Lidar-based positioning method, system, device, and storage medium

Abstract

The application provides a laser radar-based positioning method, system, device and storage medium, and the method comprises the following steps: acquiring laser radar point cloud data of a carrier from a laser radar; extracting a laser radar factor based on the laser radar point cloud data; selecting points with reflectivity higher than a preset reflectivity threshold as high-reflectivity points based on the laser radar point cloud data; calculating a current frame prediction position based on the high-reflectivity points as a high-reflectivity factor based on the point cloud features of the high-reflectivity points and a pre-constructed point cloud map; and performing factor graph optimization based on the high-reflectivity factor and the laser radar factor to obtain positioning data of the carrier. The application increases the high-reflectivity factor, improves the effective constraint in the forward direction of the carrier, and effectively improves the robustness of laser radar positioning.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a positioning method, system, device and storage medium based on lidar. Background Technology

[0002] With the rapid development and maturation of autonomous driving on the ground, unmanned subway operation is an inevitable trend in the future. Positioning technology is a key component in realizing subway collision avoidance warnings and automatic driving. However, in the enclosed underground environment, GPS (Global Positioning System) signals cannot be used for positioning. Positioning technologies and mapping methods based on WiFi, Bluetooth, Radio Frequency Identification (RFID), Ultra Wideband (UWB), and ultrasound also require the installation of corresponding auxiliary positioning equipment in the underground environment. While this can improve positioning accuracy, it requires significant equipment and maintenance costs. Furthermore, the harsh underground environment easily leads to cumulative errors during the positioning process.

[0003] With the development of computer performance and related optimization algorithms, Simultaneous Localization and Mapping (SLAM) technology has emerged. However, the dust and poor lighting conditions in underground environments lead to unstable feature point extraction, posing numerous challenges to purely vision-based SLAM solutions.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] To address the problems in the prior art, the present invention aims to provide a positioning method, system, device, and storage medium based on lidar. By increasing the high reflectivity factor, the effective constraint in the forward direction of the carrier is improved, which helps to enhance the robustness of lidar positioning.

[0006] This invention provides a positioning method based on lidar, comprising the following steps:

[0007] Acquire lidar point cloud data of the carrier from lidar;

[0008] LiDAR factors are extracted based on the LiDAR point cloud data.

[0009] Based on the lidar point cloud data, points with reflectivity higher than a preset reflectivity threshold are selected as high reflectivity points.

[0010] Based on the point cloud features of the high reflectivity points and the pre-constructed point cloud map, the current frame prediction position based on the high reflectivity points is calculated as the high reflectivity factor.

[0011] Based on the high reflectivity factor and the lidar factor, factor graph optimization is performed to obtain the positioning data of the carrier.

[0012] In some embodiments, the factor map optimization based on the high reflectivity factor and the lidar factor includes the following steps:

[0013] Input the high reflectivity factor and the lidar factor into the residual block;

[0014] Obtain the positioning data of the carrier output by the residual block.

[0015] In some embodiments, after acquiring the lidar point cloud data of the carrier from the lidar, the following steps are further included:

[0016] The position and angle parameters of the carrier are obtained based on the lidar point cloud data.

[0017] When inputting the high reflectivity factor and the lidar factor into the residual block, the method also includes inputting the position parameters and angle parameters of the carrier into the residual block.

[0018] In some embodiments, after inputting the high reflectivity factor and lidar factor into the residual block, the following steps are further included:

[0019] A cost function is constructed using the current frame predicted position based on the high reflectivity point, as well as the position and angle parameters of the carrier.

[0020] In some embodiments, based on the point cloud features of the high reflectivity points and a pre-built point cloud map, the predicted position of the current frame based on the high reflectivity points is calculated, including the following steps:

[0021] Construct planar point cloud features based on the high reflectivity points;

[0022] Calculate the distance from the high reflectivity point in the current frame to the nearest point in the point cloud map, and optimize the distance iteratively to achieve point cloud feature matching between the high reflectivity point and the point cloud map;

[0023] The predicted position of the current frame is calculated based on the point cloud feature matching results and high reflectivity points.

[0024] In some embodiments, extracting lidar factors based on the lidar point cloud data includes the following steps:

[0025] Selecting LiDAR point cloud data for constructing point cloud maps based on a sliding window and edge-factoring of older LiDAR point cloud data;

[0026] Construct a point cloud map based on the selected lidar point cloud data;

[0027] The point cloud data of the current frame is matched with the point cloud map for point cloud features. Based on the matching result, the predicted position of the current frame based on the LiDAR measurement data is output as the LiDAR factor.

[0028] In some embodiments, before constructing the point cloud map based on the selected lidar point cloud data, the following steps are also included:

[0029] Based on the selected lidar point cloud data, all points are divided into foreground points, background points, and ground points;

[0030] Set a time window with a preset duration;

[0031] Determine whether there exists a point in the foreground points that is not continuously updated within a time range smaller than the time window;

[0032] If so, treat the point as a dynamic object point and filter it out.

[0033] In some embodiments, constructing a point cloud map based on selected lidar point cloud data includes constructing the point cloud map into a kd-tree format based on the selected lidar point cloud data.

[0034] The step of matching the point cloud features of the current frame's lidar point cloud data with the point cloud map includes the following steps:

[0035] Establish Scan-Context, find the nearest neighbor result in the kd-tree form point cloud map, calculate the statistical score based on the search result, obtain the best matching point cloud position, complete loop closure detection, and output the LiDAR factor based on the best matching result.

[0036] This invention also provides a lidar-based positioning system for implementing the lidar-based positioning method, the system comprising:

[0037] The data acquisition module is used to acquire point cloud data of the carrier from the lidar.

[0038] The first factor extraction module is used to extract lidar factors based on the lidar point cloud data.

[0039] The second factor extraction module is used to select points with reflectivity higher than a preset reflectivity threshold based on the lidar point cloud data as high reflectivity points, and to calculate the current frame prediction position based on the point cloud features of the high reflectivity points and the pre-constructed point cloud map as a high reflectivity factor.

[0040] The carrier positioning module is used to perform factor map optimization based on the high reflectivity factor and the lidar factor to obtain the positioning data of the carrier.

[0041] This invention also provides a positioning device based on lidar, comprising:

[0042] processor;

[0043] A memory in which executable instructions of the processor are stored;

[0044] The processor is configured to execute the steps of the lidar-based positioning method by executing the executable instructions.

[0045] This invention also provides a computer-readable storage medium for storing a program, which, when executed by a processor, implements the steps of the lidar-based positioning method.

[0046] 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.

[0047] The lidar-based positioning method, system, device, and storage medium of the present invention have the following beneficial effects:

[0048] This invention first obtains the lidar factor and the high reflectivity factor, then uses both factors for factor map optimization to obtain the carrier's positioning data. By increasing the high reflectivity factor, the effective constraint in the carrier's forward direction is improved, effectively enhancing the robustness of lidar positioning. This invention can be applied to subway tunnel scenarios, such as for subway self-positioning, and can be used as a basis for collision avoidance control, providing the relative velocity of static objects in the environment using millimeter-wave radar. This invention can also be applied to other mapping and positioning scenarios, such as streets, residential areas, and enclosed spaces. Attached Figure Description

[0049] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0050] Figure 1 This is a flowchart of a positioning method based on lidar according to an embodiment of the present invention;

[0051] Figure 2 This is a single-frame lidar point cloud scanning image of a tunnel according to an embodiment of the present invention;

[0052] Figure 3 This is a flowchart of an embodiment of the present invention for extracting lidar factors based on lidar point cloud data;

[0053] Figure 4 This is a flowchart of filtering dynamic objects according to an embodiment of the present invention;

[0054] Figure 5 This is a flowchart illustrating the calculation of the current frame prediction position based on high reflectivity points, according to an embodiment of the present invention.

[0055] Figure 6 This is a schematic diagram of a positioning system based on lidar according to an embodiment of the present invention;

[0056] Figure 7 This is a schematic diagram of the structure of a positioning device based on lidar according to an embodiment of the present invention;

[0057] Figure 8 This is a schematic diagram of the structure of a computer-readable storage medium according to an embodiment of the present invention. Detailed Implementation

[0058] 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 examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0059] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0060] The flowchart shown in the attached diagram is merely an illustrative example and does not necessarily include all steps. For example, some steps may be broken down, while others may be combined or partially combined. Therefore, the actual execution order may change depending on the specific circumstances.

[0061] like Figure 1As shown, this embodiment of the invention provides a positioning method based on lidar, including the following steps:

[0062] S100: Acquires lidar point cloud data of the carrier from the lidar, which is mounted on the carrier and moves with the carrier to collect lidar point cloud data in real time.

[0063] S200: Extract lidar factors based on the lidar point cloud data;

[0064] S300: Select points with reflectivity higher than a preset reflectivity threshold based on the lidar point cloud data, and use them as high reflectivity points;

[0065] The preset reflectivity threshold value can be selected by the staff as needed. When the value is set higher, the number of high reflectivity points selected will be less, and when the value is set lower, the number of high reflectivity points selected will be more. For example, the preset reflectivity threshold can be selected as 20, 15, etc.

[0066] S400: Based on the point cloud features of the high reflectivity points and the pre-built point cloud map, calculate the current frame prediction position based on the high reflectivity points as the high reflectivity factor;

[0067] S500: Based on the high reflectivity factor and the lidar factor, perform factor map optimization to obtain the positioning data of the carrier.

[0068] In the lidar-based positioning method of this embodiment, the sequence number of each step is only for distinguishing the steps and is not a limitation on the specific execution order of the steps. The execution order between the above steps can be adjusted and changed as needed. For example, steps S200 and S300 can be performed simultaneously, or steps S200 and S400 can be performed simultaneously, or steps S300 and S400 can be performed first, followed by step S200, etc., all of which can achieve the purpose of the present invention.

[0069] Compared to visual sensors, LiDAR provides more robust, accurate, and noise-stable measurement information, and is insensitive to changes in lighting conditions. LiDAR SLAM is the most stable and reliable SLAM solution. In railway scenarios, oncoming trains create dynamic objects over extended periods, posing challenges to object filtering of the input point cloud. Therefore, this invention optimizes the LiDAR-based localization method.

[0070] Considering the degradation of the spatial structure of point cloud features in the forward direction in railway tunnel scenarios, which leads to a decrease in the prediction accuracy of the forward position of the lidar odometer, high reflectivity points can provide effective constraints in the forward direction of the train. Figure 2This is a single-frame LiDAR point cloud scanning image of a tunnel according to an embodiment of the present invention. The colors are categorized according to reflectivity, revealing high-reflectivity target objects (marked as white dots) on both sides of some sections of the tunnel. These are generally reflectors or markers, typical of tunnel infrastructure. Specifically, in the positioning method of the present invention, LiDAR factors are first obtained through steps S100 and S200, then high reflectivity factors are obtained through steps S300 and S400. In step S500, the LiDAR factors and high reflectivity factors are used together for factor map optimization to obtain the positioning data of the carrier. By increasing the high reflectivity factor, the effective constraint in the carrier's forward direction is improved, effectively enhancing the robustness of LiDAR positioning.

[0071] When applying this lidar-based positioning method to locate subway trains in railway tunnels, the carrier is the subway train itself. The resulting positioning data for the carrier includes at least the train's position (coordinates along the x, y, and z axes) and its velocity values ​​along the x, y, and z axes. This method can also be applied to other scenarios where the carrier is a vehicle or other object, such as a community patrol vehicle.

[0072] like Figure 3 As shown, in this embodiment, step S200: extracting lidar factors based on the lidar point cloud data includes the following steps:

[0073] S210: Selecting LiDAR point cloud data for constructing point cloud maps based on a sliding window and edge-difference approach for older LiDAR point cloud data;

[0074] In this embodiment, in order to ensure high performance in real time, old LiDAR frames are edged out for pose optimization instead of matching LiDAR frames with the global map. Using local matching instead of global matching can significantly improve the real-time performance of the system.

[0075] S220: Construct a point cloud map based on selected LiDAR point cloud data;

[0076] S230: Perform point cloud feature matching between the lidar point cloud data of the current frame and the point cloud map. Specifically, perform point cloud feature matching between the lidar point cloud features of the current frame and the point cloud map, and output the predicted position of the current frame based on the lidar measurement data as the lidar factor based on the matching result.

[0077] In this embodiment, the lidar point cloud features include surface feature points, line feature points, and irregular point features; where surface feature points refer to feature points corresponding to ultra-small curvature, line feature points refer to feature points corresponding to large curvature, and irregular point features refer to feature points corresponding to intermediate curvature.

[0078] In this embodiment, step S220, constructing a point cloud map based on the selected LiDAR point cloud data, includes constructing the point cloud map into a kd-tree format based on the selected LiDAR point cloud data. kd-tree, short for k-dimensional tree, is a tree-like data structure that stores instance points in k-dimensional space for fast retrieval.

[0079] In this embodiment, the transformation relationships, ground parameters, and fused positioning pose obtained from the feature registration algorithms of the current laser frame are used as constraint edges. An objective function for the error is established by constructing a covariance matrix, and solved using a nonlinear least squares algorithm to obtain the optimal estimate with the minimum overall error. The pose is optimized by employing a sliding window method and marginalizing older LiDAR scans, rather than matching the scans to the global map. Using scan matching at a local scale rather than a global scale significantly improves the system's real-time performance.

[0080] like Figure 4 As shown, in this embodiment, before step S220: constructing a point cloud map based on the selected lidar point cloud data, the method further includes filtering dynamic objects using the following steps:

[0081] S211: Based on the selected lidar point cloud data, all points are divided into foreground points, background points, and ground points; specifically, the lidar point cloud data can be classified into foreground points, background points, and ground points after Euclidean clustering.

[0082] S212: Set a time window with a preset time length. The duration of the preset time length can be selected and set as needed, for example, 3 seconds in a railway tunnel scenario, but this invention is not limited thereto.

[0083] S213: Determine whether there is a point in the foreground point that is not continuously updated within a time range smaller than the time window;

[0084] S214: If so, treat the point as a dynamic object point and filter out the dynamic object point. That is, if a voxel position in the global map cannot be continuously updated within a time window less than the threshold, then the point is regarded as a dynamic object, thereby eliminating the foreground points that are dynamic objects.

[0085] S215: If not, there are no dynamic object points, no need to filter out dynamic objects, continue to step S230.

[0086] Therefore, in the method of this embodiment, dynamic object points in the foreground can be filtered out accurately and quickly, avoiding adverse effects of dynamic objects on positioning and effectively reducing interference from dynamic objects.

[0087] Step S230: Matching the point cloud features of the current frame's lidar point cloud data with the point cloud map, including the following steps:

[0088] Establish a Scan-Context, search for the nearest neighbor in the kd-tree form of the point cloud map (specifically, search in the KeyFrame generated during the point cloud feature matching process, and use Ring-key to search for the nearest neighbor under the kd-tree), calculate the statistical score based on the search results, obtain the best matching point cloud position, complete loop closure detection, and output the LiDAR factor based on the best matching result.

[0089] Scan-Context is a structure-information-based global localization method that directly records the 3D structural information of point clouds without relying on the full set of histogram descriptors or machine learning methods. This invention constructs loop closure detection based on Scan-Context, a non-histogram global descriptor, enabling faster and more efficient searching for loops and optimizing map construction. By using scan-match at a local scale rather than a global scale, the real-time performance of the system can be significantly improved.

[0090] like Figure 5 As shown, in this embodiment, step S400: calculating the current frame predicted position based on the point cloud features of the high reflectivity points and the pre-built point cloud map, includes the following steps:

[0091] S410: Construct planar point cloud features based on the high reflectivity points;

[0092] S420: Calculate the distance from the high reflectivity point in the current frame to the nearest point in the point cloud map based on the planar point cloud features, and achieve point cloud feature matching between the high reflectivity point and the point cloud map by iteratively optimizing the distance;

[0093] S430: Calculate the predicted position of the current frame based on high reflectivity points based on point cloud feature matching results.

[0094] In this embodiment, various factors were optimized using the Ceres library, and the final positioning data of the carrier was published. Step S500: Factor graph optimization was performed based on the high reflectivity factor and the lidar factor, including the following steps:

[0095] Input the high reflectivity factor and the lidar factor into the addresidualblock;

[0096] Obtain the positioning data of the carrier output by the residual block.

[0097] Ceres is a nonlinear optimization library widely used for solving SLAM problems, including Basic Algorithm (BA) problems. SLAM (Simultaneous Localization and Mapping) primarily enables a vehicle to perform localization, mapping, and path planning in unknown environments. Solving nonlinear optimization problems using Ceres generally involves three parts: 1. Constructing the cost function, which is the objective function for optimization. 2. Constructing the optimization problem to be solved using the cost function. 3. Configuring the solver parameters and solving the problem; this step involves setting how the equations are solved and whether the solution process is output.

[0098] The parameters for AddResidualBlock are as follows:

[0099] ResidualBlockId Problem::AddResidualBlock(CostFunction*,LossFunction*,const vector<double*> parameter_blocks);

[0100] AddResidualBlock() is a method under Problem, which adds residual blocks to the Problem object as needed. The residual block has three parameters: cost_function (the cost function); loss_function (usually nullptr, defaulting to 1); and the third is the variable to be optimized, which in this example is the VBIas state parameter.

[0101] The odometer in this embodiment contains five state parameters: position P, velocity V, angle R, IMU acceleration deviation Ba, and IMU angular velocity deviation Bg. Each state has three dimensions. The VBias state parameters include velocity V, IMU acceleration deviation Ba, and IMU angular velocity deviation Bg. These three states have a total of nine dimensions. The VBias state parameters can also be used as a factor and passed into the residual block AddResidualBlock for optimization.

[0102] In this embodiment, after step S100: acquiring the lidar point cloud data of the carrier from the lidar, the following steps are also included:

[0103] The position and angle parameters of the carrier are obtained based on the lidar point cloud data.

[0104] In this embodiment, after step S500, which involves inputting the high reflectivity factor and lidar factor into the residual block, the following steps are also included:

[0105] Using the current frame predicted position based on the high reflectivity point, as well as the position and angle parameters of the carrier, the residual of the current frame predicted position based on the high reflectivity point, as well as the position and angle parameters of the carrier, is calculated to construct a cost function.

[0106] In step S500, when inputting the high reflectivity factor and the lidar factor into the residual block, the position parameters and angle parameters of the carrier are also input into the residual block as other factors.

[0107] like Figure 6 As shown, this embodiment of the invention also provides a lidar-based positioning system for implementing the lidar-based positioning method described above. The system includes:

[0108] The data acquisition module M100 is used to acquire lidar point cloud data of the carrier from the lidar.

[0109] The first factor extraction module M200 is used to extract lidar factors based on the lidar point cloud data.

[0110] The second factor extraction module M300 is used to select points with reflectivity higher than a preset reflectivity threshold based on the lidar point cloud data as high reflectivity points, and to calculate the current frame prediction position based on the point cloud features of the high reflectivity points and the pre-constructed point cloud map as a high reflectivity factor.

[0111] The carrier positioning module M400 is used to perform factor map optimization based on the high reflectivity factor and the lidar factor to obtain the positioning data of the carrier.

[0112] In the positioning method of the present invention, the lidar factor is first obtained through the data acquisition module M100 and the first factor extraction module M200, and then the high reflectivity factor is obtained through the second factor extraction module M300. The lidar factor and the high reflectivity factor are used together by the carrier positioning module M400 to perform factor map optimization and obtain the positioning data of the carrier. By increasing the high reflectivity factor, the effective constraint in the forward direction of the carrier is improved, which effectively improves the robustness of lidar positioning.

[0113] When applying this lidar-based positioning method to locate subway trains in railway tunnels, the carrier is the subway train itself. The resulting positioning data for the carrier includes at least the train's position (coordinates along the x, y, and z axes) and its velocity values ​​along the x, y, and z axes. This method can also be applied to other scenarios where the carrier is a vehicle or other object, such as a community patrol vehicle.

[0114] In the lidar-based positioning system of the present invention, the functions of each module can be implemented using the specific implementation methods of the lidar-based positioning method described above, which will not be elaborated here.

[0115] This invention also provides a lidar-based positioning device, including a processor; a memory storing executable instructions of the processor; wherein the processor is configured to perform the steps of the lidar-based positioning method by executing the executable instructions.

[0116] 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 in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, collectively referred to herein as a "circuit," "module," or "platform."

[0117] The following reference Figure 7 To describe an electronic device 600 according to this embodiment of the present invention. Figure 7 The electronic device 600 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.

[0118] like Figure 7 As shown, the electronic device 600 is presented in the form of a general-purpose computing device. The components of the electronic device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including storage unit 620 and processing unit 610), a display unit 640, etc.

[0119] The storage unit stores program code that can be executed by the processing unit 610, causing the processing unit 610 to perform the steps described in the above-described section on the lidar-based positioning method according to various exemplary embodiments of the present invention. For example, the processing unit 610 can perform actions such as... Figure 1 The steps are shown in the figure.

[0120] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache storage unit 6202, and may further include a read-only memory unit (ROM) 6203.

[0121] The storage unit 620 may also include a program / utility 6204 having a set (at least one) program module 6205, such program module 6205 including but not limited to: an 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.

[0122] Bus 630 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.

[0123] Electronic device 600 can also communicate with one or more external devices 700 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 600, and / or with any device that enables electronic device 600 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 650. Furthermore, electronic device 600 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 660. Network adapter 660 can communicate with other modules of electronic device 600 via bus 630. 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 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0124] In the aforementioned lidar-based positioning device, when the program in the memory is executed by the processor, it implements the steps of the lidar-based positioning method. Therefore, the device can also achieve the technical effects of the lidar-based positioning method described above.

[0125] This invention also provides a computer-readable storage medium for storing a program that, when executed by a processor, implements the steps of the lidar-based positioning method. In some possible implementations, various aspects of the invention can also be implemented as a program product comprising program code that, when executed on a terminal device, causes the terminal device to perform the steps described in the lidar-based positioning method section of this specification according to various exemplary embodiments of the invention.

[0126] refer to Figure 8 As shown, a program product 800 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 be executed 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.

[0127] 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.

[0128] The computer-readable storage medium 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. The readable storage medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0129] 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).

[0130] When the program in the computer storage medium is executed by the processor, it implements the steps of the laser radar-based positioning method. Therefore, the computer storage medium can also achieve the technical effects of the laser radar-based positioning method.

[0131] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A positioning method based on lidar, characterized in that, When applied to tunnel scenarios, the method includes the following steps: Acquire lidar point cloud data of the carrier from lidar; Extracting lidar factors based on the lidar point cloud data includes: selecting lidar point cloud data for constructing a point cloud map based on a sliding window and edge-diffusion of old lidar point cloud data; constructing a point cloud map based on the selected lidar point cloud data; performing point cloud feature matching between the lidar point cloud data of the current frame and the point cloud map; and outputting the predicted position of the current frame based on lidar measurement data as the lidar factor based on the matching result. Based on the lidar point cloud data, points with reflectivity higher than a preset reflectivity threshold are selected as high reflectivity points, and the high reflectivity points correspond to reflectors or markers in the tunnel. Based on the point cloud features of the high reflectivity points and a pre-constructed point cloud map, the predicted position of the current frame based on the high reflectivity points is calculated as a high reflectivity factor. This includes: constructing planar point cloud features based on the high reflectivity points; calculating the distance from the high reflectivity points in the current frame to the nearest point in the point cloud map, and iteratively optimizing the distance to achieve point cloud feature matching between the high reflectivity points and the point cloud map; and calculating the predicted position of the current frame based on the point cloud feature matching result. Based on the high reflectivity factor and the lidar factor, factor graph optimization is performed to obtain the positioning data of the carrier.

2. The positioning method based on lidar according to claim 1, characterized in that, The factor graph optimization based on the high reflectivity factor and the lidar factor includes the following steps: Input the high reflectivity factor and the lidar factor into the residual block; Obtain the positioning data of the carrier output by the residual block.

3. The positioning method based on lidar according to claim 2, characterized in that, After acquiring the lidar point cloud data of the carrier from the lidar, the following steps are also included: The position and angle parameters of the carrier are obtained based on the lidar point cloud data. When inputting the high reflectivity factor and the lidar factor into the residual block, the method also includes inputting the position parameters and angle parameters of the carrier into the residual block.

4. The positioning method based on lidar according to claim 3, characterized in that, After inputting the high reflectivity factor and lidar factor into the residual block, the following steps are also included: A cost function is constructed using the current frame predicted position based on the high reflectivity point, as well as the position and angle parameters of the carrier.

5. The positioning method based on lidar according to claim 1, characterized in that, Before constructing the point cloud map based on the selected lidar point cloud data, the following steps are also included: Based on the selected lidar point cloud data, all points are divided into foreground points, background points, and ground points; Set a time window with a preset duration; Determine whether there exists a point in the foreground points that is not continuously updated within a time range smaller than the time window; If so, treat the point as a dynamic object point and filter it out.

6. The positioning method based on lidar according to claim 1, characterized in that, The construction of a point cloud map based on selected lidar point cloud data includes constructing the point cloud map into a kd-tree format based on the selected lidar point cloud data. The step of matching the point cloud features of the current frame's lidar point cloud data with the point cloud map includes the following steps: Establish Scan-Context, find the nearest neighbor result in the kd-tree form point cloud map, calculate the statistical score based on the search result, obtain the best matching point cloud position, complete loop closure detection, and output the LiDAR factor based on the best matching result.

7. A positioning system based on lidar, characterized in that, The system for implementing the lidar-based positioning method according to any one of claims 1 to 6, the system comprising: The data acquisition module is used to acquire point cloud data of the carrier from the lidar. The first factor extraction module is used to extract lidar factors based on the lidar point cloud data. The second factor extraction module is used to select points with reflectivity higher than a preset reflectivity threshold based on the lidar point cloud data as high reflectivity points, and to calculate the current frame prediction position based on the point cloud features of the high reflectivity points and the pre-constructed point cloud map as a high reflectivity factor. The carrier positioning module is used to perform factor map optimization based on the high reflectivity factor and the lidar factor to obtain the positioning data of the carrier.

8. A positioning device based on lidar, characterized in that, include: processor; A memory in which executable instructions of the processor are stored; The processor is configured to perform the steps of the lidar-based positioning method according to any one of claims 1 to 6 by executing the executable instructions.

9. A computer-readable storage medium for storing a program, characterized in that, When the program is executed by the processor, it implements the steps of the lidar-based positioning method as described in any one of claims 1 to 6.

Images