WO2022165793A1

WO2022165793A1 - Extrinsic parameter calibration method and apparatus and computer readable storage medium

Info

Publication number: WO2022165793A1
Application number: PCT/CN2021/075766
Authority: WO
Inventors: 薛连杰; 万富华; 李延召
Original assignee: 深圳市大疆创新科技有限公司
Priority date: 2021-02-07
Filing date: 2021-02-07
Publication date: 2022-08-11

Abstract

Disclosed in embodiments of the present application is an extrinsic parameter calibration method, comprising: obtaining first coordinates of a reference object in an actual coordinate system of a laser radar; determining second coordinates of the reference object in a design coordinate system of the laser radar; determining a first rotation matrix between the actual coordinate system and the design coordinate system according to the first coordinates and the second coordinates; and determining, according to the first rotation matrix and design extrinsic parameters, a target extrinsic parameter between the actual coordinate system and a coordinate system of a movable platform carrying the laser radar, the design extrinsic parameters being extrinsic parameters between the design coordinate system and the coordinate system of the movable platform. According to the method disclosed in the embodiments of the present application, the extrinsic parameter between the actual coordinate system of the laser radar and the coordinate system of the movable platform can be calibrated with high precision.

Description

External parameter calibration method, device and computer readable storage medium

technical field

The present application relates to the technical field of external parameter calibration, and in particular, to an external parameter calibration method, device, and computer-readable storage medium.

Background technique

The mobile platform can perceive the environment through the onboard lidar. When installing the lidar on the movable platform, the lidar needs to be installed according to the designed angle and position. However, there is an error in the installation of the lidar. This error causes the coordinates of the point cloud points collected by the lidar to be converted by designing external parameters, and the position of the point cloud point indicated by the converted coordinates relative to the movable platform will be Wrong, it is not conducive to the use of point clouds for mobile platforms such as obstacle avoidance.

SUMMARY OF THE INVENTION

In view of this, the embodiments of the present application provide an external parameter calibration method, device, and computer-readable storage medium, one of which is to calibrate the actual coordinate system of the laser radar with higher precision and the coordinate system of the movable platform. of external parameters.

A first aspect of the embodiments of the present application provides an external parameter calibration method, including:

Obtain the first coordinate of the reference object in the actual coordinate system of the lidar;

determining the second coordinate of the reference object in the design coordinate system of the lidar;

determining a first rotation matrix between the actual coordinate system and the design coordinate system according to the first coordinate and the second coordinate;

According to the first rotation matrix and design extrinsic parameters, determine the target extrinsic parameters between the actual coordinate system and the coordinate system of the movable platform carrying the lidar, and the design extrinsic parameters are the design coordinate system and The extrinsic parameter between the coordinate systems of the movable platform.

A second aspect of an embodiment of the present application provides an external parameter calibration device, comprising: a processor and a memory storing a computer program, where the processor implements the following steps when executing the computer program:

A third aspect of the embodiments of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, implements the external parameter calibration method provided in the first aspect.

The external parameter calibration method provided in the embodiment of the present application does not directly solve the conversion relationship between the actual coordinate system of the lidar and the coordinate system of the movable platform, but first considers the relationship between the actual coordinate system of the lidar and the design coordinate system The conversion between the design coordinate system and the coordinate system of the movable platform is combined with the conversion between the coordinate system of the movable platform, and the target external parameter is finally obtained. Since the design extrinsic parameters for conversion between the design coordinate system and the coordinate system of the movable platform are known and accurate, and the translation vector can be ignored between the actual coordinate system and the design coordinate system, only Determine the rotation matrix (the first rotation matrix), so in the whole process of determining the external parameters of the target, only the first rotation matrix needs to be newly determined for the external parameters, which simplifies the calculation process and improves the calculation efficiency. There are few external parameters that need to be recalculated, so the error introduced in the target external parameters is small, and the accuracy of the calibrated target external parameters is improved.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative labor.

FIG. 1 is a schematic diagram of the installation angle and position of the lidar on the unmanned vehicle provided by the embodiment of the present application.

FIG. 2 is a flowchart of an external parameter calibration method provided by an embodiment of the present application.

FIG. 3 is a schematic structural diagram of a target provided in an embodiment of the present application.

FIG. 4 is a side view of a reference object layout provided by an embodiment of the present application.

FIG. 5 is a front view of a reference object layout provided by an embodiment of the present application.

FIG. 6 is a schematic diagram of a scene in which a reference object is measured by a laser radar according to an embodiment of the present application.

FIG. 7 is a schematic diagram of a first direction vector of a reference object in an actual coordinate system of a lidar according to an embodiment of the present application.

FIG. 8 is a schematic diagram of a second direction vector of a reference object provided in an embodiment of the present application in a design coordinate system of a lidar.

FIG. 9 is a comparison diagram of a first direction vector and a second direction vector provided by an embodiment of the present application.

FIG. 10 is a schematic structural diagram of an external parameter calibration device provided by an embodiment of the present application.

Detailed ways

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. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

Movable platforms, such as unmanned vehicles, drones, unmanned ships, robots, etc., need to perceive the environment. In one embodiment, the movable platform can be equipped with a laser radar, and the environment can be scanned by the laser radar to obtain point cloud data corresponding to the environment. Information about the object, such as the location of the object, the category of the object, and so on.

When installing lidar on a movable platform, the lidar needs to be installed according to the designed angle and position. Here, an unmanned vehicle (autonomous vehicle) can be taken as an example. As shown in Figure 1, in the design, the lidar can be installed in the left front of the unmanned vehicle, and the installation angle of the lidar can be the ortho-radiation direction of the lidar. The forward direction of the unmanned vehicle is the same, that is, when the lidar is installed on the unmanned vehicle at the designed angle, the coordinate system (XYZ) of the lidar and the body coordinate system (X'Y'Z') of the unmanned vehicle are different. The axis directions are the same. Here, the coordinate system of the lidar at the designed angle and the designed position can be called the design coordinate system of the lidar.

If the lidar can be installed at the design position of the movable platform at the designed angle without error, the coordinates of the point cloud points collected by the lidar are the coordinates under the design coordinate system of the lidar, and the coordinates in the design coordinate system After the coordinates are converted into the coordinate system of the movable platform through the design external parameters, the position of the point cloud point relative to the movable platform can be accurately described. However, there is an error in the installation of the lidar. This error makes the coordinate system where the coordinates of the point cloud points collected by the lidar are located is not the design coordinate system. If the coordinates are still converted through the design external parameters, the conversion The position of the point cloud point indicated by the latter coordinates relative to the movable platform will be wrong. Further, if the movable platform performs obstacle recognition based on the coordinates of these point cloud points, the identified obstacle position will also be incorrect. is wrong, resulting in an obstacle avoidance accident. Here, the coordinate system corresponding to the laser radar at the actual installation position and installation angle can be called the actual coordinate system of the laser radar.

In order to solve the above problems, in one embodiment, the external parameters (target external parameters) between the actual coordinate system of the lidar and the coordinate system of the movable platform can be calibrated, and the calibrated target external parameters can be used to The coordinates of the point cloud points collected by the lidar are converted from the actual coordinate system of the lidar to the coordinate system of the movable platform, and the coordinates converted to the coordinate system of the movable platform and the point cloud points are relative to where the movable platform is located. match the real location. Here, if the movable platform is an unmanned vehicle, the coordinate system of the movable platform can be called the body coordinate system, and if the movable platform is a drone, the coordinate system of the movable platform can be called the body coordinate system.

When calibrating the target external parameter, in one embodiment, a reference object can be arranged in the calibration site, and the reference object can be various objects that can be detected by lidar, which can be used as the object to be measured during the calibration process. target, such as a target. After the reference object is arranged, the accurate third coordinate of the reference object in the coordinate system of the movable platform can be measured by a high-precision measuring device, and the point cloud of the scene can be collected by the lidar mounted on the movable platform to obtain The first coordinate of the reference object in the actual coordinate system of the lidar, so that the target external parameter between the actual coordinate system and the coordinate system of the movable platform can be determined according to the first coordinate and the third coordinate .

The external parameters can be used to convert coordinates in different coordinate systems, for example, the target external parameters can be used to convert coordinates in the actual coordinate system and the coordinate system of the movable platform. The external parameter may include a rotation matrix and a translation vector. In the above embodiment, when determining the target external parameter according to the first coordinate and the third coordinate, both the rotation matrix and the translation vector in the target external parameter need to be determined. To optimize, it is necessary to solve the rotation matrix and translation vector in the target external parameter through the optimization algorithm. However, in the process of optimization and solution, the rotation matrix and translation vector may not be well converged, and the final solution of the target external parameter may not have sufficient accuracy, that is, the point collected by the lidar using the target external parameter. After the coordinates of the cloud point are converted, the position of the point cloud point indicated by the converted coordinates relative to the movable platform is still inaccurate.

Considering the above problems, an embodiment of the present application provides a method for calibrating external parameters. Referring to FIG. 2, FIG. 2 is a flowchart of the method for calibrating external parameters provided by the embodiment of the present application. The method may include the following steps:

S202: Acquire the first coordinates of the reference object in the actual coordinate system of the lidar.

S204. Determine the second coordinate of the reference object in the design coordinate system of the lidar.

S206. Determine a first rotation matrix between the actual coordinate system and the design coordinate system according to the first coordinate and the second coordinate.

S208 , according to the first rotation matrix and the design external parameters, determine the target external parameters between the actual coordinate system and the coordinate system of the movable platform carrying the lidar.

As mentioned above, when calibrating the target external parameter, a reference object can be arranged in the calibration field. In order to make the first coordinate of the reference object measured by the lidar more accurate, in an embodiment, the reference object may include a first part and a second part, wherein the reflectivity of the first part is higher than that of the second part. Since the reflectivity of the second part is lower than that of the first part, the point cloud corresponding to the first part can have a clearer outline under the background of the point cloud corresponding to the second part, so that the reference object measured by the lidar can be obtained. The first coordinate of is more accurate. Further, the first part can be surrounded by the second part, so that the entire outline of the point cloud corresponding to the first part can be more clearly set off against the point cloud corresponding to the second part. In one example, the first part may be located in the middle of the reference object, and other parts other than the middle part may be the second part.

Referring to FIG. 3 , FIG. 3 is a schematic structural diagram of a target provided in an embodiment of the present application. As shown in FIG. 3 , the reference can be a target, and the target can include a first part 310 and a second part 320 , the first part can be located at the center of the target, and the second part can surround the first part and located at the border of the target. In one example, the second part of the target can be made of black foam and the first part can be made of a totally reflective material. In one example, the second part of the target can be black foam, the black foam can be used as the bottom plate of the target, the first part can be a total reflection patch, and the total reflection patch can be attached to the center of the black foam. In the example shown in Figure 3, the target is a 50CM*50CM square plate, but it can be understood that the size and shape of the target are not limited to the design shown in Figure 3, which can be determined according to the scanning density of the lidar and the calibration site. Area and other flexible designs for rectangular plates, circular plates and various other sizes and shapes.

When arranging the reference objects, in one embodiment, the number of the reference objects to be arranged may be N, where N is an integer greater than or equal to 3. Considering that lidar has different accuracy characteristics at different distances and different positions in the field of view, the arrangement of the reference objects can cover as many distances and more positions as possible in the field of view. For example, in one embodiment, the N reference objects may be arranged into multiple reference object rows, each row may contain one or more reference objects, and each row may correspond to a distance in front of the lidar field of view. Referring to Figure 4, the N reference objects (targets) in Figure 4 can be arranged into 2 reference object rows, wherein the first reference object row can be located at a distance of 7m in front of the lidar field of view, and the second reference object row can be located at a distance of 7m. The distance of 10m in front of the lidar field of view.

Considering that the reference objects between different reference object rows may form occlusions, for example, in Figure 4, the reference objects of the first reference object row may form occlusions to the reference objects of the second reference object row, resulting in the lidar being unable to detect the second reference object. Therefore, in an embodiment, the reference objects between the reference object rows can be arranged in a cross manner. As shown in Figure 5, the positions of the reference objects are staggered from each other, so that each reference object can be scanned by the lidar, and there is no situation that the reference objects in the rear row are blocked by the reference objects in the front row.

In one embodiment, the N reference objects may be distributed at different positions in the lidar field of view. As shown in Figure 5, in the field of view of the lidar, some reference objects are distributed above the field of view, some reference objects are distributed on the left side of the field of view, and some reference objects are distributed on the right side of the field of view, so that the Comprehensive coverage of each area in the lidar field of view, so that the final calibrated target external parameters have higher accuracy.

After the reference object arrangement of the calibration site is completed, the movable platform can be parked at the designated position, and the point cloud corresponding to the scene can be collected by the lidar carried on the movable platform. The first coordinate in the actual coordinate system of . You can refer to Figure 6. As shown in Figure 6, the movable platform is an unmanned vehicle. After the unmanned vehicle is parked in a designated position, the reference object (target) in front of the field of view can be measured by the lidar in front of it. Thus, the first coordinates of the reference object can be determined. It should be noted that the layout of the reference objects shown in FIG. 6 is only an example, and the actual layout of the reference objects can be designed according to actual needs.

When the first coordinates of the reference object are determined according to the collected point cloud, in one embodiment, the coordinates of the center point of the first part of the reference object with high reflectivity may be determined as the first coordinates of the reference object, as shown in FIG. 3 . shown, that is, the coordinates of the bullseye of the target can be determined as the coordinates of the target. Specifically, after the point cloud corresponding to the scene is collected by the lidar, the target point cloud point whose reflectivity is higher than the threshold can be extracted according to the reflectivity of the point cloud point in the point cloud. It can be understood that the target point cloud point corresponds to the point cloud point corresponding to the first part of the reference object. Since there are multiple reference objects, the extracted target point cloud points can be clustered according to the distance between the point cloud points to obtain multiple categories. Here, each category can correspond to the first reference object. part. For each of the multiple categories, the coordinates of the center point of the point cloud in the category may be calculated, and the coordinates of the center point may be determined as the first coordinates of the reference object corresponding to the category.

When calculating the coordinates of the center point of a point cloud in a class, there can be various implementations. In one embodiment, the average value of the coordinates of each point cloud point in the category may be calculated, and the coordinates corresponding to the average value may be determined as the coordinates of the center point of the category. In one embodiment, the point cloud contour of the category can be determined, so that the coordinates of the center point corresponding to the category are calculated according to the coordinates of the point cloud points located in the contour.

It should be noted that, in the above embodiment, the first coordinate of the reference object needs to be determined according to the point cloud point corresponding to the reference object collected by the lidar. Therefore, in order to make the determined first coordinate of the reference object more accurate, the lidar The collected point cloud points should cover the entire reference object more evenly. In one embodiment, the lidar may be a non-repetitive scanning lidar, and the point cloud points in the area scanned by the lidar may be denser or more uniformly distributed, so the calculated first coordinate of the reference object is more accurate. In one embodiment, the lidar can be a repetitive scanning lidar, so it needs to configure enough laser lines to reduce the interval between the point cloud point row and the point cloud point row obtained by scanning, otherwise, calculate The first coordinates of the reference object may not be accurate enough.

For the second coordinate of the reference object under the design coordinate system of the lidar, in one embodiment, the third coordinate of the reference object under the coordinate system of the movable platform can be obtained by converting the reference object according to the design external parameters. As mentioned above, the coordinates of the reference object in the coordinate system of the movable platform can be measured by a high-precision measuring device. Here, the measuring device can be a total station, a three-dimensional laser scanner, or the like. For example, if the movable platform is an unmanned vehicle, the measuring device can be fixed at the origin of the body coordinate system when the unmanned vehicle is parked at the specified position, so that the position of the reference object can be measured by the measuring device, then The measured coordinates of the reference object are the third coordinates of the reference object in the body coordinate system of the unmanned vehicle. The third coordinate can be converted from the coordinate system of the movable platform to the design coordinate system of the lidar through the design external parameters. The design external parameters are the design coordinate system of the lidar and the coordinate system of the movable platform that have been determined in the design stage. After the external parameter between and the third coordinate is converted by the design external parameter, the converted coordinate is the second coordinate of the reference object in the design coordinate system of the lidar.

Considering that the translation error of the lidar during installation is small, and the translation error has little effect on the accuracy after coordinate conversion, therefore, the translation error of the lidar can be ignored during installation, and only the angle error of the lidar during installation is considered. , the angle error is the deviation between the installation angle and the design angle of the lidar. When only the deviation between the installation angle and the design angle of the lidar is considered, only the rotation transformation relationship between the actual coordinate system and the design coordinate system of the lidar can be considered, and the translation transformation relationship between the two can be ignored. Therefore, When determining the external parameter between the actual coordinate system and the design coordinate system, only the rotation matrix (ie, the first rotation matrix) in the external parameter may be determined, and the translation vector in the external parameter is ignored.

The first rotation matrix may be determined according to the first coordinates of the reference object in the actual coordinate system of the lidar and the second coordinates of the reference object in the design coordinate system of the lidar. Specifically, in one embodiment, the first coordinate of the reference object in the actual coordinate system can be converted into a direction vector (first direction vector) of the reference object in the actual coordinate system. Correspondingly, refer to The second coordinate of the object under the design coordinate system can also be converted into the direction vector (second direction vector) of the reference object under the design coordinate system, so that the first the direction vector and the second direction vector to obtain the first rotation matrix. The so-called registration of the first direction vector and the second direction vector is to solve a rotation matrix so that after the rotation matrix acts on the first direction vector, the first direction vector can be rotated to be as close to the second direction vector as possible. coincide. 7 to 9, Fig. 7 shows the first direction vector of the reference object in the actual coordinate system of the lidar, Fig. 8 shows the second direction vector of the reference object in the design coordinate system of the lidar, Fig. 9 A comparison diagram showing the first direction vector and the second direction vector.

When registering the first direction vector and the second direction vector, specifically, the first direction vector may be normalized, the normalized first direction vector may be recorded as β, and the first direction vector may be recorded as β. The two direction vectors are normalized, and the normalized second direction vector can be recorded as α. A plurality of the first direction vectors may be denoted as matrix B, and a plurality of the second direction vectors may be denoted as matrix A, and matrix B and matrix A are as follows:

As mentioned above, the first rotation matrix to be determined is so that after the first rotation matrix acts on the first direction vector, the first direction vector and the second direction vector can overlap as much as possible, which can be expressed in mathematical form as passing through The problem that the inner product of the first direction vector and the second direction vector after rotation is the largest. The problem can be expressed as follows:

Wherein, R represents the first rotation matrix, and the matrix BAT can be decomposed by singular value, that is, SVD decomposition, so as to obtain ^UDVT , U is the left singular value vector, D is the diagonal matrix, and V is the right singular value vector. Substitute UDVT into the above formula to get:

Since D is a diagonal matrix, the solution for R is:

R= ^VUT

The rotation matrix R is the external parameter rotation matrix between the actual coordinate system and the design coordinate system of the lidar, that is, the first rotation matrix.

Using the first rotation matrix, the coordinates in the actual coordinate system measured by the lidar can be converted into the design coordinate system, and the coordinates in the design coordinate system can be converted into the movable platform through the known design external parameters Under the coordinate system, the transformation of coordinates from the actual coordinate system of the lidar to the coordinate system of the movable platform is realized. In other words, the target extrinsic parameter between the actual coordinate system and the coordinate system of the movable platform includes the first rotation matrix and the design extrinsic parameter, and the first rotation matrix is used for coordinates between the actual coordinate system and the design coordinate For conversion between systems, the design external parameters are used for coordinate conversion between the design coordinate system and the coordinate system of the movable platform.

In one embodiment, the target external parameter may include a target rotation matrix and a target translation vector, the design external parameter may include a design rotation matrix and a design translation vector, and the target rotation matrix in the target external parameter may be the first rotation The matrix and the design rotation matrix are obtained by fusion. For example, in an example, the target rotation matrix Rtarget, the first rotation matrix R and the design rotation matrix Rd may have the following relationship: Rtarget=R*Rd. Since the translation transformation between the actual coordinate system and the design coordinate system is ignored, the target translation vector in the target extrinsic parameter can reuse the design translation vector of the design extrinsic parameter, that is, the target translation vector ttarget can be the same as the design translation vector td is the same, ie ttarget=td.

Referring to FIG. 10, FIG. 10 is a schematic structural diagram of an external parameter calibration apparatus provided in an embodiment of the present application. The apparatus may include: a processor 1010 and a memory 1020 storing a computer program. When the processor executes the computer program Implement the following steps:

Optionally, the second coordinate of the reference object under the design coordinate system is obtained by converting the third coordinate of the reference object under the coordinate system of the movable platform through the design external parameter.

Optionally, the third coordinate of the reference object in the coordinate system of the movable platform is obtained by measuring using a measuring device.

Optionally, the measurement equipment includes: a total station or a three-dimensional laser scanner.

Optionally, the reference object includes a first part and a second part, and the reflectivity of the first part is higher than that of the second part.

Optionally, the first portion is surrounded by the second portion.

Optionally, the first part is located in the middle of the reference object.

Optionally, the first part includes a total reflection patch, and the second part includes black foam.

Optionally, the first coordinates of the reference object are determined according to coordinates of target point cloud points whose reflectivity is higher than a threshold in the point cloud collected by the lidar.

Optionally, when the processor determines the first coordinates of the reference object according to the coordinates of the target point cloud point:

Clustering the target point cloud points according to the distance between the point cloud points;

The coordinates of the center point are respectively determined for each category obtained by clustering, and the coordinates of the center point of the category are determined as the first coordinates of the reference object.

Optionally, the number of the reference objects is N, and N is an integer greater than or equal to 3.

Optionally, the N reference objects form at least two reference object rows at different distances in front of the lidar field of view.

Optionally, the reference objects between the reference object rows are arranged crosswise.

Optionally, the N reference objects are distributed at different positions in the field of view of the lidar.

Optionally, when determining the first rotation matrix between the actual coordinate system and the design coordinate system according to the first coordinate and the second coordinate, the processor is configured to:

determining the first direction vector of the reference object in the actual coordinate system according to the first coordinate;

determining a second direction vector of the reference object in the design coordinate system according to the second coordinate;

The first direction vector and the second direction vector are registered to obtain the first rotation matrix.

Optionally, the lidar includes: a non-repetitive scanning lidar.

Optionally, the target external parameter includes a target rotation matrix and a target translation vector, the target rotation matrix is obtained by fusing the first rotation matrix and the design rotation matrix in the design external parameter, and the target translation vector The same as the design translation vector in the design extrinsic parameter.

Optionally, the movable platform carrying the lidar includes an unmanned vehicle, and the coordinate system of the movable platform includes a vehicle body coordinate system.

Optionally, the design coordinate system of the laser radar is a coordinate system corresponding to the designed position and designed angle of the laser radar.

For the specific implementation of the external parameter calibration apparatuses of the various embodiments provided above, reference may be made to the relevant descriptions above, which will not be repeated here.

The external parameter calibration device provided in the embodiment of the present application does not directly solve the conversion relationship between the actual coordinate system of the lidar and the coordinate system of the movable platform, but first considers the relationship between the actual coordinate system of the lidar and the design coordinate system The conversion between the design coordinate system and the coordinate system of the movable platform is combined with the conversion between the coordinate system of the movable platform, and the target external parameter is finally obtained. Since the design extrinsic parameters for conversion between the design coordinate system and the coordinate system of the movable platform are known and accurate, and the translation vector can be ignored between the actual coordinate system and the design coordinate system, only Determine the rotation matrix (the first rotation matrix), so in the whole process of determining the external parameters of the target, only the first rotation matrix needs to be newly determined for the external parameters, which simplifies the calculation process and improves the calculation efficiency. There are few external parameters that need to be recalculated, so the error introduced in the target external parameters is small, and the accuracy of the calibrated target external parameters is improved.

Embodiments of the present application further provide a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, implements the external parameter calibration method provided by the embodiments of the present application.

Various implementations are provided above for each protection subject. On the basis of no conflict or contradiction, those skilled in the art can freely combine various implementations according to the actual situation, thereby forming various technical solutions. . However, this application document is limited in space and cannot describe all the technical solutions obtained by combination, but it can be understood that these technical solutions that cannot be developed also belong to the scope disclosed in the embodiments of this application.

Embodiments of the present application may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having program code embodied therein. Computer-usable storage media includes both persistent and non-permanent, removable and non-removable media, and storage of information can be accomplished by any method or technology. Information may be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), Flash Memory or other memory technology, Compact Disc Read Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any relationship between these entities or operations. any such actual relationship or sequence exists. The terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion such that a process, method, article or device comprising a list of elements includes not only those elements, but also other not expressly listed elements, or also include elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

The methods and devices provided by the embodiments of the present invention have been described in detail above. The principles and implementations of the present invention are described with specific examples in this paper. The descriptions of the above embodiments are only used to help understand the methods of the present invention and its implementation. At the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation and application scope. To sum up, the content of this description should not be construed as a limitation to the present invention. .

Claims

A kind of external parameter calibration method, is characterized in that, comprises:

Obtain the first coordinate of the reference object in the actual coordinate system of the lidar;

determining the second coordinate of the reference object in the design coordinate system of the lidar;

determining a first rotation matrix between the actual coordinate system and the design coordinate system according to the first coordinate and the second coordinate;

According to the first rotation matrix and design extrinsic parameters, determine the target extrinsic parameters between the actual coordinate system and the coordinate system of the movable platform carrying the lidar, and the design extrinsic parameters are the design coordinate system and The extrinsic parameter between the coordinate systems of the movable platform.
The method according to claim 1, wherein the second coordinate of the reference object in the design coordinate system is the third coordinate of the reference object in the coordinate system of the movable platform passing through the The design extrinsic parameter conversion is obtained.
The method according to claim 2, wherein the third coordinate of the reference object in the coordinate system of the movable platform is obtained by measuring using a measuring device.
The method according to claim 3, wherein the measurement equipment comprises: a total station or a three-dimensional laser scanner.
The method of claim 1, wherein the reference object comprises a first part and a second part, and the reflectivity of the first part is higher than the reflectivity of the second part.
6. The method of claim 5, wherein the first portion is surrounded by the second portion.
The method of claim 6, wherein the first portion is located in the middle of the reference object.
6. The method of claim 5, wherein the first portion comprises a total reflection patch and the second portion comprises black foam.
The method according to claim 5, wherein the first coordinates of the reference object are determined according to the coordinates of the target point cloud points whose reflectivity is higher than a threshold in the point cloud collected by the lidar.
The method according to claim 9, wherein determining the first coordinates of the reference object according to the coordinates of the target point cloud point comprises:

Clustering the target point cloud points according to the distance between the point cloud points;

The coordinates of the center point are respectively determined for each category obtained by clustering, and the coordinates of the center point of the category are determined as the first coordinates of the reference object.
The method according to claim 1, wherein the number of the reference objects is N, and the N is an integer greater than or equal to 3.
The method according to claim 11, wherein the N reference objects form at least two reference object rows at different distances in front of the lidar field of view.
The method according to claim 12, wherein the reference objects between the reference object rows are arranged crosswise.
The method according to claim 11, wherein the N reference objects are distributed at different positions in the field of view of the lidar.
The method according to claim 1, wherein the determining, according to the first coordinates and the second coordinates, a first rotation matrix between the actual coordinate system and the design coordinate system comprises:

determining the first direction vector of the reference object in the actual coordinate system according to the first coordinate;

determining a second direction vector of the reference object in the design coordinate system according to the second coordinate;

The first direction vector and the second direction vector are registered to obtain the first rotation matrix.
The method of claim 1, wherein the lidar comprises: a non-repetitive scanning lidar.
The method according to claim 1, wherein the target external parameter comprises a target rotation matrix and a target translation vector, and the target rotation matrix is the first rotation matrix and the design rotation matrix in the design external parameter After fusion, the target translation vector is the same as the design translation vector in the design external parameter.
The method according to claim 1, wherein the movable platform on which the lidar is mounted comprises an unmanned vehicle, and the coordinate system of the movable platform comprises a vehicle body coordinate system.
The method according to claim 1, wherein the design coordinate system of the laser radar is a coordinate system corresponding to the designed position and designed angle of the laser radar.
A device for calibrating external parameters, comprising: a processor and a memory storing a computer program, wherein the processor implements the following steps when executing the computer program:

Obtain the first coordinate of the reference object in the actual coordinate system of the lidar;

determining the second coordinate of the reference object in the design coordinate system of the lidar;

determining a first rotation matrix between the actual coordinate system and the design coordinate system according to the first coordinate and the second coordinate;

According to the first rotation matrix and design extrinsic parameters, the target extrinsic parameters between the actual coordinate system and the coordinate system of the movable platform carrying the lidar are determined, and the design extrinsic parameters are the design coordinate system and The extrinsic parameter between the coordinate systems of the movable platform.
The device according to claim 20, wherein the second coordinate of the reference object in the design coordinate system is the third coordinate of the reference object in the coordinate system of the movable platform passing through the It is obtained by converting the design external parameters.
The apparatus according to claim 21, wherein the third coordinate of the reference object in the coordinate system of the movable platform is obtained by measuring using a measuring device.
The apparatus according to claim 22, wherein the measurement equipment comprises: a total station or a three-dimensional laser scanner.
21. The device of claim 20, wherein the reference object comprises a first portion and a second portion, and the reflectivity of the first portion is higher than the reflectivity of the second portion.
25. The device of claim 24, wherein the first portion is surrounded by the second portion.
26. The device of claim 25, wherein the first portion is located in the middle of the reference.
25. The device of claim 24, wherein the first portion comprises a total reflection patch and the second portion comprises black foam.
The device according to claim 24, wherein the first coordinate of the reference object is determined according to the coordinate of the target point cloud point whose reflectivity is higher than a threshold in the point cloud collected by the lidar.
The apparatus according to claim 28, wherein when the processor determines the first coordinates of the reference object according to the coordinates of the target point cloud points:

Clustering the target point cloud points according to the distance between the point cloud points;

The coordinates of the center point are respectively determined for each category obtained by clustering, and the coordinates of the center point of the category are determined as the first coordinates of the reference object.
The device according to claim 20, wherein the number of the reference objects is N, and the N is an integer greater than or equal to 3.
The device according to claim 30, wherein the N reference objects form at least two reference object rows at different distances in front of the lidar field of view.
The apparatus of claim 31 , wherein the reference objects between the reference object rows are arranged in a cross-arrangement.
The apparatus according to claim 30, wherein the N reference objects are distributed at different positions in the field of view of the lidar.
The apparatus according to claim 20, wherein the processor determines a first rotation matrix between the actual coordinate system and the design coordinate system according to the first coordinate and the second coordinate When used for:

Determine the first direction vector of the reference object in the actual coordinate system according to the first coordinate;

determining a second direction vector of the reference object in the design coordinate system according to the second coordinate;

The first direction vector and the second direction vector are registered to obtain the first rotation matrix.
The apparatus of claim 20, wherein the lidar comprises: a non-repetitive scanning lidar.
The apparatus according to claim 20, wherein the target external parameter comprises a target rotation matrix and a target translation vector, and the target rotation matrix is the first rotation matrix and the design rotation matrix in the design external parameter After fusion, the target translation vector is the same as the design translation vector in the design external parameter.
The device according to claim 20, wherein the movable platform on which the lidar is mounted comprises an unmanned vehicle, and the coordinate system of the movable platform comprises a vehicle body coordinate system.
The device according to claim 20, wherein the design coordinate system of the laser radar is a coordinate system corresponding to the designed position and designed angle of the laser radar.
A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the external parameter calibration method according to any one of claims 1-19 is realized .