达和惯性测量单元的参数标定方法包括操作S210~S220。
在操作S210,获取N组重力向量数据,其中,该N组重力向量数据中的每组重力向量数据是在改变雷达和惯性测量单元的姿态之后,并使雷达和惯性测量单元处于静止状态下得到的,每组重力向量数据包括在雷达坐标系下的第一重力向量和在惯性测量单元坐标系下的第二重力向量,N为大于2的整数。
根据本公开的实施例,在每一次改变雷达和惯性测量单元的姿态之后,可以将雷达和惯性测量单元在当前姿态下处于静止状态,然后利用雷达对标定物进行扫描,基于扫描得到的数据计算得到在雷达坐标系下的第一重力向量。与此同时,可以利用惯性测量单元测量得到在惯性测量单元坐标系下的第二重力向量。
根据本公开的实施例,由于是在雷达和惯性测量单元(即IMU)静止状态下获得的N组重力向量数据,可以避免惯性测量单元在移动过程中测量重力向量存在的漂移问题。
根据本公开的实施例,改变姿态的次数可以根据最终确定的旋转变换参数的标定精度进行确定。例如,姿态的次数可以是3次,5次等。根据本公开的实施例,每改变一次雷达和惯性测量单元的姿态,可以获得一组或多组重力向量数据。
根据本公开的实施例,雷达的种类不做限定,例如,可以是激光雷达,毫米波雷达,超声波雷达等等。根据本公开的实施例,惯性测量单元可以包括加速度计,在雷达和惯性测量单元处于静止状态下,加速度计可以测量得到重力向量在惯性测量单元坐标系下的数据。根据本公开的实施例,N的大小可以根据最终确定的旋转变换参数的标定精度进行确定。具体地,N的大小可以是3,10,20,30或者50等其他整数。
在操作S220,根据N组重力向量数据确定雷达坐标系和惯性测量单元坐标系之间的旋转变换参数。
根据本公开的实施例,可以利用静止状态下雷达扫描采集得到的点云数据信息以及IMU采集的重力加速度信息,通过设置一定的先验标定物,联合解算两个传感器的旋转变换参数。
根据本公开的实施例,还可以确定雷达坐标系和惯性测量单元坐标系之间的平移变换参数,将旋转变换参数和平移变换参数确定为雷达坐标系和惯性测量单元坐标系之间的坐标变换参数。从而可以解决雷达和惯性测量单元之间进行多传感器融合时由于坐标系不统一而导致测量数据不准确的问题。
根据本公开的实施例,雷达和惯性测量单元可以固定连接,雷达和惯性测量单元设置的应用场景不做限定。例如,在一可选的实施例中,可以将雷达和惯性测量单元固定连接后设置在无人机上。在另一可选的实施例中,可以将雷达和惯性测量单元固定连接后设置在无人车上。
通过本公开的实施例,在改变雷达和惯性测量单元的姿态之后,将雷达和惯性测量单元保持静止状态,通过获取雷达和惯性测量单元在各自坐标系下的重力向量,根据重力向量数据确定雷达坐标系和惯性测量单元坐 标系之间的旋转变换参数。由于是在静止状态下获得雷达和惯性测量单元在各自坐标系下的重力向量,并通过重力向量数据确定旋转变换参数,使得在参数标定过程中不涉及点云配准计算以及IMU数据的积分问题,避免了繁琐的点云匹配过程以及IMU积分计算过程,解决了采用相关技术进行参数标定时,标定精度较低,导致传感器测量结果的准确性较低的技术问题,达到了简化标定流程,获得稳定且较高的标定精度的技术效果。
下面参考图3~图5,结合具体实施例对图2所示的方法做进一步说明。
图3示意性示出了根据本公开实施例的获取N组重力向量数据的流程图。
如图3所示,获取N组重力向量数据包括操作S310~操作S340。
该方法包括操作S310~操作S340。
在操作S310,在每一次改变雷达和惯性测量单元的姿态之后,使雷达和惯性测量单元处于静止状态。
在操作S320,获取K组点云数据集,其中,K组点云数据集是由雷达针对K个标定参照物中的每个标定参照物分别进行数据采集得到的,其中,K为大于1的整数。
根据本公开的实施例,可以预先将K个标定参照物非平行放置,并使得将每个标定参照物的面法向量与重力方向垂直。K个标定参照物可以是垂直标定物。
根据本公开的实施例,标定参照物的类型至少包括铅锤板和/或建筑墙面等等。K个标定参照物可以属于相同类型的参照物,也可以是不同类型的参照物。例如,K个标定参照物可以都是铅锤板,或者,K个标定参照物中可以部分是铅锤板,部分是建筑墙面。
根据本公开的实施例,K的大小不做限定,例如,可以是2,3或者5等其它大于1的整数。
根据本公开的实施例,可以获取雷达针对K个标定参照物中的每个标定参照物分别进行数据采集持续预定时长后得到的K组点云数据集。
例如,K个标定参照物包括第一标定参照物和第二标定参照物。雷达可以持续预定时长对第一标定参照物进行数据采集,得到针对第一标定参 照物的一组点云数据集。然后,雷达可以持续预定时长对第二标定参照物进行数据采集,得到针对第二标定参照物的一组点云数据集。
根据本公开的实施例,预定时长不做限定,例如,可以是2秒,3秒,或者5秒等等,可以根据雷达的类型和采集效果进行确定。
根据本公开的实施例,以激光雷达采集K个标定参照物为例,当激光雷达在当前姿态下静止时,在一定时长内,在采集每个标定参照物的过程中,激光雷达的扫描路径不重复。该激光雷达可以是非重复扫描式的激光雷达。
针对非重复扫描式的激光雷达,由于其具有静态情况下可获得完全覆盖视场数据的测量特点,可以具有更好的标定精度。
根据本公开的实施例,在采集每个标定参照物的过程中,激光雷达的扫描路径可以呈曲线。
根据本公开的实施例,当激光雷达在当前姿态下静止时,在一定时长内,激光雷达的视场的扫描密度逐渐增加。
根据本公开的实施例,基于非重复扫描式的激光雷达的特点,在静止状态下,在很短的时间内即可达到激光雷达视场内接近100%的点云覆盖率,因此通过将标定参照物放入激光雷达视场内的方式可快速获取标定参照物的点云数据,提高了标定效率。
在操作S330,根据K组点云数据集确定在雷达坐标系下的第一重力向量。
在操作S340,获取惯性测量单元在雷达进行数据采集的过程中测量得到的第二重力向量。
根据本公开的实施例,可以通过惯性测量单元的加速度计测量得到重力向量在惯性测量单元坐标系下的数据,即测量得到第二重力向量。
根据本公开的实施例,可以将每一次改变雷达和惯性测量单元的姿态得到的第一重力向量和第二重力向量作为一组重力向量数据。
图4示意性示出了根据本公开实施例的通过激光雷达采集第一标定参照物和第二标定参照物的点云数据集的示意图。
如图4所示,激光雷达411和惯性测量单元412集成在同一个装置410上。可以在激光雷达411的视场内,分别对第一标定参照物420和第二标 定参照物430进行扫描。第一标定参照物420和第二标定参照物430可以非平行放置,保证其面法向量与重力向量垂直,同时要求在数据采集过程中保持静止。
基于非重复扫描式的激光雷达411的特点,在静止状态下,可以在很短的时间内即可达到视场内接近100%的点云覆盖率,因此通过将第一标定参照物420和第二标定参照物430放入激光雷达411的视场内的方式可快速获取标定参照物的点云数据。一次静态数据采集过程如图4所示,一次静态数据采集时长可以为t,例如可以是3秒。根据本公开的实施例,可以获取第一标定参照物420上的X个点,第二标定参照物430上的Y个点。通过惯性测量单元412可以获得t时间内a个加速度计测量值,基于a个加速度计测量值可以确定第二重力向量。
图5示意性示出了根据本公开实施例的根据K组点云数据集确定在雷达坐标系下的第一重力向量的流程图。
如图5所示,根据K组点云数据集确定在雷达坐标系下的第一重力向量包括操作S510~操作S520。
在操作S510,确定与每组点云数据集对应的平面法向量,得到K个平面法向量。
在操作S520,根据K个平面法向量计算在雷达坐标系下的第一重力向量。
根据本公开的实施例,以K个标定参照物包括第一标定参照物和第二标定参照物为例。
在雷达坐标系下,可以根据第一标定参照物的点云数据集拟合计算出该第一标定参照物的平面法向量V1。根据第二标定参照物的点云数据集拟合计算出该第二标定参照物的平面法向量V2。通过将平面法向量V1和平面法向量V2进行叉乘运算,计算得到第一重力向量在雷达坐标系下的表达,可表示为 Lv。
需要说明的是,本公开的标定参照物不限于第一标定参照物和第二标定参照物,还可以包括第三标定参照物等多个标定参照物。以K个标定参照物还包括第三标定参照物为例。在雷达坐标系下,可以根据第三标定参照物的点云数据集拟合计算出该第三标定参照物的平面法向量V3。通过将 平面法向量V1、平面法向量V2和平面法向量V3中的任意两个进行叉乘运算,得到多个计算结果,根据多个计算结果确定第一重力向量在雷达坐标系下的表达。例如,可以计算多个计算结果的平均值,将该平均值确定为第一重力向量。
图6示意性示出了根据本公开实施例的获取惯性测量单元在雷达进行数据采集的过程中测量得到的第二重力向量的流程图。
如图6所示,获取惯性测量单元在雷达进行数据采集的过程中测量得到的第二重力向量包括操作S610~操作S620。
在操作S610,获取惯性测量单元在雷达进行数据采集的过程中测量得到的多个重力测量向量。
根据本公开的实施例,在多次改变惯性测量单元和雷达的姿态之后,在雷达进行数据采集的过程中,惯性测量单元可以测量得到多个重力测量向量,可以用 Iv i表示在惯性测量单元坐标系下的重力测量向量。
在操作S620,根据多个重力测量向量确定第二重力向量。
根据本公开的实施例,可以从多个重力测量向量中确定出方差最小的重力测量向量作为第二重力向量。
根据本公开的实施例,或者,也可以计算多个重力测量向量的向量平均值,然后将向量平均值作为第二重力向量。根据本公开的实施例,可以将第二重力向量表示为 Iv。
根据本公开的实施例,在N组重力向量数据之后,其中,每组重力向量数据包括第一重力向量 Lv和第二重力向量 Iv,可以根据N组重力向量数据确定雷达坐标系和惯性测量单元坐标系之间的旋转变换参数。
具体地,可以利用N组重力向量数据,通过最小二乘算法求解得到雷达坐标系和惯性测量单元坐标系之间的旋转变换参数,求解方程可以为如下形式。
Figure PCTCN2020078068-appb-000001
根据本公开的实施例,通过最小二乘算法,使用N组重力向量数据求得上述式(一)的值最大,从而可以求解得到最优的旋转矩阵q,即雷达坐标系和惯性测量单元坐标系之间的旋转变换参数。
下面参考一具体实施例对参数标定方法作进一步说明。
图7示意性示出了根据本公开另一实施例的参数标定方法的流程图。
如图7所示,参数标定方法包括操作S710~操作S770。
在操作S710,选择具有先验信息的标定参照物,搭建标定场景。例如,在本公开中可使用两块吊起的铅锤板作为标定参照物,保证其面法向量与重力向量垂直,两块标定板非平行放置,同时要求在数据采集过程中保持静止,若建筑墙面为竖直,也可使用墙面作为标定参照物。
在操作S720,保持传感器静止不动,激光雷达采集标定参照物的点云数据和惯性测量单元测量重力加速度。
在操作S730,进行点云平面拟合,计算平面法向量后计算第一重力向量。
在操作S740,对重力加速度进行处理,计算在惯性测量单元坐标系下的第二重力向量。
在操作S750,改变激光雷达的姿态,重复上述操作S720~操作S740。
在操作S760,判断数据组数量是否满足数量要求。在满足的情况下执行S770,否则,执行操作S720。
在操作S770,使用N组重力向量数据,联合求解得到旋转变换参数。
通过本公开的实施例,由于是在静止状态下获得雷达和惯性测量单元在各自坐标系下的重力向量,并通过重力向量数据确定旋转变换参数,使得在参数标定过程中不涉及点云配准计算以及IMU数据的积分问题,避免了繁琐的点云匹配过程以及IMU积分计算过程,解决了采用相关技术进行参数标定时,标定精度较低,导致传感器测量结果的准确性较低的技术问题,达到了简化标定流程,获得稳定且较高的标定精度的技术效果。
图8示意性示出了根据本公开实施例的参数标定装置的框图。
如图8所示,参数标定装置800包括处理器810和存储器820。存储器820用于存储一个或多个程序,其中,当所述一个或多个程序被所述处理器810执行时,使得所述处理器实现如上所述的参数标定方法。
具体地,处理器810例如可以包括通用微处理器、指令集处理器和/或相关芯片组和/或专用微处理器(例如,专用集成电路(ASIC)),等等。处理器810还可以包括用于缓存用途的板载存储器。处理器810可以是用于执行根据本公开实施例的方法流程的不同动作的单一处理单元或者是 多个处理单元。
图9示意性示出了根据本公开实施例的参数标定系统的框图。
如图9所示,参数标定系统900包括雷达910、惯性测量单元920和参数标定装置930。
根据本公开的实施例,惯性测量单元920与雷达910固定连接。
根据本公开的实施例,参数标定装置930与图8所示的参数标定装置800相同。
根据本公开的实施例,还提供了一种可读存储介质,其上存储有可执行指令,该指令被处理器执行时使处理器执行如上所述的参数标定方法。
该可读存储介质可以是上述实施例中描述的设备/装置/系统中所包含的;也可以是单独存在,而未装配入该设备/装置/系统中。上述可读存储介质承载有一个或者多个程序,当上述一个或者多个程序被执行时,实现根据本公开实施例的方法。
根据本公开的实施例,可读存储介质可以是非易失性的可读存储介质,例如可以包括但不限于:便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本公开中,可读存储介质可以是任何包含或存储程序的有形介质,该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
附图中的流程图和框图,图示了按照本公开各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上,流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分,上述模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意,在有些作为替换的实现中,方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如,两个接连地表示的方框实际上可以基本并行地执行,它们有时也可以按相反的顺序执行,这依所涉及的功能而定。也要注意的是,框图或流程图中的每个方框、以及框图或流程图中的方框的组合,可以用执行规定的功能或操作的专用的 基于硬件的系统来实现,或者可以用专用硬件与计算机指令的组合来实现。
本领域技术人员可以清楚地了解到,为描述的方便和简洁,仅以上述各功能模块的划分进行举例说明,实际应用中,可以根据需要而将上述功能分配由不同的功能模块完成,即将装置的内部结构划分成不同的功能模块,以完成以上描述的全部或者部分功能。上述描述的装置的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
最后应说明的是:以上各实施例仅用以说明本公开的技术方案,而非对其限制;尽管参照前述各实施例对本公开进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分或者全部技术特征进行等同替换;在不冲突的情况下,本公开实施例中的特征可以任意组合;而这些修改或者替换,并不使相应技术方案的本质脱离本公开各实施例技术方案的范围。

Claims (17)

  1. 一种用于雷达和惯性测量单元的参数标定方法,其中,所述雷达和所述惯性测量单元固定连接,所述方法包括:
    获取N组重力向量数据,其中,所述N组重力向量数据中的每组重力向量数据是在改变所述雷达和所述惯性测量单元的姿态之后,并使所述雷达和所述惯性测量单元处于静止状态下得到的,所述每组重力向量数据包括在雷达坐标系下的第一重力向量和在惯性测量单元坐标系下的第二重力向量,所述N为大于2的整数;以及
    根据所述N组重力向量数据确定所述雷达坐标系和所述惯性测量单元坐标系之间的旋转变换参数。
  2. 根据权利要求1所述的方法,其中,获取N组重力向量数据包括:
    在每一次改变所述雷达和所述惯性测量单元的姿态之后,使所述雷达和所述惯性测量单元处于静止状态;
    获取K组点云数据集,其中,所述K组点云数据集是由所述雷达针对K个标定参照物中的每个标定参照物分别进行数据采集得到的,其中,K为大于1的整数;
    根据所述K组点云数据集确定在所述雷达坐标系下的第一重力向量;以及
    获取所述惯性测量单元在所述雷达进行数据采集的过程中测量得到的第二重力向量,
    其中,将每一次改变所述雷达和所述惯性测量单元的姿态得到的第一重力向量和第二重力向量作为一组重力向量数据。
  3. 根据权利要求2所述的方法,其中,根据所述K组点云数据集确定在所述雷达坐标系下的第一重力向量包括:
    确定与每组点云数据集对应的平面法向量,得到K个平面法向量;以及
    根据所述K个平面法向量计算在所述雷达坐标系下的第一重力向量。
  4. 根据权利要求2所述的方法,其中,获取K组点云数据集包括:
    获取所述雷达针对K个标定参照物中的每个标定参照物分别进行数据采集持续预定时长后得到的K组点云数据集。
  5. 根据权利要求2所述的方法,其中,获取所述惯性测量单元在所述雷达进行数据采集的过程中测量得到的第二重力向量包括:
    获取所述惯性测量单元在所述雷达进行数据采集的过程中测量得到的多个重力测量向量;以及
    根据所述多个重力测量向量确定所述第二重力向量。
  6. 根据权利要求5所述的方法,其中,根据所述多个重力测量向量确定所述第二重力向量:
    计算所述多个重力测量向量的向量平均值;以及
    将所述向量平均值作为所述第二重力向量。
  7. 根据权利要求2所述的方法,其中,所述每个标定参照物的面法向量与重力方向垂直,所述K个标定参照物非平行放置。
  8. 根据权利要求2所述的方法,其中,所述标定参照物的类型至少包括铅锤板和/或建筑墙面。
  9. 根据权利要求1所述的方法,其中,所述雷达包括激光雷达。
  10. 根据权利要求9所述的方法,其中,当所述激光雷达在当前姿态下静止时,在一定时长内,所述激光雷达的扫描路径不重复。
  11. 根据权利要求10所述的方法,其中,所述激光雷达的扫描路径呈曲线。
  12. 根据权利要求9所述的方法,其中,当所述激光雷达在当前姿态下静止时,在一定时长内,所述激光雷达的视场的扫描密度逐渐增加。
  13. 根据权利要求1所述的方法,其中,根据所述N组重力向量数据确定所述雷达坐标系和所述惯性测量单元坐标系之间的旋转变换参数包括:
    利用所述N组重力向量数据,通过最小二乘算法求解得到所述雷达坐标系和所述惯性测量单元坐标系之间的旋转变换参数。
  14. 根据权利要求1所述的方法,还包括:
    确定所述雷达坐标系和所述惯性测量单元坐标系之间的平移变换参数;以及
    将所述旋转变换参数和所述平移变换参数确定为所述雷达坐标系和所述惯性测量单元坐标系之间的坐标变换参数。
  15. 一种参数标定装置,包括:
    处理器;
    存储器,用于存储一个或多个程序,其中,当所述一个或多个程序被所述处理器执行时,使得所述处理器实现权利要求1至14中任一项所述的方法。
  16. 一种参数标定系统,包括:
    雷达;
    惯性测量单元,与所述雷达固定连接;
    参数标定装置,包括:
    处理器;
    存储器,用于存储一个或多个程序,其中,当所述一个或多个程序被所述处理器执行时,使得所述处理器实现权利要求1至14中任一项所述的方法。
  17. 一种可读存储介质,其上存储有可执行指令,该指令被处理器执行时使处理器执行权利要求1至14中任一项所述的方法。
PCT/CN2020/078068 2020-03-05 2020-03-05 参数标定方法、装置、系统和存储介质 WO2021174507A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202080004275.6A CN113767264A (zh) 2020-03-05 2020-03-05 参数标定方法、装置、系统和存储介质
PCT/CN2020/078068 WO2021174507A1 (zh) 2020-03-05 2020-03-05 参数标定方法、装置、系统和存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/078068 WO2021174507A1 (zh) 2020-03-05 2020-03-05 参数标定方法、装置、系统和存储介质

Publications (1)

Publication Number Publication Date
WO2021174507A1 true WO2021174507A1 (zh) 2021-09-10

Family

ID=77612815

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/078068 WO2021174507A1 (zh) 2020-03-05 2020-03-05 参数标定方法、装置、系统和存储介质

Country Status (2)

Country Link
CN (1) CN113767264A (zh)
WO (1) WO2021174507A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113870359A (zh) * 2021-09-23 2021-12-31 厦门大学 基于随机梯度下降的激光雷达外参标定方法及装置
CN114111672A (zh) * 2021-11-26 2022-03-01 南京航空航天大学 一种多位移传感器法向测量的传感器安装位置参数快速标定方法
CN114413887A (zh) * 2021-12-24 2022-04-29 北京理工大学前沿技术研究院 一种传感器外部参数标定方法、设备及介质
CN114581513A (zh) * 2022-03-07 2022-06-03 清华大学 一种空间坐标定位方法、装置及电子设备
CN115291196A (zh) * 2022-07-06 2022-11-04 南京牧镭激光科技有限公司 一种激光净空雷达安装姿态的标定方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9568592B1 (en) * 2014-11-04 2017-02-14 Google Inc. Automotive sensor alignment with external IMU tool
CN110109143A (zh) * 2019-04-30 2019-08-09 北京云迹科技有限公司 多线激光雷达与imu的位姿标定方法及装置
CN110244284A (zh) * 2019-07-29 2019-09-17 南通润邦重机有限公司 一种用于多线激光雷达和gps\ins标定的标定板及其方法
CN110686704A (zh) * 2019-10-18 2020-01-14 深圳市镭神智能系统有限公司 激光雷达与组合惯导的位姿标定方法、系统及介质
US20200041623A1 (en) * 2018-02-05 2020-02-06 Centre Interdisciplinaire De Developpement En Cartographie Des Oceans (Cidco) METHOD AND APPARATUS FOR AUTOMATIC CALIBRATION OF MOBILE LiDAR SYSTEMS
CN110849363A (zh) * 2019-12-03 2020-02-28 深圳市镭神智能系统有限公司 激光雷达与组合惯导的位姿标定方法、系统及介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9645250B2 (en) * 2015-04-07 2017-05-09 GM Global Technology Operations LLC Fail operational vehicle speed estimation through data fusion of 6-DOF IMU, GPS, and radar
CN109901139B (zh) * 2018-12-28 2023-07-04 文远知行有限公司 激光雷达标定方法、装置、设备和存储介质
CN110780285B (zh) * 2019-10-24 2022-10-18 深圳市镭神智能系统有限公司 激光雷达与组合惯导的位姿标定方法、系统及介质
CN110849362B (zh) * 2019-11-28 2022-01-04 湖南率为控制科技有限公司 一种基于车载惯性的激光雷达与视觉组合导航算法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9568592B1 (en) * 2014-11-04 2017-02-14 Google Inc. Automotive sensor alignment with external IMU tool
US20200041623A1 (en) * 2018-02-05 2020-02-06 Centre Interdisciplinaire De Developpement En Cartographie Des Oceans (Cidco) METHOD AND APPARATUS FOR AUTOMATIC CALIBRATION OF MOBILE LiDAR SYSTEMS
CN110109143A (zh) * 2019-04-30 2019-08-09 北京云迹科技有限公司 多线激光雷达与imu的位姿标定方法及装置
CN110244284A (zh) * 2019-07-29 2019-09-17 南通润邦重机有限公司 一种用于多线激光雷达和gps\ins标定的标定板及其方法
CN110686704A (zh) * 2019-10-18 2020-01-14 深圳市镭神智能系统有限公司 激光雷达与组合惯导的位姿标定方法、系统及介质
CN110849363A (zh) * 2019-12-03 2020-02-28 深圳市镭神智能系统有限公司 激光雷达与组合惯导的位姿标定方法、系统及介质

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113870359A (zh) * 2021-09-23 2021-12-31 厦门大学 基于随机梯度下降的激光雷达外参标定方法及装置
CN114111672A (zh) * 2021-11-26 2022-03-01 南京航空航天大学 一种多位移传感器法向测量的传感器安装位置参数快速标定方法
CN114413887A (zh) * 2021-12-24 2022-04-29 北京理工大学前沿技术研究院 一种传感器外部参数标定方法、设备及介质
CN114413887B (zh) * 2021-12-24 2024-04-02 北京理工大学前沿技术研究院 一种传感器外部参数标定方法、设备及介质
CN114581513A (zh) * 2022-03-07 2022-06-03 清华大学 一种空间坐标定位方法、装置及电子设备
CN114581513B (zh) * 2022-03-07 2024-04-19 清华大学 一种空间坐标定位方法、装置及电子设备
CN115291196A (zh) * 2022-07-06 2022-11-04 南京牧镭激光科技有限公司 一种激光净空雷达安装姿态的标定方法

Also Published As

Publication number Publication date
CN113767264A (zh) 2021-12-07

Similar Documents

Publication Publication Date Title
WO2021174507A1 (zh) 参数标定方法、装置、系统和存储介质
CN107340522B (zh) 一种激光雷达定位的方法、装置及系统
CN109084732A (zh) 定位与导航方法、装置及处理设备
CN109073407B (zh) 惯性测量单元的漂移标定方法、设备及无人飞行器
CN104268935A (zh) 一种基于特征的机载激光点云与影像数据融合系统及方法
WO2018227576A1 (zh) 地面形态检测方法及系统、无人机降落方法和无人机
JP5762131B2 (ja) キャリブレーション装置、キャリブレーション装置のキャリブレーション方法およびキャリブレーションプログラム
JP2016048172A (ja) 画像処理装置、画像処理方法、およびプログラム
Kersten et al. Comparative geometrical investigations of hand-held scanning systems
WO2020133172A1 (zh) 图像处理方法、设备及计算机可读存储介质
JP6333396B2 (ja) モバイルプラットフォームの変位を計測する方法及び装置
CN111308415B (zh) 一种基于时间延迟的在线估计位姿的方法和设备
CN111307046B (zh) 基于半球图像的树高测量方法
CN112955711A (zh) 位置信息确定方法、设备及存储介质
CN113111513B (zh) 传感器配置方案确定方法、装置、计算机设备及存储介质
WO2017038659A1 (ja) 運動検出装置及びそれを用いた三次元形状測定装置
CN111142514A (zh) 一种机器人及其避障方法和装置
CN111025330B (zh) 一种基于深度图的目标倾斜角检测方法及设备
CN117288151B (zh) 一种投影设备的三维姿态确定方法、装置和电子设备
CN114777768A (zh) 一种卫星拒止环境高精度定位方法、系统及电子设备
CN109658507A (zh) 信息处理方法及装置、电子设备
KR101821992B1 (ko) 무인비행체를 이용한 목표물의 3차원 위치 산출 방법 및 장치
CN109891188A (zh) 移动平台、摄像路径生成方法、程序、以及记录介质
WO2022256976A1 (zh) 稠密点云真值数据的构建方法、系统和电子设备
CN111161357B (zh) 信息处理方法及装置、增强现实设备和可读存储介质

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: 20923504

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: 20923504

Country of ref document: EP

Kind code of ref document: A1