Low-wire-harness multi-laser-radar automatic calibration method

Abstract

The invention relates to a low-wire-harness multi-laser-radar automatic calibration method and device, electronic equipment and a storage medium. The method comprises the steps that after static point cloud data collected by a preset number of low-wire-harness radars are preprocessed, ground point cloud extraction is carried out based on an improved RANSAC algorithm, and a ground plane is generated; based on a ground plane extracted from the point cloud data of the preset number of low-wire-harness radars, plane matching is carried out to solve an external parameter initial value; and setting a region of interest (ROI), carrying out iterative closest point (ICP) algorithm optimization by a hierarchical search strategy to generate an external parameter matrix, and completing automatic calibration of the low-wire-harness multi-laser radar. According to the target-free on-line calibration method based on the natural scene features, no pre-arranged facility is needed, full-scene self-adaptive calibration is achieved, and the flexibility of a calibration scheme and the convenience of long-term maintenance are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of automatic driving and robot perception, and in particular, to a low-beam multi-laser radar automatic calibration method and device, an electronic device, and a computer readable storage medium. BACKGROUND

[0002] With the development of automatic driving and robot perception technology, multi-laser radar systems are widely used in environment modeling, target detection and positioning. Traditional external parameter calibration methods mainly rely on calibration boards (such as checkerboards or planar targets), and point cloud matching is performed by extracting geometric features (such as planes and corner points) on the calibration board to solve the relative pose between radars. However, low-beam radars (such as 16-line and 32-line) have the following problems due to the sparsity of point clouds: 1. Ground unevenness problem: After point cloud splicing, the traditional calibration method causes the ground plane to be discontinuous due to calibration errors, affecting the subsequent point cloud fusion and clustering effect. 2. Strong dependence on calibration board: In outdoor or no-calibration scene, a large-size calibration board needs to be made to ensure sufficient beam coverage, increasing the calibration complexity. 3. Low feature matching accuracy: The point cloud of a low-beam radar is sparse, and the feature extraction of the calibration board is unstable, making it difficult for the external parameter optimization to converge. Although existing solutions (such as ICP matching based on calibration boards) can achieve rough calibration, they cannot solve the problems of ground unevenness and poor adaptability in dynamic scenes. Therefore, there is an urgent need for a self-adaptive external parameter calibration method based on environmental features without a calibration board.

[0003] In the prior art, a multi-laser radar calibration method based on dynamic target point clouds first acquires point cloud data collected by two laser radars on a highway; a vehicle perception algorithm is used to identify the point cloud data to obtain an identification result containing the vehicle trajectory points, and the vehicle trajectory information is determined accordingly; the trajectory information is projected onto a two-dimensional plane and rasterized to match the trajectories of the two radars, and a vehicle trajectory matching result is obtained; finally, the corresponding trajectory points of the same vehicle at the same time are found in the matched overlapping grid, and the external parameter calibration between the two laser radars is completed by stacking the point clouds and calculating. The above scheme cannot solve the core splicing problem of "ground unevenness".

[0004] Meanwhile, a vehicle-mounted multi-laser radar calibration system fixes a vehicle to be calibrated in a calibration scene meeting preset data acquisition conditions through a scene calibration device to reduce dependence on a specific environment; a radar calibration device performs registration among multiple radars with a main laser radar as a reference to determine pose adjustment parameters of the remaining radars; and then a point cloud pose transformation device is used to complete pose adjustment of the multiple laser radars to achieve the calibration task. The present application solves the problem in the prior art that it is difficult to balance calibration efficiency, computing power and effect, and there is dependence on a static calibration environment, resulting in high cost and difficulty in large-scale calibration scenarios, and achieves the technical effects of ensuring calibration effect through a fixed scene, reducing computing load and improving calibration efficiency. The above scheme optimizes the efficiency of in-scene batch calibration, but its essence still depends on a calibration device and a calibration scene, and cannot support online calibration and real-time parameter correction of a vehicle in an actual working environment such as a field or a mine.

[0005] Therefore, one or more methods are needed to solve the above problems.

[0006] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present disclosure, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0007] The purpose of the present disclosure is to provide a low-beam multi-laser radar automatic calibration method, device, electronic equipment and computer readable storage medium, thereby at least partially overcoming one or more problems caused by limitations and defects of related technologies.

[0008] According to one aspect of the present disclosure, a low-beam multi-laser radar automatic calibration method is provided, comprising:

[0009] After preprocessing the static point cloud data collected by a preset number of low-beam radars, ground point cloud extraction is performed based on an improved RANSAC algorithm to generate a ground plane;

[0010] Based on the ground plane extracted from the point cloud data of the preset number of low-beam radars, plane matching is performed to solve the initial value of the external parameter;

[0011] An ROI (Region of Interest) is set, and an ICP (Iterative Closest Point) optimization is performed using a hierarchical search strategy to generate an external parameter matrix, thereby completing low-beam multi-laser radar automatic calibration.

[0012] In an exemplary embodiment of the present disclosure, preprocessing the static point cloud data collected by a preset number of low-beam radars in the method comprises:

[0013] The static point cloud data collected by a preset number of low-beam radars is preprocessed by multi-frame superposition;

[0014] Statistical filtering is applied to the superimposed point cloud to remove outlier noise points.

[0015] In one exemplary embodiment of this disclosure, the improved RANSAC algorithm in the method includes:

[0016] The distance threshold of the interior points is dynamically adjusted by dynamically reducing the distance threshold according to the proportion of the interior points.

[0017] Change random sampling to sampling from the 20% of point clouds with the lowest elevation according to weights.

[0018] In one exemplary embodiment of this disclosure, the method further includes:

[0019] Ground point cloud extraction is performed based on the improved RANSAC algorithm. Normal vector constraints are applied to the extracted planes, and only planes with an angle of less than 10 degrees between the normal vector and the gravity direction are retained as effective planes to generate ground planes.

[0020] In one exemplary embodiment of this disclosure, the method further includes:

[0021] By calculating the normal vectors of a preset number of ground planes, the X-axis rotation and Y-axis rotation between radars are solved.

[0022] The Z-axis translation is obtained by calculating the vertical displacement of a preset number of ground planes.

[0023] In one exemplary embodiment of this disclosure, the hierarchical search strategy for Iterative Closest Point (ICP) optimization further includes:

[0024] Coarse search layer: Within the preset range of the initially estimated X-axis translation, Y-axis translation, and Z-axis rotation parameters, a binary search method is used for fast search, and the root mean square error of ICP matching is used as an indicator to quickly lock in the approximate range of these parameters.

[0025] Fine search layer: Based on the coarse search results, a non-convex optimization objective function is constructed, which is composed of the weighted superposition of error terms such as RMSE of ICP matching, point cloud overlap, and edge alignment error. Through iterative optimization, the optimal values ​​of X and Y translation and Z-axis rotation parameters are finally solved.

[0026] In one exemplary embodiment of this disclosure, the method further includes:

[0027] By applying the extrinsic parameter matrix to the point cloud of the radar to be calibrated and projecting it onto the main radar coordinate system, error matching is performed based on a preset threshold to evaluate the calibration results of the extrinsic parameter matrix.

[0028] Generate a multi-view point cloud visualization interface for users to intuitively verify the alignment of point clouds and realize the visualization verification of the calibration results of the external parameter matrix.

[0029] In one aspect of this disclosure, an automatic calibration device for low-beam multi-laser radar is provided, comprising:

[0030] The ground plane extraction module is used to preprocess a preset number of static point cloud data collected by low-beam radar, and then extract the ground point cloud data based on the improved RANSAC algorithm to generate the ground plane.

[0031] The extrinsic parameter initial value solution module is used to perform plane matching to solve for the extrinsic parameter initial values ​​based on the ground plane extracted from the point cloud data of a preset number of low-beam radars.

[0032] The extrinsic parameter matrix generation module is used to set the region of interest (ROI), perform iterative nearest point algorithm (ICP) optimization using a hierarchical search strategy to generate the extrinsic parameter matrix, and complete the automatic calibration of low-beam multi-LiDAR.

[0033] In one aspect of this disclosure, an electronic device is provided, comprising:

[0034] Processor; and

[0035] A memory storing computer-readable instructions that, when executed by the processor, implement the method according to any one of the preceding claims.

[0036] In one aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the method according to any one of the preceding claims.

[0037] An exemplary embodiment of this disclosure provides an automatic calibration method for low-beam multi-LiDAR. The method includes: preprocessing static point cloud data collected by a preset number of low-beam radars; extracting ground point clouds based on an improved RANSAC algorithm to generate ground planes; performing plane matching based on the ground planes extracted from the point cloud data of the preset number of low-beam radars to solve for initial extrinsic parameters; setting a region of interest (ROI); and using a hierarchical search strategy to perform iterative nearest point algorithm (ICP) optimization to generate an extrinsic parameter matrix, thus completing the automatic calibration of the low-beam multi-LiDAR. This targetless, natural scene-based online calibration method requires no pre-deployed facilities, achieves adaptive calibration across all scenes, and significantly improves the flexibility and ease of long-term maintenance of the calibration scheme.

[0038] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0039] The above and other features and advantages of this disclosure will become more apparent from the detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0040] Figure 1 A flowchart of an automatic calibration method for low-beam multi-laser radar according to an exemplary embodiment of the present disclosure is shown;

[0041] Figure 2 A logic flowchart of an automatic calibration method for low-beam multi-lidar radar according to an exemplary embodiment of the present disclosure is shown;

[0042] Figure 3 A flowchart illustrating the algorithm for solving the initial values ​​of extrinsic parameters in an automatic calibration method for low-beam multi-lidar according to an exemplary embodiment of the present disclosure is shown.

[0043] Figure 4 A flowchart illustrating the optimized generation algorithm for the extrinsic parameter matrix of an automatic calibration method for low-beam multi-laser radar according to an exemplary embodiment of the present disclosure is shown.

[0044] Figure 5 A schematic block diagram of an automatic calibration device for low-beam multi-laser radar according to an exemplary embodiment of the present disclosure is shown;

[0045] Figure 6 A block diagram of an electronic device according to an exemplary embodiment of the present disclosure is shown schematically;

[0046] Figure 7 The illustration shows a schematic diagram of a computer-readable storage medium according to an exemplary embodiment of the present disclosure. Detailed Implementation

[0047] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted.

[0048] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details described, or other methods, components, materials, apparatuses, steps, etc., can be employed. In other instances, well-known structures, methods, apparatuses, implementations, materials, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0049] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, or in one or more software-hardened modules, or in different network and / or processor devices and / or microcontroller devices.

[0050] In this example embodiment, an automatic calibration method for low-beam multi-laser radar is first provided; refer to Figure 1 As shown, the automatic calibration method for low-beam multi-lidar can include the following steps:

[0051] Step S110: After preprocessing the static point cloud data collected by a preset number of low-beam radars, ground point cloud extraction is performed based on the improved RANSAC algorithm to generate a ground plane.

[0052] Step S120: Based on the ground plane extracted from the point cloud data of a preset number of low-beam radars, perform plane matching to solve for the initial values ​​of the extrinsic parameters.

[0053] Step S130: Set the Region of Interest (ROI), use a hierarchical search strategy to perform iterative nearest point algorithm (ICP) optimization to generate the extrinsic parameter matrix, and complete the automatic calibration of low-beam multi-LiDAR.

[0054] An exemplary embodiment of this disclosure provides an automatic calibration method for low-beam multi-LiDAR. The method includes: preprocessing static point cloud data collected by a preset number of low-beam radars; extracting ground point clouds based on an improved RANSAC algorithm to generate ground planes; performing plane matching based on the ground planes extracted from the point cloud data of the preset number of low-beam radars to solve for initial extrinsic parameters; setting a region of interest (ROI); and using a hierarchical search strategy to perform iterative nearest point algorithm (ICP) optimization to generate an extrinsic parameter matrix, thus completing the automatic calibration of the low-beam multi-LiDAR. This targetless, natural scene-based online calibration method requires no pre-deployed facilities, achieves adaptive calibration across all scenes, and significantly improves the flexibility and ease of long-term maintenance of the calibration scheme.

[0055] The following will further explain an automatic calibration method for low-beam multi-lidar in this example embodiment.

[0056] Example 1:

[0057] In step S110, the static point cloud data collected by a preset number of low-beam radars can be preprocessed, and then ground point cloud extraction can be performed based on the improved RANSAC algorithm to generate a ground plane.

[0058] In this example embodiment, the preprocessing of a preset number of static point cloud data acquired by low-beam radar includes:

[0059] Preprocess and overlay multiple frames of static point cloud data collected by a preset number of low-beam radars.

[0060] Statistical filtering is applied to the superimposed point cloud to remove outlier noise points.

[0061] In this example embodiment, the improved RANSAC algorithm in the method includes:

[0062] The distance threshold of the interior points is dynamically adjusted by dynamically reducing the distance threshold according to the proportion of the interior points.

[0063] Change random sampling to sampling from the 20% of point clouds with the lowest elevation according to weights.

[0064] In this example embodiment, the method further includes:

[0065] Ground point cloud extraction is performed based on the improved RANSAC algorithm. Normal vector constraints are applied to the extracted planes, and only planes with an angle of less than 10 degrees between the normal vector and the gravity direction are retained as effective planes to generate ground planes.

[0066] In step S120, plane matching can be performed on the ground plane extracted from the point cloud data of a preset number of low-beam radars to solve for the initial values ​​of the extrinsic parameters.

[0067] In this example embodiment, the method further includes:

[0068] By calculating the normal vectors of a preset number of ground planes, the X-axis rotation and Y-axis rotation between radars are solved.

[0069] The Z-axis translation is obtained by calculating the vertical displacement of a preset number of ground planes.

[0070] In this example embodiment, the hierarchical search strategy for Iterative Closest Point (ICP) optimization further includes:

[0071] Coarse search layer: Within the preset range of the initially estimated X-axis translation, Y-axis translation, and Z-axis rotation parameters, a binary search method is used for fast search, and the root mean square error of ICP matching is used as an indicator to quickly lock in the approximate range of these parameters.

[0072] Fine search layer: Based on the coarse search results, a non-convex optimization objective function is constructed, which is composed of the weighted superposition of error terms such as RMSE of ICP matching, point cloud overlap, and edge alignment error. Through iterative optimization, the optimal values ​​of X and Y translation and Z-axis rotation parameters are finally solved.

[0073] In step S130, the Region of Interest (ROI) can be set, and the hierarchical search strategy can be used to perform iterative nearest point algorithm (ICP) optimization to generate the extrinsic parameter matrix, thereby completing the automatic calibration of low-beam multi-LiDAR.

[0074] In this example embodiment, the method further includes:

[0075] By applying the extrinsic parameter matrix to the point cloud of the radar to be calibrated and projecting it onto the main radar coordinate system, error matching is performed based on a preset threshold to evaluate the calibration results of the extrinsic parameter matrix.

[0076] Generate a multi-view point cloud visualization interface for users to intuitively verify the alignment of point clouds and realize the visualization verification of the calibration results of the external parameter matrix.

[0077] In this example embodiment, the automatic extrinsic parameter calibration algorithm for low-beam multi-lidar radar based on ground constraints and ICP optimization of the present invention can achieve rapid and accurate calibration of the extrinsic parameters of two or more low-beam radars in field scenarios without a calibration board. It solves the problem in traditional methods where uneven ground after matching leads to inaccurate ground segmentation, requiring the clipping plane to be moved upwards, resulting in the loss of target points.

[0078] Example 2:

[0079] In the embodiments of this example, as Figure 2 As shown, the point cloud preprocessing and ground extraction steps include:

[0080] First, multiple frames of static point cloud data acquired by various low-beam radars are overlaid to enhance point cloud density, and statistical filtering is applied to the overlaid point cloud to remove outlier noise points. Then, an improved RANSAC algorithm is used for ground point cloud extraction. This improved algorithm significantly enhances the robustness and accuracy of fitting ground planes in sparse low-beam point clouds by dynamically adjusting the inlier distance threshold (dynamically reducing the threshold based on the inlier ratio) and optimizing the sampling strategy (changing random sampling to weighted sampling from the lowest 20% of the point cloud at the lowest elevation). Finally, normal vector constraints are applied to the extracted planes, retaining only planes with an angle of less than 10 degrees between their normal vector and the gravity direction as effective planes.

[0081] In the embodiments of this example, as Figure 3 As shown, the steps for solving the initial values ​​of extrinsic parameters based on ground constraints include:

[0082] Using the ground planes extracted from the two radar point clouds in step 1, plane matching is performed to solve for some initial values ​​of extrinsic parameters. Specifically, the X-axis rotation (Roll) and Y-axis rotation (Pitch) between the radars are solved by calculating the normal vectors of the two ground planes; the Z-axis translation is solved by calculating the vertical displacement of the two planes. This step is one of the core innovations, providing accurate initial values ​​for subsequent optimization by prioritizing the constraint of ground parameters, and fundamentally ensuring the flatness of the ground after point cloud stitching.

[0083] In the embodiments of this example, as Figure 4 As shown, the steps for ROI region setting and hierarchical ICP fine-tuning include:

[0084] After obtaining the initial values ​​of the extrinsic parameters, the system enters the fine-tuning stage. First, regions of interest (ROIs) are established in feature-rich non-ground areas (such as building facades and vehicle outlines) to eliminate interference from featureless areas like the ground and sky. Then, a hierarchical search strategy is employed for ICP optimization.

[0085] Coarse search layer: A binary search is used to quickly search around the initially estimated X-axis translation, Y-axis translation, and Z-axis rotation (Yaw) parameters, using the root mean square error (RMSE) of ICP matching as an indicator to quickly pinpoint the approximate range of these parameters.

[0086] Fine Search Layer: Based on the coarse search results, a comprehensive non-convex optimization objective function is constructed. This function is composed of a weighted sum of various error terms, including the RMSE of ICP matching, point cloud overlap, and edge alignment error. Through iterative optimization, the optimal X and Y translation and Z-axis rotation parameters are finally solved. The optimization process sets a maximum number of iterations and an error threshold as termination conditions.

[0087] In this example embodiment, the calibration result evaluation and visualization verification steps include:

[0088] The optimized extrinsic parameter matrix is ​​applied to the point cloud of the radar to be calibrated and projected onto the main radar coordinate system. The system automatically calculates the matching error of overlapping areas in the point cloud. For radars with different beamlines, the system adaptively adjusts the proportion or weight of points to ensure fair evaluation. If the error exceeds a preset threshold, the system will automatically trigger a recalibration process or prompt for manual intervention. Simultaneously, a multi-view point cloud visualization interface is provided for users to intuitively verify the alignment of the point cloud (especially the ground). After confirmation, the final extrinsic parameter matrix is ​​output.

[0089] In this example embodiment, the calibration system reduces ground overlap error and improves the robustness and accuracy of the calibration system by adjusting and optimizing the convergence criteria to control the accuracy and stability of automatic calibration. The system supports 16-line radar, increasing the calibration success rate. The system introduces result evaluation methods; through visualization of calibration results and manual evaluation, it can determine whether it is necessary to return to the initial process and recalibrate by modifying parameters, thus improving the automation and efficiency of the calibration system.

[0090] It should be noted that although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0091] Furthermore, in this example embodiment, an automatic calibration device for low-beam multi-laser radar is also provided. (Refer to...) Figure 5 As shown, the low-beam multi-laser radar automatic calibration device 400 may include: a ground plane extraction module 410, an external parameter initial value solution module 420, and an external parameter matrix generation module 430. Wherein:

[0092] The ground plane extraction module 410 is used to preprocess the static point cloud data collected by a preset number of low-beam radars, and then extract the ground point cloud data based on the improved RANSAC algorithm to generate the ground plane.

[0093] The extrinsic parameter initial value solving module 420 is used to solve the extrinsic parameter initial values ​​by performing plane matching based on the ground plane extracted from the point cloud data of a preset number of low-beam radars.

[0094] The extrinsic parameter matrix generation module 430 is used to set the region of interest (ROI), perform iterative nearest point algorithm (ICP) optimization using a hierarchical search strategy to generate the extrinsic parameter matrix, and complete the automatic calibration of low-beam multi-LiDAR.

[0095] The specific details of each of the aforementioned low-beam multi-laser radar automatic calibration device modules have been described in detail in the corresponding low-beam multi-laser radar automatic calibration method, so they will not be repeated here.

[0096] It should be noted that although several modules or units of a low-beam multi-laser automatic calibration device 400 have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0097] Furthermore, in an exemplary embodiment of this disclosure, an electronic device capable of implementing the above-described method is also provided.

[0098] Those skilled in the art will understand that various aspects of the present invention can be implemented as systems, methods, or program products. Therefore, various aspects of the present invention can be specifically implemented as entirely hardware embodiments, entirely software embodiments (including firmware, microcode, etc.), or embodiments combining hardware and software aspects, collectively referred to herein as “circuit,” “module,” or “system.”

[0099] The following reference Figure 6 To describe an electronic device 500 according to such an embodiment of the present invention. Figure 6 The electronic device 500 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0100] like Figure 6 As shown, the electronic device 500 is manifested in the form of a general-purpose computing device. The components of the electronic device 500 may include, but are not limited to: at least one processing unit 510, at least one storage unit 520, a bus 530 connecting different system components (including storage unit 520 and processing unit 510), and a display unit 540.

[0101] The storage unit stores program code that can be executed by the processing unit 510, causing the processing unit 510 to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of the present invention. For example, the processing unit 510 can perform actions such as... Figure 1 Steps S110 to S130 are shown in the diagram.

[0102] Storage unit 520 may include a readable medium in the form of a volatile storage unit, such as random access memory (RAM) 5201 and / or cache memory 5202, and may further include a read-only memory (ROM) 5203.

[0103] Storage unit 520 may also include a program / utility 5204 having a set (at least one) program module 5205, such program module 5205 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0104] Bus 530 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0105] Electronic device 500 can also communicate with one or more external devices 570 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 500, and / or with any device that enables electronic device 500 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 550. Furthermore, electronic device 500 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 560. As shown, network adapter 560 communicates with other modules of electronic device 500 via bus 530. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 500, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0106] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0107] In exemplary embodiments of this disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the methods described above is stored. In some possible embodiments, various aspects of the invention may also be implemented as a program product comprising program code that, when the program product is run on a terminal device, causes the terminal device to perform the steps of the various exemplary embodiments of the invention described in the "Exemplary Methods" section above.

[0108] refer to Figure 7 As shown, a program product 600 for implementing the above-described method according to an embodiment of the present invention is described. It may employ a portable compact disc read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, the readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0109] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0110] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0111] The program code contained on the readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0112] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0113] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0114] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0115] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. An automatic calibration method for low-beam multi-laser radar, characterized in that, The method includes: After preprocessing the static point cloud data collected by a preset number of low-beam radars, ground point cloud extraction is performed based on the improved RANSAC algorithm to generate ground planes. Based on the ground plane extracted from the point cloud data of a preset number of low-beam radars, plane matching is performed to solve for the initial values ​​of the extrinsic parameters. Set the Region of Interest (ROI), use a hierarchical search strategy to perform iterative nearest point algorithm (ICP) optimization to generate the extrinsic parameter matrix, and complete the automatic calibration of low-beam multi-LiDAR.

2. The method as described in claim 1, characterized in that, The method includes preprocessing the static point cloud data collected by a preset number of low-beam radars, including: Preprocess and overlay multiple frames of static point cloud data collected by a preset number of low-beam radars. Statistical filtering is applied to the superimposed point cloud to remove outlier noise points.

3. The method as described in claim 2, characterized in that, The improved RANSAC algorithm in the method includes: The distance threshold of the interior points is dynamically adjusted by dynamically reducing the distance threshold according to the proportion of the interior points. Change random sampling to sampling from the 20% of point clouds with the lowest elevation according to weights.

4. The method as described in claim 3, characterized in that, The method further includes: Ground point cloud extraction is performed based on the improved RANSAC algorithm. Normal vector constraints are applied to the extracted planes, and only planes with an angle of less than 10 degrees between the normal vector and the gravity direction are retained as effective planes to generate ground planes.

5. The method as described in claim 1, characterized in that, The method further includes: By calculating the normal vectors of a preset number of ground planes, the X-axis rotation and Y-axis rotation between radars are solved. The Z-axis translation is obtained by calculating the vertical displacement of a preset number of ground planes.

6. The method as described in claim 1, characterized in that, The hierarchical search strategy used in the method for iterative nearest point (ICP) algorithm optimization also includes: Coarse search layer: Within the preset range of the initially estimated X-axis translation, Y-axis translation, and Z-axis rotation parameters, a binary search method is used for fast search, and the root mean square error of ICP matching is used as an indicator to quickly lock in the approximate range of these parameters. Fine search layer: Based on the coarse search results, a non-convex optimization objective function is constructed, which is composed of the weighted superposition of error terms such as RMSE of ICP matching, point cloud overlap, and edge alignment error. Through iterative optimization, the optimal values ​​of X and Y translation and Z-axis rotation parameters are finally solved.

7. The method as described in claim 1, characterized in that, The method further includes: By applying the extrinsic parameter matrix to the point cloud of the radar to be calibrated and projecting it onto the main radar coordinate system, error matching is performed based on a preset threshold to evaluate the calibration results of the extrinsic parameter matrix. Generate a multi-view point cloud visualization interface for users to intuitively verify the alignment of point clouds and realize the visualization verification of the calibration results of the external parameter matrix.

8. An automatic calibration device for low-beam multi-laser radar, characterized in that, The device includes: The ground plane extraction module is used to preprocess a preset number of static point cloud data collected by low-beam radar, and then extract the ground point cloud data based on the improved RANSAC algorithm to generate the ground plane. The extrinsic parameter initial value solution module is used to perform plane matching to solve for the extrinsic parameter initial values ​​based on the ground plane extracted from the point cloud data of a preset number of low-beam radars. The extrinsic parameter matrix generation module is used to set the region of interest (ROI), perform iterative nearest point algorithm (ICP) optimization using a hierarchical search strategy to generate the extrinsic parameter matrix, and complete the automatic calibration of low-beam multi-LiDAR.

9. An electronic device, characterized in that, include Processor; and A memory storing computer-readable instructions that, when executed by the processor, implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images