CN114296056A - Laser radar external parameter calibration method, device, equipment and storage medium - Google Patents
Laser radar external parameter calibration method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114296056A CN114296056A CN202111473725.9A CN202111473725A CN114296056A CN 114296056 A CN114296056 A CN 114296056A CN 202111473725 A CN202111473725 A CN 202111473725A CN 114296056 A CN114296056 A CN 114296056A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- plane
- cloud data
- plane information
- rectangular
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Images
Abstract
The embodiment of the application discloses a laser radar external reference calibration method, a device, equipment and a storage medium, wherein the method comprises the following steps: obtaining first plane information of each plane according to the rectangular targets positioned on a plurality of different planes; acquiring original point cloud data acquired by a laser radar based on a plurality of rectangular targets on different planes; dividing the original point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, wherein each batch of rectangular target point cloud data corresponds to one first plane information; respectively carrying out plane fitting on each batch of rectangular target point cloud data to obtain second plane information of each batch of rectangular target point cloud data; and registering according to the first plane information and the second plane information corresponding to each plane to obtain the external parameter data of the laser radar. By adopting the method, the laser radar is subjected to external reference calibration by utilizing the plane information, so that available sample points of the external reference calibration of the laser radar are increased, and the robustness of the external reference calibration is enhanced.
Description
Technical Field
The present disclosure relates to the field of laser radar technologies, and in particular, to a method, an apparatus, a device, and a storage medium for calibrating external parameters of a laser radar.
Background
The traditional laser radar external reference calibration scheme is to perform positioning and calibration by identifying geometric information of a target object, and needs to perform fitting aiming at pattern information detected by the laser radar. And the external reference calibration of the laser radar is completed only by extracting and fitting the graph, so that the error in space is very large, and the calibration is easily misaligned. Therefore, in the prior art, the accuracy of the laser radar external reference calibration only depending on the extraction and fitting of the graph is not high.
Disclosure of Invention
The invention provides a laser radar external parameter calibration method, a device, equipment and a storage medium, which aim to solve the problems.
In a first aspect, an embodiment of the present application provides a laser radar external reference calibration method, including: obtaining first plane information of each plane according to the rectangular targets positioned on a plurality of different planes; acquiring original point cloud data acquired by the laser radar based on the rectangular targets of the plurality of different planes; dividing the original point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, wherein each batch of rectangular target point cloud data corresponds to one first plane information; respectively carrying out plane fitting on each batch of rectangular target point cloud data to obtain second plane information of each batch of rectangular target point cloud data; and registering according to the first plane information and the second plane information corresponding to each plane to obtain the external parameter data of the laser radar.
In a second aspect, an embodiment of the present application further provides a laser radar external reference calibration apparatus, where the apparatus includes: the device comprises a first plane acquisition unit, an original data acquisition unit, a target data acquisition unit, a second plane acquisition unit and an external parameter calibration unit. The first plane acquisition unit is used for acquiring first plane information of each plane according to the rectangular targets positioned on a plurality of different planes; the original data acquisition unit is used for acquiring original point cloud data acquired by the laser radar based on the rectangular targets of the plurality of different planes; the target data acquisition unit is used for segmenting the original point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, and each batch of rectangular target point cloud data corresponds to one first plane information; the second plane acquisition unit is used for respectively carrying out plane fitting on each batch of rectangular target point cloud data to acquire second plane information of each batch of rectangular target point cloud data; and the external parameter calibration unit is used for carrying out registration according to the first plane information and the second plane information corresponding to each plane to obtain external parameter data of the laser radar.
In a third aspect, an embodiment of the present application further provides an electronic device, where the electronic device includes: one or more processors, memory, lidar, and one or more applications. Wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the method of the first aspect as described above.
In a fourth aspect, embodiments of the present application further provide a computer-readable storage medium, in which a program code is stored, where the program code is called by a processor to execute the method according to the first aspect.
According to the technical scheme, first plane information of each plane is obtained according to the rectangular targets located on a plurality of different planes; acquiring original point cloud data acquired by the laser radar based on the rectangular targets of the plurality of different planes; dividing the original point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, wherein each batch of rectangular target point cloud data corresponds to one first plane information; respectively carrying out plane fitting on each batch of rectangular target point cloud data to obtain second plane information of each batch of rectangular target point cloud data; and registering according to the first plane information and the second plane information corresponding to each plane to obtain the external parameter data of the laser radar. Therefore, by adopting the method, the external reference calibration of the laser radar is realized by detecting and registering the plane, so that the available samples are increased when the external reference calibration of the laser radar is carried out, and the robustness of the external reference calibration of the laser radar is enhanced.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 shows a schematic flow chart of a laser radar external reference calibration method according to an embodiment of the present application;
fig. 2 shows a block diagram of a lidar external reference calibration apparatus according to an embodiment of the present disclosure;
fig. 3 is a block diagram illustrating an electronic device according to an embodiment of the present application;
fig. 4 shows a block diagram of a computer storage medium according to an embodiment of the present application.
Detailed Description
In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In recent years, the continuous progress of laser radar technology provides a good sensing foundation for unmanned automobiles and intelligent mobile robots. The advantages of the laser radar are: the distance between the radar main body and an external environment, an obstacle and the like can be accurately measured, and the output point cloud can describe a three-dimensional space environment and can be visually displayed; the normal and efficient operation of the equipment cannot be influenced by the laser radar due to the change of the illumination intensity and the dynamic environment.
The external reference calibration of the laser radar is relative transformation relation of a laser radar measurement coordinate system relative to other sensor measurement coordinate systems, so that specific position information of the laser radar is obtained. The traditional laser radar external reference calibration scheme is to perform positioning and calibration by identifying geometric information (such as circle center, square edge, square center and the like) of a target object, and needs to perform fitting aiming at pattern information detected by the laser radar. And the external reference calibration of the laser radar is completed only by extracting and fitting the graph, so that the error in space is very large, and the calibration is easily misaligned. Therefore, in the prior art, the accuracy of the laser radar external reference calibration only depending on the extraction and fitting of the graph is not high.
In order to alleviate the above problems, the inventors of the present application propose a method, an apparatus, a device, and a storage medium for calibrating an external parameter of a laser radar provided by the present application. Obtaining first plane information of each plane according to the rectangular targets located in a plurality of different planes; acquiring original point cloud data acquired by the laser radar based on the rectangular targets of the plurality of different planes; dividing the original point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, wherein each batch of rectangular target point cloud data corresponds to one first plane information; respectively carrying out plane fitting on each batch of rectangular target point cloud data to obtain second plane information of each batch of rectangular target point cloud data; and registering according to the first plane information and the second plane information corresponding to each plane to obtain the external parameter data of the laser radar. Therefore, by adopting the method, the external reference calibration is carried out on the laser radar by utilizing the plane information, so that the available sample points for the external reference calibration of the laser radar are increased, and the robustness of the external reference calibration of the laser radar is enhanced.
Embodiments of the present application will be described in detail below with reference to the accompanying drawings.
Referring to fig. 1, an embodiment of the present application provides a laser radar external reference calibration method, which is applicable to a terminal device having a laser radar, and the present embodiment describes a process of steps at a terminal device side, where the method may include steps S110 to S150.
Step S110: first plane information of each plane is obtained from a rectangular target located in a plurality of different planes.
In the embodiment of the application, the rectangular targets located in a plurality of different planes can be simultaneously arranged in a randomly designed scene. The rectangular targets located on the plurality of different planes have the same reflectivity. Specifically, the reflectance of the rectangular targets located on the plurality of different planes is the same, and the material of the rectangular targets located on the plurality of different planes may be the same, or the paint of the rectangular targets located on the plurality of different planes may be the same. Illustratively, the rectangular targets located in the plurality of different planes are 90% diffusely reflective.
When the rectangular targets with the same reflectivity are used as the targets for calibrating the external parameters of the laser radar, the laser radar can measure objects on all sides of the laser radar, so that the rectangular targets on a plurality of different planes are used as the targets for calibrating the external parameters of the laser radar. The laser radar transmits detection signals (laser beams) to the rectangular targets on the different planes, then receives the signals reflected back from the rectangular targets on the different planes, and the distances between the rectangular targets on the different planes and the laser radar and the materials and physical properties of the rectangular targets on the different planes can be analyzed according to the reflected signals, wherein the physical properties can comprise the reflectivity of the rectangular targets on the different planes.
The first plane information of each plane is obtained according to the rectangular targets located on the plurality of different planes, the first plane information of each rectangular target can be obtained by selecting at least three points which are not located on the same straight line from the rectangular targets and fitting the first plane information of each rectangular target, and the first plane information of each rectangular target can also be obtained by fitting the coordinates of four corner points of each rectangular target in the rectangular targets located on the plurality of different planes in a world coordinate system. The first plane information may specifically include a plane true value, that is, a value of each parameter in a plane expression of the plane in the world coordinate system, which is obtained by using four corner points. Taking the general expression Ax + By + Cz + D of the plane equation as 0 as an example, the first plane information includes a first plane equation a1x+B1y+C1z+D10 in A1，B1，C1，D1The values of these four data.
Three points can determine a plane, but when the surface of the rectangular target is not flat or the rectangular target is not a vertical plane, the error of the plane of the rectangular target determined by the three points is large. Therefore, four or more points on each rectangular target are selected to determine the first planar information of each rectangular target. For example, when a plane is determined by four points, a plane can be found in space so that the four points are almost all on the plane, e.g., the plane where the four points are located can be determined according to the minimum sum of the distances from the four points to the plane in space.
Specifically, coordinates of four corner points of each rectangular target in a plurality of rectangular targets on different planes in a world coordinate system are obtained; and fitting and calculating the four corner points of each rectangular target by using a least square method, and minimizing the error of the first plane information of each rectangular target fitted by using the four corner points of each rectangular target by minimizing the sum of squares of distances from the four corner points of each rectangular target to the point plane of the surface fitted by using the four corner points.
In some embodiments, the first plane information of each plane is obtained according to the rectangle targets located in a plurality of different planes, and the first plane information of each rectangle target can also be fitted according to coordinates of four corner points of each rectangle target in a customized coordinate system in the rectangle targets located in a plurality of different planes.
Step S120: and acquiring original point cloud data acquired by the laser radar based on the rectangular targets of the plurality of different planes.
In the embodiment of the application, the specific position of the laser radar is unknown, but the prior position information of the laser radar can be obtained according to the terminal equipment for installing the laser radar, namely the position information of the terminal equipment for installing the laser radar, the laser radar with the known prior position information is utilized to scan the rectangular targets on a plurality of different planes, and the original point cloud data collected by the laser radar based on the rectangular targets on the plurality of different planes is obtained. For example, obtaining raw point cloud data collected by a laser radar mounted on a vehicle based on the plurality of different planes of rectangular targets, obtaining raw point cloud data collected by a laser radar mounted on a robot based on the plurality of different planes of rectangular targets, obtaining raw point cloud data collected by a laser radar mounted on an aircraft based on the plurality of different planes of rectangular targets, and the like.
Step S130: and segmenting the original point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, wherein each batch of rectangular target point cloud data corresponds to one first plane information.
In the embodiment of the application, the reflectivity of the rectangular targets located on a plurality of different planes is the same, the original point cloud data is segmented according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, and the target point cloud data can be obtained by screening the original point cloud data according to the reflectivity; and segmenting the target point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, wherein each batch of rectangular target point cloud data corresponds to one first plane information.
Specifically, the working principle of the laser radar is as follows: after a detection signal (laser beam) is transmitted to a target through the laser radar, a signal reflected from the target is received, the detection signal transmitted by the laser radar is compared with the reflected signal, and after appropriate processing is carried out, relevant information of the target, such as parameters of target distance, direction, height, speed, attitude, reflectivity and the like, can be obtained. The laser radar comprises a plurality of rectangular targets, wherein the plurality of rectangular targets are arranged on the same plane, the original point cloud data acquired by the laser radar based on the plurality of different planes comprises point cloud data of the rectangular targets and point cloud data of other objects, target point cloud data corresponding to the rectangular targets are required to be extracted from the original point cloud data, the reflectivity of the rectangular targets can be used as a support for extracting the target point cloud data, the point cloud data of the reflectivity corresponding to the rectangular targets are screened out, and then the target point cloud data of the rectangular targets are obtained, namely the target point cloud data with the reflectivity identical to that of the rectangular targets are screened out from the original point cloud data.
In some embodiments, since the signal reflected by the lidar may change due to other factors, there may be a certain error in screening the target point cloud data in the original point cloud data by using the fixed reflectivity. Therefore, the target point cloud data in the original point cloud data can be screened according to the reflectivity in a certain threshold range. The threshold range may be obtained through third-party experimental data, for example, if rectangular targets located in a plurality of different planes are 90% diffuse reflection materials, original point cloud data acquired by the laser radar based on the rectangular targets in the plurality of different planes are obtained, the original point cloud data is screened according to the reflectivity to obtain target point cloud data, the original point cloud data may be screened to leave point cloud data with reflectivity of 90% as target point cloud data, or the original point cloud data may be screened to leave point cloud data with reflectivity of 85% to 95% as target point cloud data.
Specifically, the target point cloud data may be filtered in a traversal manner, and the original point cloud data is filtered according to the reflectivity from the original point cloud data acquired by the laser radar based on the plurality of rectangular targets in different planes to obtain the target point cloud data.
Illustratively, the original point cloud data is subjected to traversal clustering filtration based on the characteristic that the reflectivity of the clustered point cloud data is 90%, and then the original point cloud data with the reflectivity of 90% is screened out to serve as target point cloud data.
In some embodiments, the target point cloud data is segmented according to each piece of the first plane information to obtain a plurality of batches of rectangular target point cloud data, each batch of rectangular target point cloud data corresponds to one piece of the first plane information, and the target point cloud data may be segmented according to each piece of the first plane information to obtain a target point cloud data set corresponding to each piece of the first plane information; and deleting outliers in the target point cloud data set corresponding to each piece of first plane information to obtain a batch of rectangular target point cloud data corresponding to each piece of first plane information.
The target point cloud data is segmented according to the first plane information to obtain a target point cloud data set corresponding to the first plane information, the first plane information of each rectangular target can be fitted according to coordinates of four fixed points of each rectangular target in a plurality of rectangular targets located on different planes in a world coordinate system, the target point cloud data corresponding to the rectangular targets of the planes is estimated according to the approximate position of a laser radar and the first plane information of each rectangular target, and the target point cloud data is segmented according to the first plane information of each rectangular target. Specifically, the target point cloud data may be segmented by roughly estimating positions of the rectangular targets located on a plurality of different planes in the field of view of the laser radar according to the determined plane information of each rectangular target, and segmenting the target point cloud data according to the estimated positions by using a RanSaC algorithm, an euclidean algorithm, an DoN algorithm, a super-volume clustering segmentation method, or the like to obtain a target point cloud data set corresponding to each piece of first plane information. For example, euclidean algorithm is to segment point cloud data by distance.
The method for removing outliers in the target point cloud data set corresponding to each piece of first plane information includes removing outliers in the target point cloud data set corresponding to each piece of first plane information to obtain a batch of rectangular target point cloud data corresponding to each piece of first plane information, eliminating the outliers in the target point cloud data set corresponding to each piece of first plane information by utilizing the outlier property of the points, filtering the outliers by adopting a distance filtering method, and removing the outliers by utilizing methods such as a PCL (linear Carrier computation) algorithm, an MAD (maximum intensity distribution) algorithm, a Gaussian distribution method, a standard deviation method or a percentile method. For example, there are many methods for point cloud denoising in PCL, wherein one of the statistical methods is to eliminate outliers by using a function statistical outlierremove (filter removes outliers), and the principle is to calculate the distance distribution from each point to the nearby points in the input data to obtain the average distance from each point to all the nearby points; assuming that the result is a gaussian distribution with a shape determined by the mean and standard deviation, the mean distance outside the standard range is defined as the outlier and removed from the data set.
Step S140: and respectively carrying out plane fitting on each batch of rectangular target point cloud data to obtain second plane information of each batch of rectangular target point cloud data.
In the embodiment of the present application, the plane fitting is performed on each batch of rectangular target point cloud data, and the second plane information obtained from each batch of rectangular target point cloud data may be the rectangular target point cloud number corresponding to the first plane informationAnd according to the sampling consistency plane fitting scheme, obtaining robust plane fitting data, and thus obtaining second plane information of each batch of rectangular target point cloud data. The second plane information may specifically include a plane detection value, that is, a value of each parameter in a plane expression of a plane in the world coordinate system, which is obtained by performing sampling consistency plane fitting using rectangular target point cloud data. The expression of the plane corresponding to the second plane information is A2x+B2y+C2z+D2For example, if 0, the second plane information includes a2，B2，C2，D2The values of these four data.
Specifically, the fitting is to connect a series of points on a plane with a smooth curve. The sampling consistency plane fitting scheme may be to perform plane fitting on the rectangular target point cloud data corresponding to the first plane information By using a method such as a ranac algorithm, an IPC algorithm, or a least square method, so as to obtain robust plane fitting data, where the plane fitting data may be an intercept equation of a plane (x/a + y/B + z/C is a value of three data, i.e., a, B, and C, in 1), may be a general equation of a plane (Ax + By + Cz + D is a value of 4 data, i.e., a, B, C, and D, in 0), or may be a normal equation of a plane (xcos α + ycos β + zcos γ is a value of four data, i.e., cos α, cos β, cos γ, and p, in p). If N points in the point cloud data are subjected to plane fitting by using a normal method, if the normal of the point cloud data is obtained and outliers of the point cloud data are eliminated, calculating the centroid P of the point cloud data, the mean value of the minimum variance of the point cloud data and the direction N of the normal of the point cloud data; describing a three-dimensional plane by using a point method, or calculating a normal line type xcos α + ycos β + zcos γ of the plane as the value of four data of cos α, cos β, cos γ, P in P according to the centroid P and the normal direction N.
It should be understood that sampling consistency plane fitting is performed on the rectangular target point cloud data corresponding to the first plane information, that is, the sum of point-to-plane distances from the rectangular target point cloud data corresponding to the first plane information to a certain plane distance is minimized.
Step S150: and registering according to the first plane information and the second plane information corresponding to each plane to obtain the external parameter data of the laser radar.
In the embodiment of the present application, the registration is performed according to the first plane information and the second plane information corresponding to each plane, and the external parameter data of the laser radar is obtained, where the first plane information and the second plane information corresponding to each plane are registered, so as to obtain a target rotation angle and a target translation distance at which the plane corresponding to the second plane information coincides with the plane corresponding to the first plane information; and obtaining three-axis rotation angle information of the laser radar under a world coordinate system according to the first plane information, the target rotation angle and the target translation distance corresponding to each plane.
And performing plane fitting on first plane information of each target in the plurality of rectangular target points in different planes, which is obtained according to the rectangular targets positioned on the different planes, and each batch of rectangular target point cloud data respectively, and calibrating second plane information of each batch of rectangular target point cloud data by adopting plane point cloud registration, and simultaneously minimizing a loss function for external parameter calibration of the laser radar, so as to obtain a target rotation angle and a target translation distance, wherein the plane corresponding to the second plane information is overlapped with the plane corresponding to the first plane information.
Specifically, the loss function may be composed of two parts, where the first part is a point-to-surface distance from the target point cloud data to the first plane information after the target point cloud data passes through a rotation angle of a rotating target or a translation distance of a translating target; and the second part is the size of an included angle between the second plane information of each batch of rectangular target point cloud data and one first plane information corresponding to each batch of rectangular target point cloud data after the second plane information passes through a rotating target rotating angle or a translating target translating distance. The loss function of external parameter calibration of the laser radar is minimized, even if the sum of the point-to-surface distances of the target point cloud data and the first plane information is minimized after the target point cloud data is rotated by a target rotation angle or translated by a target translation distance, and the surface-to-surface included angle sum of the second plane information corresponding to each plane and the first plane information corresponding to the plane is minimized after the second plane information corresponding to each plane is rotated by the target rotation angle or translated by the target translation distance.
As an embodiment, the first plane information and the second plane information corresponding to each plane are registered, where the second plane information of each batch of rectangular target point cloud data is respectively rotated by a certain angle or translated by a certain distance, so that the second plane information of each batch of rectangular target point cloud data is respectively overlapped with the first plane information corresponding to each batch of rectangular target point cloud data, and the first plane information corresponding to each plane is registered with the second plane information. The target rotation angle and the target translation distance at which the second plane information corresponding to each plane coincides with the first plane information corresponding to the plane are obtained, the target rotation angle may be obtained by obtaining a mode, a median, an average, a geometric mean, or the like of a rotation angle at which the second plane information corresponding to each plane coincides with the first plane information corresponding to the plane is rotated, or the target translation distance may be obtained by obtaining a mode, a median, a mean, a geometric mean, or the like of a translation distance at which the second plane information corresponding to each plane coincides with the first plane information corresponding to the plane is translated, respectively.
As another embodiment, the first plane information and the second plane information corresponding to each plane may be registered by using, as plane truth values, the first plane information of a plurality of planes obtained from a plurality of rectangular targets located on a plurality of different planes, registering the second plane information obtained by dividing the target point cloud data as plane detection values, and rotating or translating the plane detection values by a certain angle or a certain distance by using the plane point cloud registration to minimize an included angle between the plane detection values rotated by a certain angle or translated by a certain distance and the plane truth values or maximize an area of coincidence. And the translation distance which is the smallest included angle or the largest overlapped area between the translated plane detection value and the plane truth value is the target translation distance.
As another embodiment, the first plane information and the second plane information corresponding to each plane are registered, where the second plane information of each batch of rectangular target point cloud data is respectively rotated by a certain angle or translated by a certain distance, so that the second plane information of each batch of rectangular target point cloud data is respectively overlapped with the first plane information corresponding to each batch of rectangular target point cloud data, and the first plane information corresponding to each plane is registered with the second plane information. The target rotation angle and the target translation distance at which the second plane information corresponding to each plane is overlapped with the first plane information corresponding to the plane are obtained, wherein when the second plane information corresponding to one plane is randomly selected to be overlapped with the first plane information corresponding to the plane, the rotation angle of the second plane information is taken as the target rotation angle; and randomly selecting the translation distance of the second plane information as the target translation distance when the second plane information corresponding to a plane is superposed with the first plane information corresponding to the plane.
In this embodiment of the application, the three-axis rotation angle information of the laser radar in the world coordinate system is obtained according to the first plane information, the target rotation angle, and the target translation distance corresponding to each plane. Wherein, triaxial rotation angle information under the world coordinate system includes: yaw, the angle of rotation along the Z-axis of the world coordinate system, i.e. around the Z-axis; pitch, the angle of rotation along the Y axis of its own coordinate system (the coordinate system forward of the X axis), i.e. around the Y axis; roll, the angle of rotation along the X-axis of its own coordinate system (the coordinate system in front of the X-axis), i.e. around the X-axis. The three-axis rotation angle information in the world coordinate system can reflect the relation between the coordinate system where the laser radar is located and the world coordinate system (ground coordinate system), and can also reflect the posture of the laser radar relative to the ground. And determining a yaw angle, a pitch angle and a roll angle of the laser radar in the world coordinate system according to the world coordinate system in which the first plane information corresponding to each plane is located, the target rotation angle and the target translation, namely external reference data of the laser radar in the world coordinate system.
Illustratively, the three-axis rotation angle information of the laser radar in the world coordinate system is obtained according to the first plane information, the target rotation angle and the target translation distance corresponding to each plane, which may be according toAnd the three-axis rotation angle information of the laser radar under the world coordinate system is obtained by the plane true value included in the first plane information and the plane detection value included in the second plane information. Wherein the plane truth value included in the first plane information includes a first plane equation A1x+B1y+C1z+D10 in A1，B1，C1，D1The values of the four data, the second plane information includes a plane detection value including a second plane equation A2x+B2y+C2z+D20 in A2，B2，C2，D2The target rotation angle is obtained by calculating the included angle between the plane corresponding to the first plane equation and the plane corresponding to the second plane equation according to the values of the four data, and the target translation distance is obtained by calculating the distance between the plane corresponding to the first plane equation and the plane corresponding to the second plane equation; and calibrating the yaw angle, the pitch angle and the roll angle of the laser radar in the world coordinate system according to the target rotation angle, the target translation distance and the world coordinate system in which the first plane information is located.
According to the technical scheme, first plane information of each plane is obtained according to the rectangular targets located on a plurality of different planes; acquiring original point cloud data acquired by the laser radar based on the rectangular targets of the plurality of different planes; dividing the original point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, wherein each batch of rectangular target point cloud data corresponds to one first plane information; respectively carrying out plane fitting on each batch of rectangular target point cloud data to obtain second plane information of each batch of rectangular target point cloud data; and registering according to the first plane information and the second plane information corresponding to each plane to obtain the external parameter data of the laser radar. Therefore, by adopting the method, the external parameter of the laser radar can be calibrated only by utilizing the plane information, the extraction and fitting of the graph are not relied on, the calibration error is reduced, and meanwhile, the robustness of the external parameter calibration of the laser radar is enhanced under the condition that the available sample points for calibrating the external parameter of the laser radar are increased.
Referring to fig. 2, a block diagram of a lidar external reference calibration apparatus according to an embodiment of the present invention is shown, where the apparatus 200 includes: a first plane acquisition unit 210, a raw data acquisition unit 220, a target data acquisition unit 230, a second plane acquisition unit 240, and an external reference calibration unit 250. Specifically, the first plane obtaining unit 210 is configured to obtain first plane information of each plane according to rectangular targets located in a plurality of different planes; an original data obtaining unit 220, configured to obtain original point cloud data acquired by the laser radar based on the plurality of rectangular targets in different planes; a target data obtaining unit 230, configured to segment the original point cloud data according to each piece of the first plane information to obtain multiple batches of rectangular target point cloud data, where each batch of rectangular target point cloud data corresponds to one piece of first plane information; a second plane obtaining unit 240, configured to perform plane fitting on each batch of rectangular target point cloud data respectively to obtain second plane information of each batch of rectangular target point cloud data; and the external parameter calibration unit 250 is configured to perform registration according to the first plane information and the second plane information corresponding to each plane, so as to obtain external parameter data of the laser radar.
As an embodiment, the first plane obtaining unit 210 performs fitting calculation on four corner points of each rectangular target by using a least square method according to coordinates of the four corner points of each rectangular target in the world coordinate system, located in a plurality of rectangular targets with the same reflectivity on different planes, to obtain first plane information of each rectangular target. The raw data acquiring unit 220 includes a laser radar, and the raw data acquiring unit 220 acquires raw point cloud data acquired by the laser radar based on the plurality of rectangular targets in different planes. The target data acquisition unit 230 screens out target point cloud data having the same reflectivity as that of the rectangular target from the original point cloud data; dividing the target point cloud data according to the first plane information to obtain a target point cloud data set corresponding to the first plane information; and deleting outliers in the target point cloud data set corresponding to each piece of first plane information to obtain a batch of rectangular target point cloud data corresponding to each piece of first plane information. The second plane obtaining unit 230 obtains second plane information of each batch of rectangular target point cloud data by respectively adopting a plane fitting scheme for each batch of rectangular target point cloud data. The external reference calibration unit 250 registers the first plane information and the second plane information corresponding to each plane to obtain a target rotation angle and a target translation distance of the plane corresponding to the second plane information and the plane corresponding to the first plane information; and obtaining three-axis rotation angle information of the laser radar under a world coordinate system according to the first plane information, the target rotation angle and the target translation distance corresponding to each plane.
The technical scheme that this application provided only needs to utilize plane information can carry out the demarcation of laser radar external reference to all points in the plane information can all regard as the sample point of demarcation, compare the pattern and draw final only a central point a point and utilize and mark laser radar, can reduce the error to laser radar external reference demarcation, make the robustness reinforcing to laser radar's external reference demarcation simultaneously.
It should be noted that, the embodiments of the present disclosure are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. For the device-like embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment. For any processing manner described in the method embodiment, all the processing manners may be implemented by corresponding processing modules in the apparatus embodiment, and details in the apparatus embodiment are not described again.
Referring to fig. 3, based on the above-mentioned lidar external reference calibration method, another electronic device 300 including a processor capable of executing the above-mentioned lidar external reference calibration method is provided, where the electronic device 300 further includes one or more processors 310, a memory 320, a lidar 330, and one or more application programs. The memory 320 stores a program that can execute the content in the foregoing embodiments, the laser radar 330 can collect point cloud data in the foregoing embodiments, and the processor 310 can execute the program stored in the memory 320. The electronic device 300 may be a vehicle, an intelligent robot, an aircraft, or the like.
Processor 310 may include, among other things, one or more cores for processing data and a message matrix unit. The processor 310 connects various parts within the overall electronic device using various interfaces and lines, and performs various functions of the electronic device 300 and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 320 and calling data stored in the memory 320. Alternatively, the processor 310 may be implemented in hardware using at least one of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA). The processor 310 may integrate one or more of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a modem, and the like. Wherein, the CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing display content; the modem is used to handle wireless communications. It is to be understood that the modem may be implemented by a communication chip without being integrated into the processor.
The Memory 320 may include a Random Access Memory (RAM) or a Read-Only Memory (Read-Only Memory). The memory 320 may be used to store instructions, programs, code sets, or instruction sets. The memory 320 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (e.g., obtaining first plane information for each plane based on a rectangular target located in a plurality of different planes), instructions for implementing various method embodiments described below, and the like. The storage data area may also store data created by the terminal in use (such as first plane information for each plane, original point cloud data, target point cloud data set), and the like.
The laser radar 330 may be composed of a laser transmitter, an optical receiver, a turntable, and an information processing system, and is a radar system that detects characteristic quantities such as a position, a speed, and the like of an object by transmitting a laser beam. The working principle is that a detection signal (laser beam) is emitted to a target, then a received signal (target echo) reflected from the target is compared with the emitted signal, and after appropriate processing, relevant information of the target, such as target distance, azimuth, height, speed, attitude, even shape and other parameters, can be obtained, so that the targets of airplanes, missiles and the like are detected, tracked and identified. The laser changes the electric pulse into a light pulse to be emitted, and the light receiver restores the light pulse reflected from the target into the electric pulse.
Referring to fig. 4, a block diagram of a computer-readable storage medium 400 according to an embodiment of the present application is shown. The computer-readable storage medium 400 has stored therein a program code 410, said program code 410 being invokable by the processor for performing the method described in the above-described method embodiments.
The computer-readable storage medium 400 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. Alternatively, the computer-readable storage medium 400 includes a non-volatile computer-readable storage medium. The computer readable storage medium 400 has storage space for program code 410 for performing any of the method steps of the method described above. The program code 410 can be read from or written to one or more computer program products. Program code 410 may be compressed, for example, in a suitable form.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not necessarily depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (10)
1. A laser radar external reference calibration method is characterized by comprising the following steps:
obtaining first plane information of each plane according to the rectangular targets positioned on a plurality of different planes;
acquiring original point cloud data acquired by the laser radar based on the rectangular targets of the plurality of different planes;
dividing the original point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, wherein each batch of rectangular target point cloud data corresponds to one first plane information;
respectively carrying out plane fitting on each batch of rectangular target point cloud data to obtain second plane information of each batch of rectangular target point cloud data;
and registering according to the first plane information and the second plane information corresponding to each plane to obtain the external parameter data of the laser radar.
2. The lidar external reference calibration method according to claim 1, wherein reflectivity of rectangular targets located on a plurality of different planes is the same, the segmenting the original point cloud data according to each first plane information to obtain a plurality of batches of rectangular target point cloud data, each batch of rectangular target point cloud data corresponding to one first plane information comprises:
screening the original point cloud data according to the reflectivity to obtain target point cloud data;
and segmenting the target point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, wherein each batch of rectangular target point cloud data corresponds to one first plane information.
3. The lidar external reference calibration method according to claim 2, wherein the dividing the target point cloud data according to each of the first plane information to obtain a plurality of batches of rectangular target point cloud data, each batch of rectangular target point cloud data corresponding to one first plane information comprises:
dividing the target point cloud data according to the first plane information to obtain a target point cloud data set corresponding to the first plane information;
and deleting outliers in the target point cloud data set corresponding to each piece of first plane information to obtain a batch of rectangular target point cloud data corresponding to each piece of first plane information.
4. The lidar external reference calibration method according to claim 2, wherein the screening the original point cloud data according to the reflectivity to obtain target point cloud data comprises:
and screening target point cloud data with the same reflectivity as that of the rectangular target from the original point cloud data.
5. The lidar external reference calibration method according to claim 1, wherein the obtaining first plane information of each plane based on the rectangular target located in a plurality of different planes comprises:
and fitting first plane information of each rectangular target according to coordinates of four corner points of each rectangular target in the plurality of rectangular targets on different planes in the world coordinate system.
6. The lidar external reference calibration method according to claim 5, wherein the fitting of the first plane information of each rectangular target according to the coordinates of the four corner points of each rectangular target in the world coordinate system comprises:
acquiring coordinates of four corner points of each rectangular target in a plurality of rectangular targets on different planes in a world coordinate system;
and respectively performing fitting calculation on four corner points of each rectangular target by using a least square method to obtain first plane information of each rectangular target.
7. The lidar external reference calibration method according to claim 6, wherein the registering according to the first plane information and the second plane information corresponding to each plane to obtain the external reference data of the lidar comprises:
registering first plane information and second plane information corresponding to each plane to obtain a target rotation angle and a target translation distance of a plane corresponding to the second plane information and a plane corresponding to the first plane information;
and obtaining three-axis rotation angle information of the laser radar under a world coordinate system according to the first plane information, the target rotation angle and the target translation distance corresponding to each plane.
8. A laser radar external reference calibration device is characterized by comprising:
the first plane acquisition unit is used for acquiring first plane information of each plane according to the rectangular targets positioned on a plurality of different planes;
the original data acquisition unit is used for acquiring original point cloud data acquired by the laser radar based on the rectangular targets of the plurality of different planes;
the target data acquisition unit is used for segmenting the original point cloud data according to the first plane information to obtain a plurality of batches of rectangular target point cloud data, and each batch of rectangular target point cloud data corresponds to one first plane information;
the second plane acquisition unit is used for respectively carrying out plane fitting on each batch of rectangular target point cloud data to acquire second plane information of each batch of rectangular target point cloud data;
and the external parameter calibration unit is used for carrying out registration according to the first plane information and the second plane information corresponding to each plane to obtain external parameter data of the laser radar.
9. An electronic device, comprising:
one or more processors;
a memory;
a laser radar;
one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the method of any of claims 1-7.
10. A computer-readable storage medium, having stored thereon program code that can be invoked by a processor to perform the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111473725.9A CN114296056A (en) | 2021-12-02 | 2021-12-02 | Laser radar external parameter calibration method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111473725.9A CN114296056A (en) | 2021-12-02 | 2021-12-02 | Laser radar external parameter calibration method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114296056A true CN114296056A (en) | 2022-04-08 |
Family
ID=80965823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111473725.9A Pending CN114296056A (en) | 2021-12-02 | 2021-12-02 | Laser radar external parameter calibration method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114296056A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114994648A (en) * | 2022-08-05 | 2022-09-02 | 聚时科技(深圳)有限公司 | External parameter calibration method for 2D laser radar on linear motion mechanism |
| CN115002906A (en) * | 2022-08-05 | 2022-09-02 | 中昊芯英(杭州)科技有限公司 | Object positioning method, device, medium and computing equipment |
| CN115032615A (en) * | 2022-05-31 | 2022-09-09 | 中国第一汽车股份有限公司 | Laser radar calibration point determining method, device, equipment and storage medium |
| CN115661220A (en) * | 2022-12-28 | 2023-01-31 | 深圳煜炜光学科技有限公司 | Point cloud data registration method, device, equipment and storage medium |
| CN116400334A (en) * | 2023-06-01 | 2023-07-07 | 未来机器人（深圳）有限公司 | Calibration verification method and device for laser external parameters, electronic equipment and storable medium |
| CN116878402A (en) * | 2023-07-11 | 2023-10-13 | 北京博科测试系统股份有限公司 | Non-contact wheel arch measuring sensor and method |
-
2021
- 2021-12-02 CN CN202111473725.9A patent/CN114296056A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115032615A (en) * | 2022-05-31 | 2022-09-09 | 中国第一汽车股份有限公司 | Laser radar calibration point determining method, device, equipment and storage medium |
| CN114994648A (en) * | 2022-08-05 | 2022-09-02 | 聚时科技(深圳)有限公司 | External parameter calibration method for 2D laser radar on linear motion mechanism |
| CN115002906A (en) * | 2022-08-05 | 2022-09-02 | 中昊芯英(杭州)科技有限公司 | Object positioning method, device, medium and computing equipment |
| CN114994648B (en) * | 2022-08-05 | 2022-11-08 | 聚时科技(深圳)有限公司 | External parameter calibration method for 2D laser radar on linear motion mechanism |
| CN115002906B (en) * | 2022-08-05 | 2022-11-15 | 中昊芯英(杭州)科技有限公司 | Object positioning method, device, medium and computing equipment |
| CN115661220A (en) * | 2022-12-28 | 2023-01-31 | 深圳煜炜光学科技有限公司 | Point cloud data registration method, device, equipment and storage medium |
| CN115661220B (en) * | 2022-12-28 | 2023-03-17 | 深圳煜炜光学科技有限公司 | Point cloud data registration method, device, equipment and storage medium |
| CN116400334A (en) * | 2023-06-01 | 2023-07-07 | 未来机器人（深圳）有限公司 | Calibration verification method and device for laser external parameters, electronic equipment and storable medium |
| CN116400334B (en) * | 2023-06-01 | 2023-09-12 | 未来机器人（深圳）有限公司 | Calibration verification method and device for laser external parameters, electronic equipment and storable medium |
| CN116878402A (en) * | 2023-07-11 | 2023-10-13 | 北京博科测试系统股份有限公司 | Non-contact wheel arch measuring sensor and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114296056A (en) | Laser radar external parameter calibration method, device, equipment and storage medium | |
| CN109143207B (en) | Laser radar internal reference precision verification method, device, equipment and medium | |
| CN109598765B (en) | Monocular camera and millimeter wave radar external parameter combined calibration method based on spherical calibration object | |
| CN108872991A (en) | Target analyte detection and recognition methods, device, electronic equipment, storage medium | |
| CN112513679B (en) | Target identification method and device | |
| CN110865393A (en) | Positioning method and system based on laser radar, storage medium and processor | |
| CN112308916B (en) | Target pose recognition method based on image target | |
| CN110673107B (en) | Road edge detection method and device based on multi-line laser radar | |
| CN107527368B (en) | Three-dimensional space attitude positioning method and device based on two-dimensional code | |
| CN112379352B (en) | Laser radar calibration method, device, equipment and storage medium | |
| CN112686950B (en) | Pose estimation method, pose estimation device, terminal equipment and computer readable storage medium | |
| CN113743385A (en) | Unmanned ship water surface target detection method and device and unmanned ship | |
| CN111046877A (en) | Millimeter wave image suspicious article detection method and system | |
| CN111968224A (en) | Ship 3D scanning point cloud data processing method | |
| CN115685160A (en) | Target-based laser radar and camera calibration method, system and electronic equipment | |
| CN111880195A (en) | Tower crane anti-collision method and system based on laser radar | |
| CN111913177A (en) | Method and device for detecting target object and storage medium | |
| CN113902782A (en) | Rapid registration method and system for three-dimensional point cloud of obstacles around excavator | |
| CN113870357A (en) | Camera external parameter calibration method and device, sensing equipment and storage medium | |
| Sun et al. | High-accuracy three-dimensional measurement based on multi-directional cooperative target with weighted SfM algorithm | |
| CN114972532A (en) | Method, device and equipment for calibrating external parameters between laser radars and storage medium | |
| CN107765257A (en) | A kind of laser acquisition and measuring method based on the calibration of reflected intensity accessory external | |
| CN117095038A (en) | Point cloud filtering method and system for laser scanner | |
| CN113205563B (en) | Automatic driving sensor combined calibration target and calibration method | |
| CN114384486A (en) | Data processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |