Calibration method and calibration system for a laser radar

Abstract

A calibration method and a calibration system of a laser radar, the calibration method comprising: obtaining a calibration function based on imaging system parameters and detector array parameters of the laser radar, in combination with position information of a calibration target relative to the laser radar; performing at least three calibration receiving operations, the calibration receiving operations being suitable for obtaining position relationships between sampling detectors and the calibration target, the sampling detectors in each calibration receiving operation being different detectors; and fitting the calibration function based on the position relationships obtained in the at least three calibration receiving operations, to obtain internal parameters of the laser radar. Through fitting of the calibration function, all internal parameters of the laser radar can be directly obtained. Moreover, due to the relationship between the internal parameters and technical parameters of the laser radar, all internal parameters can be obtained through fitting of data collected in a limited number of calibration receiving operations, thereby effectively improving calibration speed and improving calibration efficiency.

Description

Technical Field

[0001] This invention relates to the field of lidar, and in particular to a lidar calibration method and calibration system. Background Technology

[0002] The calibration parameters of a lidar include internal and external parameters. Internal parameters define the positional relationship between each detector and the optical center coordinate system, determining the angle information corresponding to the data points acquired by each detector (pixel). External parameters refer to the coordinate system transformation relationship between the lidar and the calibration equipment. External parameters can generally be measured mechanically, but internal parameters cannot be directly measured and must be calibrated after the lidar is assembled.

[0003] Furthermore, the position of each detector may not be ideal. For example, in traditional mechanical radar, the detector is fixed to the circuit board with a patch, and its position may not be ideal. In other words, due to assembly and adjustment errors, lidar will always have a certain angular error. Therefore, after the lidar is assembled, each detector needs to be tested individually to obtain the angular information corresponding to each detector, so that the final point cloud map obtained by the lidar is closer to the actual situation.

[0004] For traditional mechanical or hybrid solid-state radars, the vertically arranged detectors determine their vertical field of view and angular resolution. Horizontal scanning is achieved through rotation (either the entire lidar transceiver module or by using a rotating mirror). Depending on the required line count, such lidars typically include 64 or 128 detectors, and calibration can be completed by calibrating each detector individually.

[0005] However, this method of calibrating detectors one by one is slow and inefficient; moreover, it cannot calibrate lidar with detector arrays. Summary of the Invention

[0006] The problem addressed by this invention is how to improve the calibration speed and efficiency of lidar; and how to calibrate lidar with detector arrays.

[0007] To address the above problems, this invention provides a calibration method for lidar, comprising:

[0008] Based on the imaging system parameters and detector array parameters of the lidar, and combined with the position information of at least one calibration target relative to the lidar, a calibration function is obtained; at least three calibration receiving operations are performed, the calibration receiving operations being suitable for obtaining the positional relationship between the sampling detector and the calibration target, and the sampling detector in each calibration receiving operation is a different detector; based on the positional relationship obtained from at least three calibration receiving operations, the calibration function is fitted to obtain the internal parameters of the lidar.

[0009] Optionally, the step of obtaining the calibration function includes: obtaining the external parameters of the lidar during the calibration process; and obtaining the calibration function based on the imaging system parameters and detector array parameters of the lidar, combined with the external parameters of the lidar and the position information of the calibration target.

[0010] Optionally, the step of obtaining the calibration function includes: obtaining the calibration function based on the imaging principle of a pinhole lens.

[0011] Optionally, the imaging system of the lidar includes a wide-angle lens.

[0012] Optionally, the calibration function is:

[0013]

[0014] Where i represents the number of calibration receiving operations, (Pix, Piy, Piz,) are the coordinates of the target object in the optical center coordinate system in the i-th calibration receiving operation, and m i Let n be the number of rows of the sampling detector during the i-th calibration receiving operation. i Let be the column number of the sampling detector during the i-th calibration and reception operation, and (x(mi,ni), y(mi,ni), z(mi,ni)) be the coordinates of the sampling detector in the optical center coordinate system during the i-th calibration and reception operation.

[0015] Optionally, the Levenberg-Marquardt algorithm may be used to solve the calibration function during the fitting process.

[0016] Optionally, the positional relationship between the sampling detector and the calibration target includes: the attitude angle corresponding to the sampling detector, wherein the attitude angle is the angle through which the lidar rotates when the sampling detector receives the echo signal; the calibration receiving operation includes: determining the sampling detector; rotating the lidar so that the sampling detector receives the echo signal reflected by the calibration target; and obtaining the attitude angle.

[0017] Optionally, in the step of performing at least 3 calibration reception operations, at least 6 calibration reception operations are performed; the step of fitting the calibration function includes: obtaining the internal parameters and external parameters of the lidar.

[0018] Optionally, the calibration target is fixed on the extension line of the optical axis of the lidar receiving optical component.

[0019] Optionally, in the calibration receiving operation, the step of rotating the lidar includes: rotating the lidar around a first axis and a second axis respectively, so that the lidar rotates, wherein the first axis and the second axis and the optical axis of the lidar are orthogonal to each other; in the calibration receiving operation, the step of obtaining the attitude angle includes: the attitude angle includes: a first attitude angle and a second attitude angle, wherein the first attitude angle is the angle through which the lidar rotates around the first axis, and the second attitude angle is the angle through which the lidar rotates around the second axis.

[0020] Optionally, it further includes: before obtaining the calibration function, obtaining the position information of the calibration target relative to the lidar, wherein the step of obtaining the position information of the calibration target relative to the lidar includes: establishing a world coordinate system with the intersection of the first rotation axis and the second rotation axis as the origin, and the first rotation axis being parallel to one coordinate axis of the world coordinate system, and the second rotation axis being parallel to the other coordinate axis of the world coordinate system; obtaining the coordinates of the calibration target and the lidar in the world coordinate system to obtain the position information of the calibration target relative to the lidar.

[0021] Optionally, the calibration target has a preset size so that only one detector can acquire the echo signal.

[0022] Optionally, the size of the calibration target is smaller than the angular resolution of the lidar.

[0023] Optionally, in the calibration receiving operation, in the step of rotating the lidar so that the sampling detector receives the echo signal, the lidar is rotated so that the echo signal received by the sampling detector is the strongest.

[0024] Optionally, the sampling detector is selected from the apex detector, midpoint detector, and center detector of the detector array, wherein the apex detector is the detector located at the apex of the detector array, the midpoint detector is the detector located at the midpoint of each side of the detector array, and the center detector is the detector located at the center of symmetry of the detector array.

[0025] Optionally, in the step of performing at least 3 calibration and reception operations, the calibration and reception operations are performed 4 to 9 times.

[0026] Accordingly, the present invention also provides a calibration system for lidar, comprising:

[0027] A mechanical module is suitable for fixing the lidar to be calibrated and at least one calibration target; the mechanical module performs at least three calibration receiving operations, the calibration receiving operations being suitable for obtaining the positional relationship between the sampling detector and the calibration target, and the sampling detector in each calibration receiving operation is a different detector; a fitting module is suitable for fitting a preset calibration function based on the positional relationship obtained from at least three calibration receiving operations to obtain the internal parameters of the lidar, wherein the calibration function is obtained based on the imaging system parameters and detector array parameters of the lidar, combined with the positional information of the calibration target relative to the lidar.

[0028] Optionally, the calibration function is obtained based on the pinhole lens imaging principle.

[0029] Optionally, the imaging system of the lidar includes a wide-angle lens.

[0030] Optionally, the calibration function is:

[0031]

[0032] Where i represents the number of calibration receive operations. Let m be the coordinates of the target object in the optical center coordinate system during the i-th calibration receiving operation. i Let n be the number of rows of the sampling detector during the i-th calibration receiving operation. i Let x(m) be the number of columns of the sampling detector during the i-th calibration receiving operation. i n i ), y(m i n i ), z(m i n i )) represents the coordinates of the downsampling detector in the optical center coordinate system during the i-th calibration receiving operation, and

[0033] Optionally, the fitting module uses the Levenberg-Marquardt algorithm to solve the problem.

[0034] Optionally, the positional relationship between the sampling detector and the calibration target includes: the attitude angle corresponding to the sampling detector, wherein the attitude angle is the angle through which the lidar rotates when the sampling detector receives the echo signal; the calibration system further includes: an operation module, wherein the operation module is adapted to perform the calibration receiving operation through the mechanical module to obtain the attitude angle.

[0035] Optionally, the operation module includes: a sampling unit adapted to determine the sampling detector; an operation unit that rotates the lidar via the mechanical module so that the sampling detector receives the echo signal reflected by the calibration target; and a data unit that obtains the attitude angle via the mechanical module.

[0036] Optionally, the sampling detector is selected from the apex detector, midpoint detector, and center detector of the detector array, wherein the apex detector is the detector located at the apex of the detector array, the midpoint detector is the detector located at the midpoint of each side of the detector array, and the center detector is the detector located at the center of symmetry of the detector array.

[0037] Optionally, at least six calibration reception operations are performed; the fitting module obtains the internal and external parameters of the lidar.

[0038] Optionally, the mechanical module includes: a turntable having a table surface, and the turntable also having a vertical axis perpendicular to the table surface and a horizontal axis parallel to the table surface; a lidar fixed to the table surface of the turntable, the optical axis of the lidar being parallel to the table surface of the turntable and perpendicular to the horizontal axis; the attitude angle includes: a first attitude angle and a second attitude angle, the first attitude angle being the angle through which the table surface of the turntable rotates about the vertical axis, and the second attitude angle being the angle through which the table surface of the turntable rotates about the horizontal axis.

[0039] Optionally, it also includes: a measurement module, which is adapted to obtain the coordinates of the calibration target and the coordinates of the lidar in a world coordinate system to obtain the position information of the calibration target relative to the lidar. The measurement module is pre-set with a world coordinate system, the origin of which is the intersection of the vertical rotation axis and the horizontal rotation axis, one coordinate axis of the world coordinate system is parallel to the vertical rotation axis, and the other coordinate axis of the world coordinate system is parallel to the horizontal rotation axis.

[0040] Optionally, the calibration target has a preset size so that only one detector can acquire the echo signal.

[0041] Optionally, the size of the calibration target is smaller than the angular resolution of the lidar.

[0042] Compared with the prior art, the technical solution of the present invention has the following advantages:

[0043] In the technical solution of this invention, the calibration function is obtained based on the imaging system parameters and detector array parameters of the lidar, combined with the preset position information of the calibration target relative to the lidar. Therefore, by fitting the calibration function, all internal parameters of the lidar can be directly obtained. Moreover, due to the relationship between the internal parameters and the technical parameters of the lidar, all internal parameters can be obtained by fitting the data collected through as few as three calibration receiving operations, thereby effectively improving the calibration speed and efficiency.

[0044] In an optional embodiment of the present invention, the data collected through as few as six calibration receiving operations can not only fit and obtain all the internal parameters of the lidar, but also obtain the external parameters of the lidar during the calibration process, thereby effectively improving data accuracy and reducing random errors caused by measurement.

[0045] In an optional embodiment of the present invention, the calibration target is positioned on the extension line of the lidar optical axis. Positioning the calibration target on the extension line of the lidar optical axis, utilizing the symmetry of the lidar imaging system's optical path, effectively simplifies the calibration function used for fitting, reduces the amount of fitting calculations, and improves calculation speed and efficiency.

[0046] In an optional embodiment of the present invention, the size of the calibration target is smaller than the angular resolution of the lidar, and the calibration target has a preset size so that only one detector can acquire the echo signal. Using a sufficiently small calibration target, making its size smaller than the angular resolution of the lidar, allows only one detector to acquire the echo signal reflected by the calibration target, effectively reducing the difficulty of calibration reception and improving calibration accuracy and precision. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating an embodiment of the lidar calibration method of the present invention;

[0048] Figure 2 yes Figure 1 A schematic diagram showing the relative positions of the lidar and the calibration target in the embodiment of the calibration method.

[0049] Figure 3 yes Figure 1 The diagram shows the structure of the detector array inside the lidar in the embodiment of the calibration method.

[0050] Figure 4 This is a schematic diagram of the optical path of the lidar's internal pinhole imaging and f-theta lens imaging in the calibration method embodiment shown;

[0051] Figure 5 yes Figure 1 A flowchart illustrating the calibration receiving operation steps in the embodiment of the calibration method shown;

[0052] Figure 6 yes Figure 1 A schematic diagram of the location of the sampling detector determined in the embodiment of the calibration method shown;

[0053] Figure 7 This is a schematic diagram of an embodiment of the lidar calibration system of the present invention;

[0054] Figure 8 yes Figure 7 The diagram shows the functional block diagram of the operation module in the calibration system embodiment. Detailed Implementation

[0055] As the background technology shows, existing methods for calibrating lidar suffer from slow calibration speed and low efficiency, and they are also unable to calibrate lidar with detector arrays. This paper analyzes the reasons for the slow speed, low efficiency, and inability to calibrate lidar with detector arrays using existing lidar calibration methods:

[0056] Current methods for calibrating lidar involve calibrating each detector at the lidar receiver individually, resulting in slow speed and low efficiency. Furthermore, for lidar receivers without any mechanical scanning components, the number and area of ​​the receiver detectors determine the lidar's vertical field of view and angular resolution. Lidar receivers typically employ integrated detector arrays, where each detector is very small; a detector array may contain tens of thousands, or even millions, of detectors arranged in an array. Therefore, for lidar receivers with detector arrays, calibrating angle information by calibrating each detector individually is impractical.

[0057] To solve the aforementioned technical problem, the present invention provides a calibration method for lidar, comprising:

[0058] Based on the imaging system parameters and detector array parameters of the lidar, and combined with the position information of at least one pre-set calibration target relative to the lidar, a calibration function is obtained; at least three calibration receiving operations are performed, the calibration receiving operations being suitable for obtaining the positional relationship between the sampling detector and the calibration target, and the sampling detector in each calibration receiving operation is a different detector; based on the positional relationship obtained from at least three calibration receiving operations, the calibration function is fitted to obtain the internal parameters of the lidar.

[0059] In the technical solution of this invention, the calibration function is obtained based on the imaging system parameters and detector array parameters of the lidar, combined with the preset position information of the calibration target relative to the lidar. Therefore, by fitting the calibration function, all internal parameters of the lidar can be directly obtained. Moreover, due to the relationship between the internal parameters and the technical parameters of the lidar, all internal parameters can be obtained by fitting the data collected through as few as three calibration receiving operations, thereby effectively improving the calibration speed and efficiency.

[0060] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0061] refer to Figure 1 The diagram shows a flowchart of an embodiment of the lidar calibration method of the present invention.

[0062] The lidar calibration method includes: first, executing step S110, obtaining a calibration function based on the lidar's imaging system parameters and detector array parameters, combined with preset position information of the calibration target relative to the lidar; then executing step S120, performing at least three calibration receiving operations, the calibration receiving operations being suitable for obtaining the positional relationship between the sampling detector and the calibration target, with each calibration receiving operation using a different sampling detector; and then executing step S130, fitting the calibration function based on the positional relationship obtained from at least three calibration receiving operations to obtain the lidar's internal parameters.

[0063] The internal parameters of the lidar can be obtained directly by fitting the calibration function. Data acquisition can be achieved with as few as three calibration receiving operations, which can effectively improve calibration speed and efficiency. In particular, for the detector array of solid-state lidar, it makes the calibration of its internal parameters possible.

[0064] It should be noted that in some embodiments of the present invention, the receiving end of the lidar includes a detector array. Therefore, the area of ​​the detector array and the number of detectors therein determine the field of view and angular resolution of the lidar. In some embodiments of the present invention, the detector array is an integrated detector array, in which each detector is very small, and the detectors contained in the detector array are arranged in a two-dimensional array, and the number is huge, reaching tens of thousands. It is evident that the technical solution of obtaining all internal parameters of the lidar by fitting a calibration function based on data collected from a finite number of calibration receiving operations makes the calibration of the area array detector possible, effectively improving the measurement accuracy of solid-state lidar.

[0065] In some embodiments of the present invention, the lidar is a solid-state lidar, that is, the lidar does not include any mechanical rotating or scanning components.

[0066] The calibration function is suitable for characterizing the relationship between the imaging system parameters of the lidar, the detector array parameters, the position information of the calibration target relative to the lidar, and the positional relationship between the sampling detector and the calibration target.

[0067] Since the calibration function is related to the position information of the calibration target relative to the lidar, therefore... Figure 1 As shown, in some embodiments of the present invention, the calibration method further includes: before obtaining the calibration function, performing step S101 to fix the relative positions of the lidar and the calibration target; and performing step S102 to obtain the position information of the calibration target relative to the lidar.

[0068] Reference Figure 2 , showed Figure 1 A schematic diagram showing the relative positions of the lidar and the calibration target in the embodiment of the calibration method.

[0069] In some embodiments of the present invention, in the step of fixing the relative positions of the lidar and the calibration target, the lidar 100 is fixed on a turntable. For example... Figure 2 As shown, the turntable has a platform 110, a vertical axis 111 perpendicular to the platform, and a horizontal axis 112 parallel to the platform. The platform 110 rotates about the vertical axis 111 to achieve horizontal rotation, and the platform 110 rotates about the horizontal axis 112 to achieve pitch rotation. The lidar 100 is fixed on the platform 110, and the optical axis of the lidar is perpendicular to both the vertical axis 111 and the horizontal axis 112; that is, the vertical axis 111, the horizontal axis 112, and the optical axis of the lidar are all orthogonal to each other.

[0070] The turntable's platform 110 is capable of horizontal and vertical rotation. The lidar 100 is fixed to the turntable's platform 110, and the lidar 100's horizontal and vertical attitude is adjusted by rotating the turntable. Using a turntable capable of horizontal and vertical rotation to adjust the lidar 100's attitude allows for precise acquisition of the adjusted attitude position, which is beneficial for improving calibration accuracy. Moreover, high-precision turntables are readily available and easy to operate, thus reducing calibration difficulty and equipment costs.

[0071] Since the lidar 100 is fixed to the platform 110 of the turntable, the lidar can rotate around the first axis and the second axis to change its posture. The first axis, the second axis, and the optical axis of the lidar are all orthogonal to each other, and the first axis and the second axis of the lidar 100 coincide with the vertical axis 111 and the horizontal axis 112 of the turntable, respectively.

[0072] The calibration target 120 is used to reflect the probe light to form an echo signal during the calibration receiving operation, and therefore the calibration target 120 is set within the field of view of the lidar.

[0073] It should be noted that, Figure 2 The schematic turntable 110 has two rotating shafts 111 and 112. In other embodiments of the invention, the horizontal rotating shaft can be... Figure 2 The schematic diagram shows the directions in which the rotating axes 112 are orthogonal. Alternatively, the turntable has three rotating axes, including a vertical axis and two horizontal axes, the two horizontal axes being orthogonal to each other in the horizontal direction.

[0074] like Figure 2 As shown, in some embodiments of the present invention, the calibration target 120 is fixed on the extension line of the optical axis of the receiving optical component of the lidar 100, so that the lidar beam is incident orthogonally on the calibration target 120, avoiding measurement deviations caused by oblique incidence. Furthermore, the fact that the calibration target 120 is fixed on the extension line of the optical axis of the receiving optical component of the lidar 100 ensures that the position parameter of the calibration target has only one dimension, thereby simplifying calculations.

[0075] In some embodiments of the present invention, the calibration target 120 has a preset size so that only one detector can acquire the echo signal. Controlling the size of the calibration target 120 so that only one detector can acquire the echo signal ensures that only one detector can measure the data in the data acquired after the calibration receiving operation, which is beneficial for improving the data acquisition accuracy during the calibration receiving operation. Specifically, the size of the calibration target 120 is smaller than the angular resolution of the lidar 100. However, using a relatively small calibration target is only one example; in other embodiments of the present invention, a slightly larger calibration target can be selected.

[0076] It should be noted that, as Figure 2 As shown, in some embodiments of the present invention, the calibration target 120 is square; in other embodiments of the present invention, the calibration target may also be circular or other shapes. The present invention does not limit the shape of the calibration target.

[0077] It should be noted that in some embodiments of the present invention, only one calibration target is provided. In other embodiments of the present invention, multiple calibration targets are provided, thereby reducing the rotation angle of the lidar during calibration and receiving operations, enabling faster calibration and receiving operations, and improving calibration efficiency.

[0078] Continue to refer to Figure 1 and Figure 2 After fixing the lidar 100 and the calibration target 120, step S102 is executed to obtain the position information of the calibration target 120 relative to the lidar 100.

[0079] Specifically, in some embodiments of the present invention, the step of obtaining the position information of the calibration target 120 relative to the lidar 100 includes: using the first axis of rotation (i.e. Figure 2 The vertical axis 111 of the turntable and the second axis (i.e., Figure 2 A world coordinate system is established with the intersection of the horizontal rotation axes 112 of the turntable as the origin. The first rotation axis is parallel to one axis of the world coordinate system, and the second rotation axis is parallel to the other axis of the world coordinate system. The coordinates of the calibration target and the coordinates of the lidar are obtained in the world coordinate system to obtain the position information of the calibration target relative to the lidar. Figure 1 and Figure 2 In the illustrated embodiment, in the step of obtaining the coordinates of the calibration target 120, the position information of the calibration target 120 relative to the lidar 100 in the world coordinate system is obtained by external mechanical measurement.

[0080] It should be noted that the first rotation axis is parallel to one coordinate axis of the world coordinate system, the second rotation axis is parallel to another coordinate axis of the world coordinate system, and the optical axis of the lidar is orthogonal to both the first and second rotation axes. Therefore, the optical axis of the lidar is parallel to the third coordinate axis of the world coordinate system.

[0081] It should also be noted that the vertical axis 111 and horizontal axis 112 of the turntable coincide with the first axis and second axis of the lidar 101, respectively. Therefore, the origin of the world coordinate system is the intersection of the vertical axis 111 and horizontal axis 112 of the turntable. The two coordinate axes of the world coordinate system are parallel to the vertical axis 111 and horizontal axis 112 of the turntable, respectively. In other words, the world coordinate system can also be regarded as a coordinate system established with the intersection of the vertical axis 111 and the horizontal axis 112 as the origin, and the vertical axis 111 being parallel to one coordinate axis and the horizontal axis 112 being parallel to the other coordinate axis.

[0082] Specifically, the position information of the calibration target 120 relative to the lidar 100 in the world coordinate system is obtained through external mechanical measurement, that is, the coordinates of the calibration target in the world coordinate system are P0(A,B,C).

[0083] It should be noted that after the coordinate system is rotated, the coordinates of the calibration target 120 can be transformed through coordinate transformation. For example, when the world coordinate system is rotated by θ and θ respectively around two coordinate axes (the two coordinate axes correspond to the first rotation axis and the second rotation axis respectively), the coordinates of the calibration target 120 can be transformed. When the angle is specified, the coordinate transformation of the calibration target 120 is P = R * P0. Where R is the rotation matrix, as follows:

[0084]

[0085] On the other hand, the internal parameters of the lidar refer to the positions of each detector of the lidar in the optical center coordinate system. Therefore, the coordinates of the calibration target 120 are further transformed into the optical center coordinate system.

[0086] Specifically, based on the offset and orientation of the optical center coordinate system relative to the world coordinate system, the coordinates of the calibration target 120 in the world coordinate system are transformed to the optical center coordinate system. The offset and orientation of the optical center coordinate system relative to the world coordinate system include: translation of the optical center coordinate system relative to the world coordinate system, i.e., the coordinates (a, b, c) between the optical center (i.e., the origin of the optical center coordinate system) and the rotation center (i.e., the origin of the world coordinate system); and rotation of the optical center coordinate system relative to the world coordinate system, i.e., the angles (θx, θy, θz) between the corresponding coordinate axes of the world coordinate system and the optical center coordinate system.

[0087] Therefore, when the world coordinate system rotates by θ around two coordinate axes (the two coordinate axes correspond to the first rotation axis and the second rotation axis respectively), respectively... When the angle is determined, the coordinates (Px, Py, Pz) of the calibration target 120 in the optical center coordinate system are obtained according to the rotation matrix R, wherein the transformation relationship P'0 is as follows:

[0088]

[0089]

[0090] Continue to refer to Figure 1 After obtaining the relative positional relationship, step S110 is executed to obtain the calibration function.

[0091] Specifically, in some embodiments of the present invention, the step of obtaining the calibration function includes: obtaining the calibration function based on the pinhole lens imaging principle, that is, obtaining the calibration function based on the optical path of pinhole imaging. Therefore, in the process of obtaining the calibration function, the sampling detector, the optical center, and the calibration target are collinear. Obtaining the calibration function based on the pinhole imaging principle appropriately simplifies the optical path structure of the lidar, which is beneficial to simplifying the calibration function and thus reducing the amount of fitting calculations.

[0092] To align the detector receiving the echo signal, the optical center, and the calibration target 102 in a straight line, the attitude of the lidar 100 needs to be adjusted. Specifically, when the lidar 100 rotates θ around the first axis and the second axis respectively... When the angle is such that the above three are collinear, it is equivalent to the world coordinate system rotating by θ around two coordinate axes (the two coordinate axes correspond to the first rotation axis and the second rotation axis, respectively). At the angle specified, the detector receiving the echo signal, the optical center, and the calibration target 102 are collinear. Therefore, as mentioned above, the coordinates (Px, Py, Pz) of the calibration target 120 in the optical center coordinate system are:

[0093]

[0094] Reference Figure 3 , showed Figure 1 The diagram shows the structure of the detector array inside the lidar in the embodiment of the calibration method.

[0095] like Figure 3 As shown, the detector array 101 of the lidar is a surface array manufactured using semiconductor technology. Each pixel of the detector array 101 is a detector. In the optical center coordinate system (e.g., Figure 2 In the optical center coordinate system (shown in coordinate system 130), assuming the position of the central detector (i.e., the central pixel) is X0(x(0,0),y(0,0),z(0,0)), the spacing vector along the row direction (as shown in the horizontal x-direction) is K(Kx,Ky,Kz), and the spacing vector along the column direction (as shown in the vertical y-direction) is L(Lx,Ly,Lz). Therefore, the coordinates X(x(m,n),y(m,n),z(m,n)) of any detector in the optical center coordinate system can be obtained as follows:

[0096] x(m,n)=x(0,0)+m×Kx+n×Lx

[0097] y(m,n)=y(0,0)+m×ky+N×Ly

[0098] z(m,n)=z(0,0)+m×Kz+n×Lz

[0099] In some embodiments of the present invention, the imaging system of the lidar includes a wide-angle lens. As an example, the wide-angle lens of the lidar includes an f-theta lens. Therefore, although the calibration function is obtained based on the pinhole imaging principle, the echo signal ultimately needs to be transmitted through the lidar's imaging system before it can be received by the corresponding detector.

[0100] Reference Figure 4 The diagram shows the optical path of pinhole imaging and f-theta lens imaging.

[0101] like Figure 4As shown, the calibration function is obtained based on the pinhole imaging principle, meaning that the detector receiving the echo signal, the optical center, and the calibration target 102 are on a straight line. However, since the imaging system of the lidar 100 includes an f-theta lens, which can be considered as refracting light onto the detector, the detector receiving the echo signal, the optical center, and the calibration target 102 are not on a straight line when using an f-theta lens. Therefore, it is necessary to transform the coordinates X(x(m,n), y(m,n), z(m,n)) of any detector in the optical center coordinate system to the coordinate position FT(m,n) corresponding to the rectilinear propagation of light through the pinhole lens:

[0102]

[0103]

[0104] Where f is the equivalent focal length of a pinhole lens, and Z(m,n) is the focal length of an f-theta lens (e.g., ...). Figure 4 As shown in the figure, FT(m,n) is the position of the detector receiving the echo signal under the equivalent pinhole lens.

[0105] Therefore, the coordinates of the calibration target 120, its optical center, and the coordinates of the detector receiving the echo signal are related, and the relationship between their coordinates can be expressed as follows:

[0106]

[0107]

[0108] The parameters of the above equation include: the positional relationship between the calibration target 120, the detector receiving the echo signal, and the optical center when they are collinear, wherein the positional relationship between the detector receiving the echo signal and the calibration target includes the attitude angle corresponding to the detector receiving the echo signal, wherein the attitude angle is the angle rotated by the lidar when the detector receiving the echo signal receives the echo signal. Specifically, as shown... Figure 2 In the illustrated embodiment, the attitude angle of the lidar is the angle at which the lidar rotates about the vertical axis 111 and the horizontal axis 112, respectively, by rotations of θ and θ. The angle.

[0109] The parameters in the above equation also include the offset and attitude of the optical center coordinate system in the world coordinate system, i.e., the external parameters (a,b,c) and (θx, θy, θz) of the lidar, and the position X0(x(0,0),y(0,0),z(0,0)) of the center pixel of the detector array 101, i.e., three internal parameters, as well as the two spacing vectors K(Kx,Ky,Kz) / L(Lx,Ly,Lz) of the detector array 101, i.e., six internal parameters. In other words, of the 12 parameters in the above equation, there are 6 external parameters and 9 internal parameters.

[0110] The spacing of the detector array 101 is one of the parameters of the detector array, which is determined during the design and production of the lidar. In other words, the spacing of the detector array 101 can be obtained when the lidar is determined. The spacing vectors K(Kx,Ky,Kz) and L(Lx,Ly,Lz) in the optical center coordinate system can be obtained using the orientation of the detector array 101 in the optical center coordinate system, i.e., the angles (α, β, γ) between the plane of the detector array 101 and each coordinate axis of the optical center coordinate system, which are three internal parameters. That is, the two spacing vectors K(Kx,Ky,Kz) / L(Lx,Ly,Lz) of the detector array 101 are related to the other three internal parameters. Therefore, the six internal parameters K(Kx,Ky,Kz) / L(Lx,Ly,Lz) can be simplified to three internal parameters (α, β, γ) for calibration.

[0111] Therefore, when the lidar 100, in a certain posture, causes one detector on the detector array 101, i.e. one pixel of the detector array 101, to receive the echo signal reflected by the calibration target 120, two equations can be obtained.

[0112] Specifically, in some embodiments of the present invention, the lidar 100, in the i-th attitude, causes one detector on the detector array 101, i.e., one pixel of the detector array 101, to receive the echo signal reflected by the calibration target 120, wherein the calibration function is:

[0113]

[0114] Among them, (P) i x ,P i y ,P i z ,) represents the coordinates of the target object used for calibration in the optical center coordinate system at the i-th attitude of the lidar 100, m i Let n be the number of rows of the sampling detector for the lidar 100 in the i-th pose.i Let x(m) be the number of columns of the sampling detectors of the lidar 100 in the i-th pose. i ,n i ),y(m i ,n i ),z(m i ,n i )) represents the coordinates of the sampling detector in the optical center coordinate system of the lidar 100 in the i-th attitude, and

[0115] Continue to refer to Figure 1 After obtaining the calibration function, step S120 is executed to perform at least three calibration receiving operations.

[0116] The calibration receiving operation is suitable for obtaining the positional relationship between the sampling detector and the calibration target, and the sampling detector used in each calibration receiving operation is a different detector.

[0117] According to the calibration function, the positional relationship between the sampling detector and the calibration target includes: the attitude angle corresponding to the sampling detector, where the attitude angle is the angle through which the lidar rotates when the sampling detector receives the echo signal.

[0118] When the lidar 100, in its i-th attitude, causes one detector on the detector array 101 to receive the echo signal reflected by the calibration target 120, two equations can be obtained. Therefore, to obtain the six internal parameters, at least three calibration reception operations are required.

[0119] Reference Figure 5 , showed Figure 1 The flowchart of the calibration receiving operation steps in the embodiment of the calibration method for the lidar shown is illustrated.

[0120] In some embodiments of the present invention, the calibration receiving operation includes: firstly, executing step S121 to determine the sampling detector; then executing step S122 to rotate the lidar so that the sampling detector receives the echo signal reflected by the calibration target; and then executing step S123 to obtain the attitude angle.

[0121] The sampling detector is a detector that receives the echo signal reflected by the calibration target during each calibration receiving operation. The sampling detector determined for each calibration receiving operation is a different detector on the detector array 101.

[0122] Reference Figure 6 , showed Figure 1 The diagram shows the location of the sampling detector in the embodiment of the lidar calibration method.

[0123] like Figure 6 As shown, in some embodiments of the present invention, the sampling detector is selected from the apex detector 101a, the midpoint detector 102b, and the center detector 101c of the detector array. The apex detector 101a is the detector located at the apex of the detector array 101, the midpoint detector 101b is the detector located at the midpoint of each side of the detector array 101, and the center detector 101c is the detector located at the center of symmetry of the detector array 101. Selecting the sampling detector within the range of the apex detector 101a, the midpoint detector 102b, and the center detector 101c can improve calibration accuracy. Specifically, referring to the reference... Figure 3 In the step of determining the sampling detector, the position coordinates of the sampling detector in the detector array 101 are (m,n).

[0124] It should be noted that, based on the selection range of the sampling detector, in some embodiments of the present invention, the calibration and reception operation is performed 4 to 9 times in the step of performing at least 3 calibration and reception operations.

[0125] Continue to refer to Figure 5 After determining the detector to be used, step S122 is executed to rotate the lidar so that the sampling detector receives the echo signal.

[0126] Rotating the lidar is to adjust its attitude so that the determined sampling detector, optical center, and calibration target 120 are collinear.

[0127] In some embodiments of the present invention, the step of rotating the lidar in the calibration receiving operation includes: rotating the lidar around a first axis and a second axis respectively, so as to rotate the lidar, wherein the first axis and the second axis and the optical axis of the lidar are orthogonal to each other.

[0128] Specifically, such as Figure 2 As shown, in some embodiments of the present invention, in the step of fixing the relative position of the lidar and the calibration target, the lidar 100 is fixed on a turntable, that is, the lidar 100 is fixed on a turntable having a vertical rotation axis 111 and a horizontal rotation axis 112, and the optical axis of the lidar is orthogonal to both the vertical rotation axis 111 and the horizontal rotation axis 112. Therefore, the attitude of the lidar can be adjusted by directly rotating the lidar on the turntable. The vertical rotation axis 111 and the horizontal rotation axis 112 of the turntable coincide with the first rotation axis and the second rotation axis of the lidar 100, respectively.

[0129] It should be noted that when the lidar is provided, its field of view (FoV) can be obtained. The approximate angle of rotation of the lidar can be determined based on the field of view of the lidar. That is, the steps of rotating the lidar include: obtaining a first adjustment angle based on the field of view of the lidar; and rotating the lidar through the first adjustment angle.

[0130] After the lidar is rotated past the first adjustment angle, it is positioned in a preset attitude. The step of rotating the lidar after it is in the preset attitude further includes fine-tuning its attitude. In this fine-tuning step, the lidar's attitude is adjusted by a small angle near the preset attitude to ensure that the sampling detector receives the echo signal.

[0131] It should be noted that in some embodiments of the present invention, the size of the calibration target 120 is small, enabling only one detector to acquire the echo signal; or, the size of the calibration target 120 is smaller than the angular resolution of the lidar 100. In the calibration receiving operation, in the step of rotating the lidar to allow the sampling detector to receive the echo signal, rotating the lidar 100 only allows the echo signal received by the sampling detector to be received.

[0132] However, using a relatively small calibration target is only one example. In other embodiments of the present invention, a slightly larger calibration target can be selected. In some other embodiments of the present invention, the calibration target is slightly larger, so in the calibration receiving operation, in the step of rotating the lidar so that the sampling detector receives the echo signal, the lidar is rotated to make the echo signal received by the sampling detector the strongest. Alternatively, in some other embodiments of the present invention, in the step of making the sampling detector receive the echo signal, the lidar is rotated so that the sampling detector is located in the middle position of a plurality of detectors whose received echo signal signal strength exceeds a preset threshold.

[0133] Continue to refer to Figure 5 After adjusting the attitude of the lidar, step S123 is executed to obtain the attitude angle corresponding to the sampling detector.

[0134] Specifically, since the attitude adjustment of the lidar is achieved by rotating the lidar 100 around the first axis and the second axis respectively, the step of obtaining the attitude angle includes: the attitude angle includes: a first attitude angle and a second attitude angle, wherein the first attitude angle is the angle through which the lidar rotates around the first axis, and the second attitude angle is the angle through which the lidar rotates around the second axis.

[0135] As mentioned earlier, at least three calibration reception operations are required to obtain the six intrinsic parameters. Specifically, in the i-th calibration reception operation, the determined position of the sampling detector is (m... i ,n i The attitude angle corresponding to the sampling detector is obtained as follows:

[0136] In each calibration reception operation, the attitude of the lidar 100 is different. Therefore, during the i-th calibration reception operation, the lidar 100 is in the i-th attitude. Thus, the calibration function is:

[0137]

[0138] Among them, (P) i x ,P i y ,P i z ,) represents the coordinates of the target object used for calibration in the optical center coordinate system at the i-th attitude of the lidar 100, m i Let n be the number of rows of the sampling detector for the lidar 100 in the i-th pose. i Let x(m) be the number of columns of the sampling detectors of the lidar 100 in the i-th pose. i ,n i ),y(m i ,n i ),z(m i ,n i )) represents the coordinates of the sampling detector in the optical center coordinate system of the lidar 100 in the i-th attitude, and

[0139] Therefore, by substituting the positions of all sampling detectors in the calibration receiving operation and the attitude angles corresponding to the sampling detectors into the above calibration function and fitting them, the internal parameters of the lidar 100 are obtained.

[0140] Specifically, each pixel of the detector array 101 is a detector. The position of the center pixel of the detector array 101 is X0(x(0,0),y(0,0),z(0,0)), the spacing vector along the row direction (the horizontal x direction in the figure) is K(Kx,Ky,Kz), and the spacing vector along the column direction (the vertical y direction in the figure) is L(Lx,Ly,Lz).

[0141] Therefore, the coordinates of each detector on each of the detector arrays 101 in the optical center coordinate system are:

[0142] x(m,n)=x(0,0)+m×Kx+n×Lx

[0143] y(m,n)=y(0,0)+m×Ky+n×Ly

[0144] z(m,n)=z(0,0)+m×Kz+n×Lz

[0145] Among them, the interval vector K(Kx,Ky,Kz) and the interval vector L(Lx,Ly,Lz) are one of the parameters of the detector array, and the position and orientation of the detector array 101, including the rotation (α, β, γ) of the detector array 101 in the optical center coordinate system and the position of the center pixel of the detector array 101 as X0(x(0,0),y(0,0),z(0,0)), are the six parameters that need to be fitted.

[0146] Since the calibration function is a nonlinear function, in some embodiments of the present invention, the Levenberg-Marquardt algorithm is used to solve the problem in the step of fitting the calibration function to address this nonlinear fitting.

[0147] It should be noted that in some embodiments of the present invention, during the process of obtaining the calibration function, the external parameters of the lidar are directly obtained through mechanical measurement. Only the internal functions of the lidar are unknown in the calibration function and need to be fitted. Therefore, the number of calibration receiving operations can be as few as three to complete the fitting. However, as mentioned above, the parameters of the calibration function also include the external parameters of the lidar. Therefore, by increasing the amount of data, the external parameters of the lidar can also be fitted, thereby improving calibration accuracy and reducing random errors caused by measurement.

[0148] Specifically, in some embodiments of the present invention, the step of performing at least 3 calibration receiving operations is performed at least 6 times; the step of fitting the calibration function includes: obtaining the internal parameters and external parameters of the lidar.

[0149] Accordingly, the present invention also provides a calibration system for lidar.

[0150] refer to Figure 7 The diagram shows a schematic representation of an embodiment of the lidar calibration system of the present invention.

[0151] The calibration system includes: a mechanical module 210, which is adapted to fix the lidar 201 to be calibrated and the calibration target 202; the mechanical module 210 performs at least three calibration receiving operations, which are adapted to obtain the positional relationship between the sampling detector and the calibration target 202, and the sampling detector in each calibration receiving operation is a different detector; and a fitting module 220, which is adapted to fit a preset calibration function based on the positional relationship obtained from at least three calibration receiving operations to obtain the internal parameters of the lidar 200, wherein the calibration function is obtained based on the imaging system parameters of the lidar 200 and the detector array parameters of the lidar, combined with the positional information of the calibration target 220.

[0152] The fitting module 220 directly obtains the internal parameters of the lidar by fitting the calibration function. Data acquisition can be achieved with as few as three calibration receiving operations, which can effectively improve calibration speed and efficiency. In particular, it makes it possible to calibrate the internal parameters of the detector array of solid-state lidar.

[0153] The mechanical module 210 is suitable for fixing the lidar 201 to be calibrated and the calibration target 202.

[0154] In some embodiments of the present invention, the receiving end of the lidar 201 includes a detector array. Therefore, the area of ​​the detector array and the number of detectors in the lidar 201 determine the field of view and angular resolution of the lidar. In some embodiments of the present invention, the detector array is an integrated detector array, in which each detector is very small, and the detectors contained in the detector array are arranged in a two-dimensional array, and the number is huge, reaching tens of thousands. It is evident that the technical solution of obtaining all internal parameters of the lidar by fitting a calibration function based on data collected from a finite number of calibration receiving operations makes the calibration of the area array detector possible, effectively improving the measurement accuracy of solid-state lidar.

[0155] In some embodiments of the present invention, the lidar 201 is a solid-state lidar, that is, the lidar 201 does not include any mechanical rotating or scanning components.

[0156] like Figure 7As shown, in some embodiments of the present invention, the mechanical module 210 includes: a turntable having a table surface 210c, the turntable also having a vertical rotating shaft 211 perpendicular to the table surface 210c and a horizontal rotating shaft 212 parallel to the table surface; the lidar 201 is fixed on the table surface 210c of the turntable, the optical axis of the lidar 201 is parallel to the table surface 210c of the turntable and perpendicular to the horizontal rotating shaft 212.

[0157] The turntable's platform 210c rotates around the vertical axis 211 to achieve horizontal rotation, and rotates around the horizontal axis 212 to achieve pitch rotation. The lidar 201 is fixed on the platform 210c, and the optical axis of the lidar 201 is perpendicular to both the vertical axis 211 and the horizontal axis 212; that is, the vertical axis 211, the horizontal axis 212, and the optical axis of the lidar 201 are all orthogonal to each other.

[0158] As can be seen, the orientation of the lidar 201 changes with the translation and pitch of the turntable surface 210. Therefore, the lidar 201 has a first axis of rotation and a second axis of rotation, which are orthogonal to each other as well as to the optical axis of the lidar 201. The vertical axis 111 and the horizontal axis 112 of the turntable coincide with the first axis of rotation and the second axis of rotation of the lidar 100, respectively.

[0159] The turntable's platform 210c is capable of horizontal and vertical rotation. The lidar 201 is fixed to the turntable's platform 210c, and its horizontal and vertical orientation can be adjusted by rotating the turntable. Using a turntable capable of tilting and swerving to adjust the lidar 201's orientation allows for precise acquisition of the adjusted position, improving calibration accuracy. Furthermore, high-precision turntables are readily available and easy to operate, thus reducing calibration difficulty and equipment costs.

[0160] It should be noted that, Figure 7 The schematic turntable 210 has two rotating shafts 211 and 212. In other embodiments of the invention, the horizontal rotating shaft can be... Figure 7 The schematic diagram shows the directions in which the rotating axes 212 are orthogonal. Alternatively, the turntable may have three rotating axes, including a vertical axis and two horizontal axes, the two horizontal axes being orthogonal to each other in the horizontal direction.

[0161] The calibration target 202 is used to reflect the probe light to form an echo signal during the calibration receiving operation, and therefore the calibration target 202 is set within the field of view of the lidar 201.

[0162] like Figure 7As shown, in some embodiments of the present invention, the calibration target 202 is fixed on the extension line of the optical axis of the receiving optical component of the lidar 201 to improve the symmetry of the optical path and reduce the number of coordinate system transformations during the fitting process, thereby simplifying the calculation and reducing the calculation cost.

[0163] In some embodiments of the present invention, the calibration target 202 has a preset size so that only one detector can acquire the echo signal. Controlling the size of the calibration target 202 so that only one detector can acquire the echo signal ensures that only one detector can measure the data in the data acquired after the calibration receiving operation, which is beneficial for improving the data acquisition accuracy during the calibration receiving operation. Specifically, the size of the calibration target 202 is smaller than the angular resolution of the lidar 201. However, using a relatively small calibration target 202 is only one example; in other embodiments of the present invention, a slightly larger calibration target can be selected.

[0164] It should be noted that, as Figure 7 As shown, in some embodiments of the present invention, the calibration target 202 is square; in other embodiments of the present invention, the calibration target may also be circular or other shapes. The present invention does not limit the shape of the calibration target.

[0165] In some embodiments of the present invention, the calibration system further includes a measurement module 230, which is adapted to obtain the coordinates of the calibration target 202 and the coordinates of the lidar 201 in a world coordinate system to obtain the position information of the calibration target 202 relative to the lidar 201. The measurement module 230 is pre-set with a world coordinate system, the origin of which is the intersection of the vertical rotation axis 211 and the horizontal rotation axis 212. One coordinate axis of the world coordinate system is parallel to the vertical rotation axis 211, and the other coordinate axis of the world coordinate system is parallel to the horizontal rotation axis 212.

[0166] It should be noted that in some embodiments of the present invention, the measurement module 230 obtains the position information between the calibration target 202 and the lidar 201. However, this is only an example. In other embodiments of the present invention, the position information of the calibration target 120 relative to the lidar 100 in the world coordinate system can be obtained directly by manual measurement.

[0167] It should also be noted that the vertical rotation axis 211 is parallel to one coordinate axis of the world coordinate system, the horizontal rotation axis 212 is parallel to another coordinate axis of the world coordinate system, and the optical axis of the lidar is orthogonal to both the vertical rotation axis 211 and the horizontal rotation axis 212. Therefore, the optical axis of the lidar is parallel to the third coordinate axis of the world coordinate system.

[0168] Furthermore, the vertical axis 211 and horizontal axis 212 of the turntable coincide with the first axis and second axis of the lidar 201, respectively. Therefore, the far point of the world coordinate system is the intersection of the vertical axis 211 and horizontal axis 212 of the turntable. The two coordinate axes of the world coordinate system are parallel to the vertical axis 211 and horizontal axis 212 of the turntable, respectively. In other words, the world coordinate system can also be regarded as a coordinate system established with the intersection of the vertical axis 211 and the horizontal axis 212 as the origin, and the vertical axis 211 is parallel to one coordinate axis and the horizontal axis 212 is parallel to another coordinate axis.

[0169] Specifically, in the world coordinate system of the measurement module 230, the position information of the calibration target 202 relative to the lidar 201 is P0(A,B,C) of the calibration target.

[0170] It should be noted that after the coordinate system is rotated, the coordinates of the calibration target 202 can be transformed through coordinate transformation. For example, when the world coordinate system is rotated by θ around two coordinate axes (the two coordinate axes correspond to the first rotation axis and the second rotation axis, respectively), the coordinates of the calibration target 202 can be transformed. When the angle is specified, the coordinate transformation of the calibration target 202 is P = R * P0. Where R is the rotation matrix, as follows:

[0171]

[0172] On the other hand, the internal parameters of the lidar refer to the positions of each detector of the lidar in the optical center coordinate system. Therefore, the measurement module 230 further transforms the coordinates of the calibration target 202 to the optical center coordinate system.

[0173] Specifically, the measurement module 230 transforms the coordinates of the calibration target 202 from the world coordinate system to the optical center coordinate system based on the offset and orientation of the optical center coordinate system relative to the world coordinate system. The offset and orientation of the optical center coordinate system relative to the world coordinate system include: translation of the optical center coordinate system relative to the world coordinate system, i.e., the coordinates (a, b, c) between the optical center (i.e., the origin of the optical center coordinate system) and the rotation center (i.e., the origin of the world coordinate system); and rotation of the optical center coordinate system relative to the world coordinate system, i.e., the angles (θx, θy, θz) between the corresponding coordinate axes of the world coordinate system and the optical center coordinate system.

[0174] Therefore, when the world coordinate system rotates by θ around two coordinate axes (the two coordinate axes correspond to the first rotation axis and the second rotation axis respectively), respectively... When the angle is determined, according to the rotation matrix R, the measurement module 230 obtains the coordinates (Px, Py, Pz) of the calibration target 202 in the optical center coordinate system, where the transformation relationship P'0 is as follows:

[0175]

[0176]

[0177] The calibration receiving operation is suitable for obtaining the positional relationship between the sampling detector and the calibration target 202, and the sampling detector in each calibration receiving operation is a different detector.

[0178] The positional relationship between the sampling detector and the calibration target includes: the attitude angle corresponding to the sampling detector, wherein the attitude angle is the angle through which the lidar rotates when the sampling detector receives the echo signal.

[0179] In some embodiments of the present invention, the calibration receiving operation is performed by manual control of the mechanical module. Specifically, refer to the references... Figure 5 The calibration receiving operation includes: first, executing step S121 to determine the sampling detector; then executing step S122 to rotate the lidar so that the sampling detector receives the echo signal reflected by the calibration target; and then executing step S123 to obtain the attitude angle.

[0180] The sampling detector is a detector that receives the echo signal reflected from the calibration target 202 during each calibration receiving operation. The sampling detector determined for each calibration receiving operation is a different detector on the detector array 101.

[0181] like Figure 6As shown, in some embodiments of the present invention, the sampling detector is selected from the apex detector 101a, the midpoint detector 102b, and the center detector 101c of the detector array. The apex detector 101a is the detector located at the apex of the detector array 101, the midpoint detector 101b is the detector located at the midpoint of each side of the detector array 101, and the center detector 101c is the detector located at the center of symmetry of the detector array 101. Selecting the sampling detector within the range of the apex detector 101a, the midpoint detector 102b, and the center detector 101c can improve calibration accuracy. Specifically, referring to the reference... Figure 3 In the step of determining the sampling detector, the position coordinates of the sampling detector in the detector array 101 are (m, n). Therefore, based on the selection range of the sampling detector, in some embodiments of the present invention, the calibration reception operation is performed 4 to 9 times.

[0182] It should be noted that the manual control method for performing the calibration receiving operation is only one example. In other embodiments of the present invention, the calibration system further includes an operation module 240, which is adapted to perform the calibration receiving operation through the mechanical module 210 to obtain the attitude angle.

[0183] Reference Figure 8 , showed Figure 7 The diagram shows the functional block diagram of the operation module in the calibration system embodiment.

[0184] The operation module 240 includes: a sampling unit 241, which is adapted to determine the sampling detector; an operation unit 242, which rotates the lidar through the mechanical module 210 so that the sampling detector receives the echo signal reflected by the calibration target; and a data unit 243, which obtains the attitude angle through the mechanical module 210.

[0185] Specifically, the operation unit 242 is connected to the sampling unit 241 and obtains the determined sampling detector from the sampling unit 241; the operation unit 242 is also connected to the mechanical module 210 and controls the mechanical unit 241 to rotate the lidar, so that the sampling detector receives the echo signal; the data unit 243 is connected to the mechanical module 210 through the operation unit 242 and obtains the attitude angle through the operation unit and the mechanical module.

[0186] It should be noted that in some embodiments of the present invention, the size of the calibration target 202 is small, enabling only one detector to acquire the echo signal; or, the size of the calibration target 202 is smaller than the angular resolution of the lidar 201. Therefore, when the operation unit 242 controls the mechanical unit 241 to rotate the lidar, only one detector can receive the echo signal, so the operation unit 242 only needs to ensure that the sampling detector can receive the signal.

[0187] However, in other embodiments of the present invention, the calibration target is slightly larger, and multiple detectors on the detector array can receive the echo signal, and the operation unit positions the sampling detector at the location where the received echo signal is strongest; in other embodiments, the operation unit positions the sampling detector at the middle position among multiple detectors where the intensity of the received echo signal exceeds a preset threshold.

[0188] It should be noted that in some embodiments of the present invention, only one calibration target is provided. In other embodiments of the present invention, multiple calibration targets are provided, thereby reducing the rotation angle of the lidar during calibration and receiving operations, enabling faster calibration and receiving operations, and improving calibration efficiency.

[0189] Continue to refer to Figure 7 The lidar 201 is fixed on the platform 210c of the turntable in the mechanical module 210. The vertical axis 211 and the horizontal axis 212 of the turntable coincide with the first axis and the second axis of the lidar, respectively. Therefore, the attitude angle includes a first attitude angle and a second attitude angle. The first attitude angle is the angle through which the platform of the turntable rotates about the vertical axis, and the second attitude angle is the angle through which the platform of the turntable rotates about the horizontal axis.

[0190] Obtaining the attitude angle using a turntable effectively ensures the accuracy of the attitude angle, which is beneficial to improving calibration accuracy. Moreover, the turntable can be a high-precision turntable, which is readily available and easy to operate. Therefore, this approach also helps to reduce system costs and operational difficulty.

[0191] Specifically, in the i-th calibration and reception operation, the determined position of the sampling detector is (m i ,n i The attitude angle corresponding to the sampling detector is obtained as follows:

[0192] Continue to refer to Figure 7 The calibration system further includes a fitting module 220.

[0193] The fitting module 220 is adapted to fit the calibration function based on the positional relationship obtained from at least three calibration receiving operations to obtain the internal parameters of the lidar 201.

[0194] The fitting module 220 has a preset calibration function, which is suitable for characterizing the relationship between the internal and external parameters of the lidar, the imaging system parameters and detector array parameters of the lidar, and the position information of the calibration target relative to the lidar.

[0195] In some embodiments of the present invention, the calibration function is obtained based on the pinhole lens imaging principle. Therefore, the calibration receiving operation performed by the mechanical module 210 ensures that the sampling detector, the optical center, and the calibration target 202 are collinear. Obtaining the calibration function based on the pinhole imaging principle appropriately simplifies the optical path structure of the lidar 201, which helps to simplify the calibration function and thus reduce the amount of fitting calculations.

[0196] It should be noted that the calibration function is also related to the structure and parameters of the lidar's imaging system. Specifically, in some embodiments of the present invention, the lidar's imaging system includes a wide-angle lens, such as an f-theta lens.

[0197] The specific technical solution for the calibration function is described in the embodiments of the aforementioned calibration method. Further details are omitted here. Specifically, in the i-th calibration receiving operation, the determined position of the sampling detector is (m... i ,n i The attitude angle corresponding to the sampling detector is obtained as follows: Moreover, during the i-th calibration and reception operation, the lidar 201 is in the i-th attitude.

[0198] The calibration function of the lidar 201 in the i-th attitude is:

[0199]

[0200] Among them, (P) i x ,P i y ,P i z ,) represents the coordinates of the target object used for calibration in the optical center coordinate system at the i-th attitude of the lidar 100, m i Let n be the number of rows of the sampling detector for the lidar 100 in the i-th pose. i Let x(m) be the number of columns of the sampling detectors of the lidar 201 in the i-th pose. i ,n i ),y(mi ,n i ),z(m i ,n i )) represents the coordinates of the sampling detector in the optical center coordinate system of the lidar 201 in the i-th attitude, and

[0201] Therefore, the fitting module 220 inputs the position of the sampling detector in all calibration receiving operations and the attitude angle corresponding to the sampling detector into the above calibration function for fitting, thereby obtaining the internal parameters of the lidar 201.

[0202] Specifically, each pixel in the detector array is a detector. For example... Figure 3 As shown, the position of the center pixel of the detector array 101 is X0(x(0,0),y(0,0),z(0,0)), the spacing vector along the row direction (the horizontal x direction in the figure) is K(Kx,Ky,Kz), and the spacing vector along the column direction (the vertical y direction in the figure) is L(Lx,Ly,Lz).

[0203] Therefore, the coordinates of each detector on each of the detector arrays 101 in the optical center coordinate system are:

[0204] x(m,n)=x(0,0)+m×Kx+n×Lx

[0205] y(m,n)=y(0,0)+m×Ky+n×Ly

[0206] z(m,n)=z(0,0)+m×Kz+n×Lz

[0207] Among them, the interval vector K(Kx,Ky,Kz) and the interval vector L(Lx,Ly,Lz) are one of the parameters of the detector array, and the position and orientation of the detector array 101, including the rotation (α,β,γ) of the detector array 101 in the optical center coordinate system and the position of the center pixel of the detector array 101 as X0(x(0,0),y(0,0),z(0,0)), are the six parameters that need to be fitted.

[0208] As can be seen, the calibration function is a nonlinear function. Therefore, for this nonlinear fitting, in some embodiments of the present invention, the fitting module 220 uses the Levenberg-Marquardt algorithm to solve it.

[0209] It should be noted that in some embodiments of the present invention, the external parameters of the lidar are obtained directly through mechanical measurement. After the fitting module 220 obtains the measured external parameters, when fitting the calibration function, only the internal function of the lidar is unknown and needs to be fitted. Therefore, the number of times the calibration receiving operation is executed can be as few as 3 to complete the fitting.

[0210] However, in other embodiments of the present invention, at least six calibration receiving operations are performed; the fitting module obtains the position information obtained from at least six calibration receiving operations and fits it to obtain the internal and external parameters of the lidar, so as to reduce the random error caused by measurement and improve the calibration accuracy.

[0211] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A calibration method for a lidar, characterized in that, include: Based on the imaging system parameters and detector array parameters of the lidar, and combined with the position information of at least one calibration target relative to the lidar, a calibration function is obtained, wherein the detector array parameters include: the spacing between the detectors in the detector array; At least three calibration receiving operations are performed, the calibration receiving operations being suitable for obtaining the positional relationship between a detector acting as a sampling detector and the calibration target object, and the sampling detector in each calibration receiving operation is a different detector; Based on the positional relationship obtained from at least three calibration reception operations, the calibration function is fitted to obtain the internal parameters of the lidar.

2. The calibration method as described in claim 1, characterized in that, The steps to obtain the calibration function include: Obtain the external parameters of the lidar during the calibration process; Based on the imaging system parameters and detector array parameters of the lidar, combined with the external parameters of the lidar and the position information of the calibration target, the calibration function is obtained.

3. The calibration method as described in claim 1, characterized in that, The steps to obtain the calibration function include: obtaining the calibration function based on the imaging principle of a pinhole lens.

4. The calibration method as described in claim 1 or 3, characterized in that, The calibration function is: ； Where i represents the number of calibration receive operations. For the i-th calibration receiving operation, the coordinates of the target object in the optical center coordinate system are given. Let be the number of rows sampled by the detector during the i-th calibration receive operation. Let be the number of columns of the sampling detector during the i-th calibration receiving operation. Let the coordinates of the downsampling detector in the optical center coordinate system be the coordinates of the i-th time calibration receiving operation, and .

5. The calibration method as described in claim 4, characterized in that, In the step of fitting the calibration function, the Levenberg-Marquardt algorithm is used for solving.

6. The calibration method as described in claim 1, characterized in that, The positional relationship between the sampling detector and the calibration target includes: the attitude angle corresponding to the sampling detector, wherein the attitude angle is the angle through which the lidar rotates when the sampling detector receives the echo signal reflected by the calibration target; The calibration receiving operation includes: determining the sampling detector; rotating the lidar so that the sampling detector receives the echo signal; and obtaining the attitude angle.

7. The calibration method as described in claim 1, characterized in that, In the step of performing at least 3 calibration and reception operations, at least 6 calibration and reception operations are performed. The step of fitting the calibration function includes obtaining the internal and external parameters of the lidar.

8. The calibration method as described in claim 1, characterized in that, The calibration target is fixed on the extension line of the optical axis of the lidar receiving optical component.

9. The calibration method as described in claim 6, characterized in that, In the calibration and receiving operation, the step of rotating the lidar includes: rotating the lidar around a first axis and a second axis respectively, so that the lidar rotates, wherein the first axis and the second axis and the optical axis of the lidar are orthogonal to each other; In the calibration and reception operation, the step of obtaining the attitude angle includes: the attitude angle includes: a first attitude angle and a second attitude angle, wherein the first attitude angle is the angle through which the lidar rotates around the first axis, and the second attitude angle is the angle through which the lidar rotates around the second axis.

10. The calibration method as described in claim 9, characterized in that, Also includes: Before obtaining the calibration function, the position information of the calibration target relative to the lidar is obtained, wherein the step of obtaining the position information of the calibration target relative to the lidar includes: A world coordinate system is established with the intersection of the first rotation axis and the second rotation axis as the origin, and the first rotation axis is parallel to one coordinate axis of the world coordinate system, and the second rotation axis is parallel to the other coordinate axis of the world coordinate system. By obtaining the coordinates of the calibration target and the lidar in the world coordinate system, the position information of the calibration target relative to the lidar can be obtained.

11. The calibration method as described in claim 1, characterized in that, The calibration target has a preset size so that only one detector can collect the echo signal reflected by the calibration target.

12. The calibration method as described in claim 1 or 11, characterized in that, The size of the calibration target is smaller than the angular resolution of the lidar.

13. The calibration method as described in claim 6, characterized in that, In the calibration and reception operation, the step of rotating the lidar so that the sampling detector receives the echo signal is to make the echo signal received by the sampling detector the strongest.

14. The calibration method as described in claim 1, characterized in that, The sampling detector is selected from the apex detector, midpoint detector and center detector of the detector array, wherein the apex detector is the detector located at the apex of the detector array, the midpoint detector is the detector located at the midpoint of each side of the detector array, and the center detector is the detector located at the center of symmetry of the detector array.

15. The calibration method as described in claim 1 or 14, characterized in that, In the step of performing at least 3 calibration and reception operations, the calibration and reception operations are performed 4 to 9 times.

16. A calibration system for a lidar, characterized in that, include: A mechanical module, the mechanical module being adapted to fix the lidar to be calibrated and at least one calibration target; The mechanical module performs at least three calibration receiving operations, which are suitable for obtaining the positional relationship between the sampling detector and the calibration target. The sampling detector used in each calibration receiving operation is a different detector. The fitting module is adapted to fit a preset calibration function based on the positional relationship obtained from at least three calibration receiving operations to obtain the internal parameters of the lidar. The calibration function is obtained based on the imaging system parameters and detector array parameters of the lidar, combined with the positional information of the calibration target relative to the lidar. The detector array parameters include the spacing between the detectors in the detector array.

17. The calibration system as described in claim 16, characterized in that, The calibration function is obtained based on the imaging principle of a pinhole lens.

18. The calibration system as described in claim 16 or 17, characterized in that, The calibration function is: ； Where i represents the number of calibration receive operations. For the i-th calibration receiving operation, the coordinates of the target object in the optical center coordinate system are given. Let be the number of rows sampled by the detector during the i-th calibration receive operation. Let be the number of columns of the sampling detector during the i-th calibration receiving operation. Let the coordinates of the downsampling detector in the optical center coordinate system be the coordinates of the i-th time calibration receiving operation, and .

19. The calibration system as described in claim 18, characterized in that, The fitting module uses the Levenberg-Marquardt algorithm to solve the problem.

20. The calibration system as described in claim 16, characterized in that, The positional relationship between the sampling detector and the calibration target includes: the attitude angle corresponding to the sampling detector, wherein the attitude angle is the angle through which the lidar rotates when the sampling detector receives the echo signal reflected by the calibration target; The calibration system further includes an operation module, which is adapted to perform the calibration receiving operation through the mechanical module to obtain the attitude angle.

21. The calibration system as described in claim 20, characterized in that, The operation module includes: A sampling unit, the sampling unit being adapted to determine a sampling detector; An operation unit, which rotates the lidar via the mechanical module, so that the sampling detector receives the echo signal reflected by the calibration target. The data unit obtains the attitude angle through the mechanical module.

22. The calibration system as described in claim 21, characterized in that, The sampling detector is selected from the apex detector, midpoint detector and center detector of the detector array, wherein the apex detector is the detector located at the apex of the detector array, the midpoint detector is the detector located at the midpoint of each side of the detector array, and the center detector is the detector located at the center of symmetry of the detector array.

23. The calibration system as described in claim 16, characterized in that, Perform at least 6 calibration reception operations; The fitting module obtains the internal and external parameters of the lidar.

24. The calibration system as described in claim 21, characterized in that, The mechanical module includes: a turntable having a table surface, and a vertical axis perpendicular to the table surface and a horizontal axis parallel to the table surface; a lidar fixed to the table surface of the turntable, the optical axis of the lidar being parallel to the table surface of the turntable and perpendicular to the horizontal axis; the attitude angle includes: a first attitude angle and a second attitude angle, the first attitude angle being the angle through which the table surface of the turntable rotates about the vertical axis, and the second attitude angle being the angle through which the table surface of the turntable rotates about the horizontal axis.

25. The calibration system as described in claim 24, characterized in that, Also includes: A measurement module is provided, which is adapted to obtain the coordinates of the calibration target and the lidar in a world coordinate system to obtain the position information of the calibration target relative to the lidar. The measurement module is equipped with a world coordinate system, the origin of which is the intersection of the vertical rotation axis and the horizontal rotation axis. One coordinate axis of the world coordinate system is parallel to the vertical rotation axis, and the other coordinate axis is parallel to the horizontal rotation axis.

26. The calibration system as described in claim 16, characterized in that, The calibration target has a preset size so that only one detector can collect the echo signal reflected by the calibration target.

27. The calibration system as described in claim 16 or 26, characterized in that, The size of the calibration target is smaller than the angular resolution of the lidar.

Images