WO2023284705A1

WO2023284705A1 - Laser radar point cloud clustering method and apparatus, laser radar, and vehicle

Info

Publication number: WO2023284705A1
Application number: PCT/CN2022/105047
Authority: WO
Inventors: 郭鑫
Original assignee: 华为技术有限公司
Priority date: 2021-07-13
Filing date: 2022-07-12
Publication date: 2023-01-19
Also published as: CN115685224A

Abstract

A laser radar (10) point cloud clustering method and apparatus, a laser radar (10), and a vehicle. The method comprises: obtaining a first distance between a first point on an obstacle and the laser radar (10), a second distance between a second point on the obstacle and the laser radar (10), and included angle information between a laser beam corresponding to the first point on the obstacle and a laser beam corresponding to the second point on the obstacle; obtaining reference data from a pre-constructed clustering parameter table according to the first distance and the included angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle (S120, S130), wherein the clustering parameter table is constructed according to intrinsic parameters of the laser radar; and clustering the second point on the obstacle and the first point on the obstacle according to the reference data and the second distance. When clustering original point cloud data collected by the laser radar (10), a large amount of computing power is saved.

Description

A laser radar point cloud clustering method, device, laser radar and vehicle

This application claims the priority of the Chinese patent application submitted to the State Intellectual Property Office of China on July 13, 2021, with the application number 202110791099.1, and the title of the application is "A Lidar Point Cloud Clustering Method, Device, Lidar and Vehicle", The entire contents of which are incorporated by reference in this application.

technical field

The present application relates to the technical field of radar imaging in perception, and in particular to a laser radar point cloud clustering method, device, laser radar and vehicle.

Background technique

With the development of unmanned driving technology, higher requirements are put forward for the detection range and angular resolution of vehicle radar. The current vehicle-mounted radar (such as lidar) will form point cloud data when imaging a target. In order to distinguish multiple targets detected at the same time, it is first necessary to cluster the point cloud data collected by the vehicle-mounted radar. for clustering.

Traditional clustering algorithms have high requirements on chip computing power, so they are not suitable for scenarios where computing resources are scarce, such as on-board computing centers.

Contents of the invention

In view of the above problems in the prior art, the present application provides a laser radar point cloud clustering method, device, laser radar and vehicle, the clustering process is simple, and the effect of saving computing power is achieved.

In order to achieve the above purpose, the first aspect of the present application provides a laser radar point cloud clustering method, including:

Obtain the first distance between the first point on the obstacle and the lidar, the second distance between the second point on the obstacle and the lidar, and the laser beam and obstacle corresponding to the first point on the obstacle Angle information between the laser beams corresponding to the second point on the obstacle; according to the first distance, the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle , to obtain reference data from the pre-built clustering parameter table; wherein, the clustering parameter table is constructed according to the internal reference of the lidar; according to the reference data and the second distance, the second point on the obstacle and the first point on the obstacle Clustering at one point.

The first aspect of the present application provides a laser radar point cloud clustering method, which constructs a clustering parameter table through the internal parameters of the laser radar, takes into account the physical characteristics of the laser radar itself, and can make the clustering more accurate. Secondly, based on the pre-built clustering parameter table, the amount of calculation in actual clustering can be simplified, which is very suitable for scenarios where computing resources are scarce.

As a possible implementation manner of the first aspect, the first point and the second point are adjacent points.

As a possible implementation of the first aspect, the internal parameters of the lidar include: angle information between adjacent laser beams of the lidar and the number of beams of the lidar.

As a possible implementation of the first aspect, the clustering parameter table is constructed based on the internal parameters of the lidar, specifically including: obtaining the angle information between adjacent laser beams of the lidar and the detection multiple of the lidar; among them, the laser The detection multiple of the radar is determined according to the detection distance of the laser beam and the detection distance resolution of the laser radar; the index of the clustering parameter table is constructed with the angle information between adjacent laser beams as the first dimension and the detection multiple as the second dimension Information; obtain the reference data in the clustering parameter table based on the angle information between adjacent laser beams and the detection multiple.

From the above, by using the angle information between adjacent laser beams and the detection multiple of the laser radar as the index information of the clustering parameter table, the position of the point cloud collected by the laser radar can be accurately described, and the point cloud identification can be unique . In addition, the angle information between adjacent laser beams and the detection multiple of the lidar can be obtained directly from the lidar factory information without additional processing. Therefore, the clustering parameter table has the advantage of being simple to make.

As a possible implementation manner of the first aspect, the included angle information includes the size of the longitudinal included angle and the size of the horizontal included angle.

As a possible implementation of the first aspect, the number of elements in the first dimension is determined according to the number of beams of the laser radar.

As a possible implementation of the first aspect, the number of elements in the second dimension is determined according to the detection multiple of the lidar.

As a possible implementation of the first aspect, the detection multiple of the laser radar is determined according to the detection distance of the laser beam and the detection distance resolution of the laser radar, including: calculating the detection distance of the laser beam and the detection distance resolution of the laser radar , to get the detection multiple of lidar.

As a possible implementation manner of the first aspect, the reference data includes a first threshold and a second threshold, where the first threshold is greater than the second threshold.

As a possible implementation of the first aspect, the reference data in the clustering parameter table is calculated and obtained according to the angle information between adjacent laser beams and the detection multiple, including: determining according to the detection multiple and the detection distance resolution of the laser radar a third distance; determining the first threshold based on the third distance and a preset distance threshold; determining a second threshold based on the third distance, angle information between adjacent laser beams, and a preset angle threshold.

From the above, a method for determining the first threshold and the second threshold in the reference data is provided. This method takes into account the detection distance of the laser radar, the angle information between adjacent laser beams, etc., so that the calculation process is scientific and reasonable.

As a possible implementation of the first aspect, according to the first distance, the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle from the pre-built Obtain reference data in the class parameter table, including: obtain the first dimension according to the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle; determine according to the first distance The detection multiple corresponding to the first distance, and obtain the second dimension according to the detection multiple corresponding to the first distance; based on the first dimension and the second dimension, obtain the corresponding reference data from the pre-built clustering parameter table lookup table.

From the above, the specific process of querying the clustering parameter table is provided. The first dimension of the clustering parameter table is located through the angle information between the laser beam corresponding to the first point and the laser beam corresponding to the second point. Through the first The second dimension of the clustering parameter table is located by the detection multiple corresponding to the distance, and accordingly, the reference data corresponding to the first dimension and the second dimension can be obtained.

As a possible implementation of the first aspect, clustering the second point on the obstacle and the first point on the obstacle according to the reference data and the second distance includes: if the second distance is within the range of the reference data, Then the second point on the obstacle is of the same type as the first point on the obstacle; otherwise, the second point on the obstacle is not the same type as the first point on the obstacle.

The second aspect of the present application provides a lidar point cloud clustering device, including: a first acquisition module, a second acquisition module and a clustering module. Among them, the first acquisition module is used to acquire the first distance between the first point on the obstacle and the laser radar, the second distance between the second point on the obstacle and the laser radar, and the first distance between the second point on the obstacle and the laser radar. Angle information between the laser beam corresponding to one point and the laser beam corresponding to the second point on the obstacle; the second acquisition module is used for the laser beam corresponding to the first point on the obstacle and the obstacle according to the first distance The angle information between the laser beams corresponding to the second point on the object, and obtain reference data from the pre-built clustering parameter table; wherein, the clustering parameter table is constructed according to the internal parameters of the lidar; the clustering module uses The second point on the obstacle and the first point on the obstacle are clustered according to the reference data and the second distance.

As a possible implementation of the second aspect, the internal parameters of the lidar in the second acquisition module include: angle information between adjacent laser beams of the lidar and the number of beams of the lidar.

As a possible implementation of the second aspect, the clustering parameter table in the second acquisition module is constructed according to the internal parameters of the lidar, specifically including: an acquisition sub-module, used to acquire the gap between adjacent laser beams of the lidar The angle information and the detection multiple of the laser radar; wherein, the detection multiple of the laser radar is determined according to the detection distance of the laser beam and the detection distance resolution of the laser radar; the sub-module is constructed to use the angle information between adjacent laser beams as In the first dimension, the index information of the clustering parameter table is constructed with the detection multiple as the second dimension; the calculation sub-module is used to calculate and obtain the reference data in the clustering parameter table according to the angle information between adjacent laser beams and the detection multiple .

As a possible implementation manner of the second aspect, the included angle information includes the size of the longitudinal included angle and the size of the horizontal included angle.

As a possible implementation of the second aspect, the number of elements in the first dimension in the construction sub-module is determined according to the number of beams of the laser radar.

As a possible implementation of the second aspect, the number of elements in the second dimension in the construction sub-module is determined according to the detection multiple of the lidar.

As a possible implementation of the second aspect, the detection multiple of the laser radar is determined according to the detection distance of the laser beam and the detection distance resolution of the laser radar, including: the detection distance of the laser beam and the detection distance resolution of the laser radar are traded , to get the detection multiple of lidar.

As a possible implementation manner of the second aspect, the reference data in the calculation submodule includes a first threshold and a second threshold, where the first threshold is greater than the second threshold.

As a possible implementation of the second aspect, the calculation submodule is specifically used to: determine the third distance according to the detection multiple and the detection distance resolution of the laser radar; determine the first threshold based on the third distance and the preset distance threshold; The second threshold is determined by the third distance, angle information between adjacent laser beams and a preset angle threshold.

As a possible implementation of the second aspect, the second acquisition module is specifically configured to: according to the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle Obtain the first dimension; determine the detection multiple corresponding to the first distance according to the first distance, and obtain the second dimension according to the detection multiple corresponding to the first distance; based on the first dimension and the second dimension, check from the pre-built clustering parameter table Get the corresponding reference data from the table.

As a possible implementation of the second aspect, the clustering module is specifically configured to: if the second distance is within the range of the reference data, the second point on the obstacle is of the same type as the first point on the obstacle.

The third aspect of the present application provides a laser radar, including: a communication interface; at least one processor, which is connected to the communication interface; and at least one memory, which is connected to the processor and stores program instructions. When the program instructions are processed by at least one When the processor is executed, at least one processor is made to execute a lidar point cloud clustering method according to any one of the first aspect.

The fourth aspect of the present application provides a vehicle, including: a laser radar, used to collect point cloud data; a processor, used to use a laser radar point cloud clustering method according to any one of the first aspect point cloud data clustering.

The fifth aspect of the present application provides a computing device, a computer-readable storage medium, on which program instructions are stored, and it is characterized in that, when the program instructions are executed by a computer, the computer executes any one of the above-mentioned first aspects. Lidar point cloud clustering method.

The sixth aspect of the present application provides a computer program product, which is characterized in that, when the computer program product is run on the computing device, the computing device is made to execute a lidar point cloud clustering method according to any one of the above-mentioned first aspects.

These and other aspects of the present application will be made more apparent in the following description of the embodiment(s).

Description of drawings

The various features of the present application and the connections between the various features are further described below with reference to the accompanying drawings. The drawings are exemplary, some features are not shown to scale, and in some drawings, features customary in the field to which the application pertains and are not necessary for the application may be omitted, or additionally shown for the The application is not an essential feature, and the combination of the various features shown in the drawings is not intended to limit the application. In addition, in the whole specification, the content indicated by the same reference numeral is also the same. The specific accompanying drawings are explained as follows:

Figure 1(a) is the front view of the three-dimensional coordinate system constructed with the lidar as the origin;

Figure 1(b) is a side view of the three-dimensional coordinate system constructed with the lidar as the origin;

Figure 1(c) is a top view of the three-dimensional coordinate system constructed with the lidar as the origin;

FIG. 2 is a schematic diagram of a scene applied by a lidar point cloud clustering method provided in an embodiment of the present application;

FIG. 3 is a schematic diagram of another scene applied by a lidar point cloud clustering method provided in the embodiment of the present application;

FIG. 4 is a flowchart of a laser radar point cloud clustering method provided in an embodiment of the present application;

Fig. 5 is the flow chart of constructing the clustering parameter table provided by the embodiment of the present application;

Fig. 6 is the clustering parameter table provided by the embodiment of the present application;

FIG. 7 is a schematic diagram of a point cloud data grid provided by an embodiment of the present application;

FIG. 8 is a schematic structural diagram of a laser radar point cloud clustering device provided in an embodiment of the present application;

FIG. 9 is a schematic structural diagram of a computing device provided by an embodiment of the present application.

detailed description

The words "first, second, third, etc." or similar terms such as module A, module B, and module C in the description and claims are only used to distinguish similar objects, and do not represent a specific ordering of objects. It can be understood that Obviously, where permitted, the specific order or sequence can be interchanged such that the embodiments of the application described herein can be practiced in other sequences than those illustrated or described herein.

In the following description, the involved reference numerals representing steps, such as S110, S120, etc., do not mean that this step must be executed, and the order of the previous and subsequent steps can be interchanged or executed simultaneously if allowed.

The term "comprising" used in the description and claims should not be interpreted as being restricted to what is listed thereafter; it does not exclude other elements or steps. Therefore, it should be interpreted as specifying the presence of said features, integers, steps or components, but not excluding the presence or addition of one or more other features, integers, steps or components and groups thereof. Therefore, the expression "apparatus comprising means A and B" should not be limited to an apparatus consisting of parts A and B only.

Reference in this specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Thus, appearances of the phrases "in one embodiment" or "in an embodiment" in various places in this specification do not necessarily all refer to the same embodiment, but may. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to one of ordinary skill in the art from this disclosure, in one or more embodiments.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs. In case of any inconsistency, the meaning stated in this manual or the meaning derived from the content recorded in this manual shall prevail. In addition, the terms used herein are only for the purpose of describing the embodiments of the application, and are not intended to limit the application.

In order to accurately describe the technical content of this application, and in order to accurately understand this application, the following explanations or definitions are given to the terms used in this specification before describing the specific implementation methods:

1) Lidar (Lidar): A radar system that emits laser beams to detect target-related feature quantities. Its working principle is to send a detection signal (laser beam) to the target, and then compare the received signal reflected from the target with the transmitted detection signal, and after proper processing, the relevant information of the target can be obtained, such as the distance of the target, Azimuth, altitude, etc.

2) The longitudinal angle, that is, the pitch angle (pitch): as shown in Figure 1, Figure 1 is a three-dimensional coordinate system constructed with the lidar as the origin, where Figure 1(a) is the main view, and Figure 1(b) It is a side view, and Figure 1(c) is a top view. Point p is a point detected by the lidar. When point p is projected onto the X-Y plane, the angle between the projection line and the Y axis is the longitudinal angle, which is α in Figure 1.

3) Horizontal angle, that is, horizontal rotation angle (azimuth) or yaw angle (yaw): as shown in Figure 1, when point p is projected onto the X-Y plane, the angle between the projection line and the laser beam from the origin of the radar to point p angle, which is ω in Figure 1.

4) Number of beams: that is, the number of laser lines that the laser radar can emit. For example: for a 16-line lidar, the corresponding number of bundles is 16; for a 64-line lidar, the corresponding number of bundles is 64; for an 8-line lidar, the corresponding number of bundles is 8, etc. .

5) Point cloud clustering: According to the characteristics, similarity or distance of the point cloud data points, the point cloud data points are grouped into several "classes" or "clusters".

Next, first analyze a method for clustering point cloud data based on lidar provided by related technologies:

The point cloud clustering method of this solution is as follows: First, convert the point cloud data collected by the lidar to the corresponding logical plane, wherein the logical plane is used to describe the origin of the lidar and the relationship between each point to be clustered Relative positional relationship; then, perform a breadth-first search on each point on the corresponding logical plane to find the same point of each point; finally output the clustering results.

In this solution, all links are calculated based on real-time online calculations, and the amount of calculation is very high, which consumes a lot of computing power for the precious on-board computing center, so it is not suitable for on-board computing.

The embodiments of the present application will be described in detail below with reference to the accompanying drawings. First, the application scenarios of a laser radar point cloud clustering method provided in the embodiments of the present application will be introduced.

A laser radar point cloud clustering method provided in an embodiment of the present application can be applied to an autonomous vehicle (AV, Autonomous Vehicle) or an intelligent driving vehicle. Its application scenario can be to cluster the point cloud data collected by the vehicle lidar, and then use the clustered data to perform subsequent calculations. For example, the subsequent calculations can be target recognition calculations.

Exemplarily, as shown in FIG. 2 , it is a scenario where a laser radar point cloud clustering method provided in the embodiment of the present application is applied. In this example, the lidar point cloud clustering method can be built into the lidar 10 entity. After the laser radar 10 collects point cloud data, the point cloud data is clustered in the laser radar 10 body, and then the clustered data is transmitted to the host terminal 20 through a protocol such as a network for subsequent processing.

Exemplarily, as shown in FIG. 3 , it is another scenario in which a lidar point cloud clustering method provided in the embodiment of the present application is applied. In this example, the laser radar point cloud clustering method can be placed on the host end 20 . The laser radar 10 transmits the collected original point cloud data to the host terminal 20, and the laser radar point cloud clustering method provided by the embodiment of the application is used at the host terminal 20 to cluster the original point cloud data to obtain the clustered The data are used for subsequent calculations.

It should be understood that Fig. 2 and Fig. 3 are only examples showing the scene where the lidar point cloud clustering method is applied, and the present application does not limit which component the lidar point cloud clustering method is stored in. In other applications In the scene, the storage location of the lidar point cloud clustering method can be changed arbitrarily according to actual needs.

A laser radar point cloud clustering method provided in an embodiment of the present application will be described in detail below with reference to each figure. In the embodiment of the present application, the laser radar is taken as an example for the following description.

As shown in FIG. 4 , it is a flow chart of the lidar point cloud clustering method provided by the embodiment of the present application. The process mainly includes steps S110-S130, each step will be introduced in sequence below:

S110: Obtain the first distance between the first point on the obstacle and the laser radar, the second distance between the second point on the obstacle and the laser radar, and the correspondence between the first point on the obstacle Angle information between the laser beam and the laser beam corresponding to the second point on the obstacle.

As an optional implementation, the first distance may be determined by the time difference between the emitted light emitted by the laser radar to the first point and the reflected light reflected from the first point back to the laser radar; similarly, the first distance The second distance can also be determined by the time difference between the emitted light emitted by the lidar to the second point and the reflected light reflected from the second point back to the lidar. Angle information between the laser beam corresponding to the first point and the laser beam corresponding to the second point can be obtained from the factory manual of the laser radar.

It should be understood that the included angle information includes longitudinal included angle information and horizontal included angle information. For example, for the velodyne 16-line mechanical lidar, it can emit 16 laser beams, and a pitch angle (longitudinal angle) can be formed between every two laser beams, but the angle of each pitch angle is different; for For the velodyne 16-line mechanical lidar, it also includes a fixed aimuth angle (horizontal angle).

Generally, adjacent points have a higher probability of being of the same class, therefore, the first point and the second point in this embodiment may be selected as adjacent points. Use one point as a reference point to perform clustering operations on the other point.

S120: According to the first distance obtained in step S110, the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle, from the pre-built Get the reference data in the cluster parameter table.

In this step, first, based on the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle, the clustering can be determined in the form of table lookup One dimension of the parameter table; then according to the lidar detection multiple corresponding to the first distance, the other dimension of the clustering parameter table can be determined in the form of table lookup; based on the determined two dimensions, the second point and The corresponding clustering reference data in the clustering parameter table when the first point is clustered.

In this embodiment, the clustering parameter table is constructed according to the internal reference of the lidar, wherein the internal reference of the lidar includes: angle information between adjacent laser beams of the lidar and the number of beams of the lidar. In this embodiment, the angle information includes the size of the longitudinal angle and the size of the horizontal angle; the number of beams of the laser radar is the number of laser beams that the laser radar can emit.

Specifically, the process of constructing the clustering parameter table may include steps S121-S123, as shown in FIG. 5 , the steps of constructing the clustering parameter table are introduced below:

S121: Obtain angle information between adjacent laser beams of the lidar and a detection multiple of the lidar.

In this step, the angle information between adjacent laser beams of the lidar may include two aspects of information: the first aspect is the positional relationship between the angles between adjacent laser beams, that is, the adjacent laser beams are longitudinal The angle is still a horizontal angle; the second aspect is the size of the angle. Generally, the angle information between adjacent laser beams has been set when the laser radar leaves the factory, and it only needs to be determined according to the model of the laser radar.

In this step, the detection multiple of the lidar refers to how many times the distance resolution of the lidar has been experienced from the origin of the lidar to each laser point. Generally, the detectable multiple of the lidar has been set when the lidar leaves the factory, and it only needs to be determined according to the model of the lidar. In order to facilitate the understanding of the concept of the detection multiple of the lidar, it can be said that the detection multiple of the lidar is determined according to the detection distance of the laser beam and the detection distance resolution of the laser radar. The detection multiple can be multiplied, and the detection distance under the detection multiple can be obtained. For example: the resolution of the laser radar is 1m, and the detection multiple to a certain laser point is 5, then the detection distance corresponding to the laser point is 1*5=5m.

S122: Construct index information of the clustering parameter table with angle information between adjacent laser beams as a first dimension and with the detection multiple as a second dimension. Wherein, the index information is used to index reference data in the clustering parameter table.

As an optional implementation manner, the first dimension may be used as a row index dimension, and the second dimension may be used as a column index dimension. As another optional implementation manner, the first dimension and the second dimension may also be interchanged, that is, the first dimension is used as a column index dimension, and the second dimension is used as a row index dimension. This application does not limit it.

In this embodiment, the number of elements in the first dimension may be determined according to the number of beams of the lidar. For example: for the velodyne 16-line mechanical lidar, the number of beams is 16, that is, the lidar can emit 16 laser beams, and each two laser beams form a longitudinal angle, then the lidar corresponds to 15 longitudinal angles Angle, velodyne16-line mechanical lidar has a fixed horizontal angle, therefore, for velodyne16-line mechanical lidar, the number of elements in the first dimension is 16.

In this embodiment, the number of elements in the second dimension is determined according to the detection multiple of the lidar. For example: for a laser radar, its detection resolution is 2cm, and the maximum detection multiple is 50, then the maximum detectable distance of the radar is 2*50=100cm. For this radar, the number of second-dimensional elements is determined by the maximum detection multiple of the lidar, so the number of second-dimensional elements is 50. It should be understood that the radar resolution and maximum detection multiple here are only virtual settings to help understand the number of elements in the second dimension, and do not represent limitations on lidar parameters.

S123: Calculate and obtain the reference data in the clustering parameter table according to the angle information between the adjacent laser beams and the detection multiple.

In this step, the reference data includes a first threshold and a second threshold, wherein the first threshold is greater than the second threshold. For the convenience of understanding, the upper limit data may be used to replace the first threshold, and the lower limit data may be used to replace the second threshold. It should be understood that the range data is constituted by the upper limit data and the lower limit data to cluster related points.

In this step, as an optional implementation manner, firstly, the detection multiple is multiplied by the resolution of the lidar to obtain the third distance. Then determine the upper limit data based on the third distance and a preset distance threshold; and determine the lower limit data based on the third distance, angle information between adjacent laser beams and a preset angle threshold.

As an optional implementation, the upper limit data upper can be determined as follows:

upper=ε+d ₁

As an optional implementation, the lower limit data floor can be determined as follows:

Among them, d ₁ =n*f, d ₁ is the detection distance corresponding to the point in the lidar point cloud, that is, the third distance, n is the detection multiple of the lidar, f is the detection distance resolution of the lidar, and ε is the preset Distance threshold, β is the angle between adjacent laser beams, θ is the preset angle threshold.

S130: According to the first distance, the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle, from the pre-built clustering parameter table Get reference data from.

In this step, when the second distance is within the upper and lower limits of the reference data, the second point on the obstacle is of the same type as the first point on the obstacle; correspondingly, when the second When the second distance is not within the upper and lower limits of the reference data, the second point on the obstacle is not of the same type as the first point on the obstacle.

Referring to FIG. 6-FIG. 7, the specific implementation of a laser radar point cloud clustering method provided in the embodiment of the present application will be described in detail.

The laser radar point cloud clustering method provided in the embodiment of the present application is clustering based on the clustering parameter table. Therefore, the construction process of the clustering parameter table is firstly introduced. It should be understood that the clustering parameter table may be pre-built and obtained.

As an optional implementation, the clustering parameter table may include two dimensions, the first dimension may be angle information between adjacent laser beams, specifically, the angle information may include the size of the horizontal angle and the size of the vertical angle. The second dimension may be the detection multiple of the laser radar, wherein the detection multiple of the laser radar is determined according to the detection distance of the laser beam and the detection distance resolution of the laser radar. For example, the maximum detectable distance of the lidar is 120m, and the detection distance resolution of the lidar is 1m, so the corresponding detection multiple of the lidar is any integer value from 0 to 120. In addition, both the angle information and the detection multiple of the lidar can be obtained from the lidar factory manual. For example: the first dimension of the velodyne 16-line mechanical radar includes 15 pitch angles and a fixed aimuth angle, that is, 16 data. Its second dimension includes 32500 data, therefore, for the velodyne 16-line mechanical radar, its clustering parameter table shares 16*32500 data (elements).

Specifically, an element a (reference data) in the clustering parameter table is composed of an upper limit data upper and a lower limit data floor, that is, a=(floor, upper). For the calculation process of floor and upper, reference may be made to the foregoing embodiments, which will not be repeated in this embodiment.

As shown in FIG. 6 , it is a schematic diagram of a clustering parameter table constructed based on the construction method of the above clustering parameter table. In this figure: the column index dimension is the angle information between adjacent laser beams, and the row index dimension is the detection multiple of the laser radar, where 0-m is the detection multiple of the laser radar, which can also be understood as m times the laser The detection distance resolution of the radar, the upper limit is the quotient of the maximum detectable distance and the distance resolution of the lidar; pitch ₁ -pitch _m is the corresponding longitudinal angle between adjacent laser beams, and aimuth ₁ -aimuth _w is The corresponding horizontal angle between adjacent laser beams; upper is the upper limit of the elements in the clustering parameter table, and floor is the lower limit of the elements in the clustering parameter table. It should be understood that the positions of the horizontal angle and the vertical angle in the clustering parameter table in this figure are just an example and do not constitute a limitation. In addition, the row and column indexes can also be interchanged, that is, the column index is the detection multiple of the lidar, and the row index is the angle information between adjacent laser beams.

Then, the process of clustering the lidar point cloud based on the clustering parameter table shown in Fig. 6 is introduced.

As shown in Figure 7, it is a schematic diagram of the point cloud data grid collected by the lidar (only part of the data is shown). In the figure, there are two adjacent points A1 and A2 around A. If A is regarded as the first point on the obstacle, that is, the reference point, the process of judging whether A1 and A2 are of the same type as A is as follows:

First, obtain the angle information between the reference point A and the points A1 and A2 to be clustered. That is, obtain the horizontal angle between the reference point A and the point A1 to be clustered and the size of the angle is x, and obtain the vertical angle between the reference point A and the point A2 to be clustered and the size of the angle is y . Based on the angle information between the reference point A and the points A1 and A2 to be clustered, corresponding column information can be determined from the clustering parameter table shown in FIG. 6 .

Then, the distance from the reference point A to the origin of the laser radar is converted into the detection multiple of the laser radar, and the row information corresponding to the reference point A can be determined from the clustering parameter table shown in FIG. 6 .

Finally, by judging whether the distance between the point A1 to be clustered and the point A2 to be clustered and the origin of the lidar is within the upper and lower limits of the corresponding elements, the relationship between the point A1 to be clustered and the point A2 to be clustered is determined. Whether the reference point A is of the same class. If the distance is within the upper and lower limits of the corresponding element, the point is of the same class as the reference point, otherwise, the two are not of the same class.

Another embodiment of the present application provides a radar point cloud clustering device, which may be implemented by a software system, may also be implemented by a hardware device, and may also be implemented by a combination of a software system and a hardware device.

It should be understood that FIG. 8 is only an exemplary structural diagram showing a lidar point cloud clustering device, and the present application does not limit the division of functional modules in the head pose measurement device. As shown in Figure 8, the lidar point cloud clustering device can be logically divided into multiple modules, each module can have different functions, and the functions of each module can be read and executed by the processor in the computing device memory Instructions in to achieve. Exemplarily, the radar point cloud clustering device includes a first acquisition module 710 , a second acquisition module 720 and a clustering module 730 . In an optional implementation manner, the lidar point cloud clustering apparatus is used to execute the content described in steps S110-S130 shown in FIG. 4 . Specifically, it may be: a first acquisition module 710, configured to acquire the first distance between the first point on the obstacle and the laser radar, and the second distance between the second point on the obstacle and the laser radar , and angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle. The second acquiring module 720 is configured to, according to the first distance, the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle, from Obtain reference data from a pre-built clustering parameter table; wherein, the clustering parameter table is constructed according to the internal reference of the lidar. A clustering module 730, configured to cluster the second point on the obstacle and the first point on the obstacle according to the reference data and the second distance.

Specifically, the internal parameters of the lidar in the second acquisition module 720 include: angle information between adjacent laser beams of the lidar and the number of beams of the lidar.

Specifically, the clustering parameter table in the second acquisition module 720 is constructed according to the internal reference of the lidar, and specifically includes: an acquisition submodule 721 , a construction submodule 722 and a calculation submodule 723 . Wherein, the obtaining sub-module 721 is used to obtain the angle information between adjacent laser beams of the lidar and the detection multiple of the laser radar; wherein, the detection multiple of the laser radar is based on the detection distance of the laser beam and the detection range resolution of the lidar is determined. The construction sub-module 722 is configured to construct the index information of the clustering parameter table with the angle information between the adjacent laser beams as the first dimension and the detection multiple as the second dimension. The calculation sub-module 723 is configured to calculate and obtain the reference data in the clustering parameter table according to the angle information between the adjacent laser beams and the detection multiple.

As an optional implementation manner, the included angle information includes the size of the longitudinal included angle and the size of the horizontal included angle.

As an optional implementation manner, the number of elements in the first dimension in the construction submodule 722 is determined according to the number of beams of the lidar.

As an optional implementation manner, the number of elements in the second dimension of 722 in the construction submodule is determined according to the detection multiple of the lidar.

As an optional implementation, the detection multiple of the laser radar is determined according to the detection distance of the laser beam and the detection distance resolution of the laser radar, including:

The detection distance of the laser beam is multiplied by the detection distance resolution of the laser radar to obtain the detection multiple of the laser radar.

As an optional implementation manner, the reference data in the calculation submodule 723 includes a first threshold and a second threshold, where the first threshold is greater than the second threshold.

As an optional implementation, the calculation submodule 723 is specifically configured to: determine a third distance according to the detection multiple and the detection range resolution of the lidar; based on the third distance and a preset distance The threshold determines the first threshold; and determines the second threshold based on the third distance, angle information between adjacent laser beams and a preset angle threshold. As an optional implementation manner, the second acquisition module 720 is specifically configured to: according to the difference between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle Obtain the first dimension from the angle information between them; determine the detection multiple corresponding to the first distance according to the first distance, and obtain the second dimension according to the detection multiple corresponding to the first distance; based on the For the first dimension and the second dimension, corresponding reference data are obtained from the pre-built clustering parameter table lookup table.

As an optional implementation manner, the clustering module 730 is specifically configured to: if the second distance is within the range of the reference data, the second point on the obstacle is The first point is the same class.

Wherein, for the specific implementation manner of each functional module in this embodiment, reference may be made to the introduction in the foregoing method embodiments, and details are not repeated in this embodiment.

Another embodiment of the present application provides a laser radar, which includes:

Communication Interface;

at least one processor connected to the communication interface; and:

At least one memory, which is connected to the processor and stores program instructions, and when the program instructions are executed by the at least one processor, the at least one processor executes the laser radar described in the above embodiments For the point cloud clustering method, this embodiment does not repeat the lidar point cloud clustering method.

Another embodiment of the present application provides a vehicle, which may include the laser radar and the processor in the previous embodiment. The processor is used to cluster the point cloud data collected by the laser radar by using the laser radar point cloud clustering method described in the above embodiment. This application will not repeat it.

FIG. 9 is a schematic structural diagram of a computing device 900 provided by an embodiment of the present application. The computing device 900 includes: a processor 910 , a memory 920 , and a communication interface 930 .

It should be understood that the communication interface 930 in the computing device 900 shown in FIG. 9 can be used to communicate with other devices.

Wherein, the processor 910 may be connected to the memory 920 . The memory 920 can be used to store the program codes and data. Therefore, the memory 920 may be a storage unit inside the processor 910, or an external storage unit independent of the processor 910, or may include a storage unit inside the processor 910 and an external storage unit independent of the processor 910. part.

Optionally, computing device 900 may further include a bus. Wherein, the memory 920 and the communication interface 930 may be connected to the processor 910 through a bus. The bus may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (Extended Industry Standard Architecture, EISA) bus or the like. The bus can be divided into address bus, data bus, control bus and so on.

It should be understood that, in this embodiment of the present application, the processor 910 may be a central processing unit (central processing unit, CPU). The processor can also be other general-purpose processors, digital signal processors (digital signal processor, DSP), application specific integrated circuits (Application specific integrated circuit, ASIC), off-the-shelf programmable gate matrix (field programmable gate Array, FPGA) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like. Alternatively, the processor 910 adopts one or more integrated circuits for executing related programs, so as to implement the technical solutions provided by the embodiments of the present application.

The memory 920 may include read-only memory and random-access memory, and provides instructions and data to the processor 910 . A portion of processor 910 may also include non-volatile random access memory. For example, processor 910 may also store device type information.

When the computing device 900 is running, the processor 910 executes the computer-executed instructions in the memory 920 to perform the operation steps of the above method.

It should be understood that the computing device 900 according to the embodiment of the present application may correspond to a corresponding body executing the methods according to the various embodiments of the present application, and the above-mentioned and other operations and/or functions of the modules in the computing device 900 are for realizing the present invention For the sake of brevity, the corresponding processes of the methods in the embodiments are not repeated here.

Those skilled in the art can appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the present application.

Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.

If the functions described above are realized in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present application or a part of the technical solution can be embodied in the form of software products, the computer software products are stored in a storage medium, and include several instructions to make a computer device (which can It is a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disc and other media that can store program codes. .

The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it is used to perform a method for clustering a laser radar point cloud. At least one of the described programs.

The computer storage medium in the embodiments of the present application may use any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (non-exhaustive list) of computer readable storage media include: electrical connections with one or more leads, portable computer disks, hard disks, 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 device, magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a data signal carrying computer readable program code in baseband or as part of a carrier wave. Such propagated data signals may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. .

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wires, optical cables, radio, etc., or any suitable combination of the foregoing.

Computer program code for performing the operations of the present application may be written in one or more programming languages or combinations thereof, including object-oriented programming languages—such as Java, Smalltalk, C++, and conventional Procedural Programming Language - such as "C" or a similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (such as via the Internet using an Internet service provider). connect).

Note that the above are only preferred embodiments and technical principles used in this application. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, all of which belong to protection scope of this application.

Claims

A laser radar point cloud clustering method is characterized in that, comprising:

Obtain the first distance between the first point on the obstacle and the laser radar, the second distance between the second point on the obstacle and the laser radar, and the laser corresponding to the first point on the obstacle Angle information between the beam and the laser beam corresponding to the second point on the obstacle;

According to the first distance, the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle, from the pre-built clustering parameter table Obtaining reference data; wherein, the clustering parameter table is constructed according to the internal reference of the lidar;

Clustering the second point on the obstacle and the first point on the obstacle according to the reference data and the second distance.
The method according to claim 1, wherein the internal reference of the lidar includes: angle information between adjacent laser beams of the lidar and the number of beams of the lidar.
The method according to claim 2, wherein the clustering parameter table is constructed according to the internal reference of the lidar, specifically comprising:

Obtain the angle information between adjacent laser beams of the lidar and the detection multiple of the laser radar; wherein, the detection multiple of the laser radar is distinguished according to the detection distance of the laser beam and the detection distance of the laser radar rate determination;

Constructing the index information of the clustering parameter table with the included angle information between the adjacent laser beams as the first dimension and the detection multiple as the second dimension;

The reference data in the clustering parameter table is calculated and obtained according to the angle information between the adjacent laser beams and the detection multiple.
The method according to any one of claims 1-3, wherein the included angle information includes the size of the longitudinal included angle and the size of the horizontal included angle.
The method according to claim 3, wherein the number of elements in the first dimension is determined according to the number of beams of the lidar.
The method according to claim 3, characterized in that the number of elements in the second dimension is determined according to the detection multiple of the lidar.
The method according to claim 3, wherein the detection multiple of the laser radar is determined according to the detection distance of the laser beam and the detection distance resolution of the laser radar, including:

The detection distance of the laser beam is multiplied by the detection distance resolution of the laser radar to obtain the detection multiple of the laser radar.
The method according to claim 3, wherein the reference data includes a first threshold and a second threshold, wherein the first threshold is greater than the second threshold.
The method according to claim 8, wherein said calculating and obtaining the reference data in the clustering parameter table according to the angle information between the adjacent laser beams and the detection multiple includes:

determining a third distance according to the detection multiple and the detection distance resolution of the lidar;

determining the first threshold based on the third distance and a preset distance threshold;

The second threshold is determined based on the third distance, angle information between adjacent laser beams and a preset angle threshold.
The method according to claim 3, characterized in that, according to the first distance, between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle Obtain reference data from the pre-built clustering parameter table, including:

Acquiring the first dimension according to angle information between a laser beam corresponding to a first point on the obstacle and a laser beam corresponding to a second point on the obstacle;

determining a detection multiple corresponding to the first distance according to the first distance, and acquiring the second dimension according to the detection multiple corresponding to the first distance;

Based on the first dimension and the second dimension, corresponding reference data is obtained from the pre-built clustering parameter table lookup table.
The method according to claim 1, wherein the second point on the obstacle and the first point on the obstacle are clustered according to the reference data and the second distance, include:

If the second distance is within the reference data range, the second point on the obstacle is of the same type as the first point on the obstacle.
A radar point cloud clustering device is characterized in that it comprises:

The first acquiring module is used to acquire the first distance between the first point on the obstacle and the laser radar, the second distance between the second point on the obstacle and the laser radar, and the Angle information between the laser beam corresponding to the first point and the laser beam corresponding to the second point on the obstacle;

The second acquisition module is configured to, according to the first distance, the angle information between the laser beam corresponding to the first point on the obstacle and the laser beam corresponding to the second point on the obstacle, from the previous Acquiring reference data in the clustering parameter table constructed; Wherein, the clustering parameter table is constructed according to the internal reference of the lidar;

A clustering module, configured to cluster the second point on the obstacle and the first point on the obstacle according to the reference data and the second distance.
The device according to claim 12, wherein the internal reference of the lidar in the second acquisition module includes: angle information between adjacent laser beams of the lidar and the number of beams of the lidar.
The device according to claim 12, wherein the clustering parameter table in the second acquisition module is constructed according to the internal reference of the lidar, specifically comprising:

The obtaining sub-module is used to obtain the angle information between adjacent laser beams of the laser radar and the detection multiple of the laser radar; wherein, the detection multiple of the laser radar is based on the detection distance of the laser beam and the The detection range resolution of lidar is determined;

Constructing a submodule for constructing index information of the clustering parameter table with the included angle information between the adjacent laser beams as the first dimension and the detection multiple as the second dimension;

The calculation sub-module is used to calculate and obtain the reference data in the clustering parameter table according to the angle information between the adjacent laser beams and the detection multiple.
The device according to any one of claims 12-14, wherein the included angle information includes a size of a longitudinal included angle and a size of a horizontal included angle.
The device according to claim 14, wherein the number of elements in the first dimension in the building sub-module is determined according to the number of beams of the lidar.
The device according to claim 14, wherein the number of elements in the second dimension in the construction submodule is determined according to the detection multiple of the lidar.
The device according to claim 14, wherein the detection multiple of the laser radar is determined according to the detection distance of the laser beam and the detection distance resolution of the laser radar, including:

The detection distance of the laser beam is multiplied by the detection distance resolution of the laser radar to obtain the detection multiple of the laser radar.
The device according to claim 14, wherein the reference data in the calculation submodule includes a first threshold and a second threshold, wherein the first threshold is greater than the second threshold.
The device according to claim 19, wherein the calculation submodule is specifically used for:

determining a third distance according to the detection multiple and the detection distance resolution of the lidar;

determining the first threshold based on the third distance and a preset distance threshold;

The second threshold is determined based on the third distance, angle information between adjacent laser beams and a preset angle threshold.
The device according to claim 14, wherein the second acquiring module is specifically used for:

Acquiring the first dimension according to angle information between a laser beam corresponding to a first point on the obstacle and a laser beam corresponding to a second point on the obstacle;

determining a detection multiple corresponding to the first distance according to the first distance, and acquiring the second dimension according to the detection multiple corresponding to the first distance;

Based on the first dimension and the second dimension, corresponding reference data is obtained from the pre-built clustering parameter table lookup table.
The device according to claim 12, wherein the clustering module is specifically used for:

If the second distance is within the reference data range, the second point on the obstacle is of the same type as the first point on the obstacle.
A laser radar, is characterized in that, comprises:

Communication Interface;

at least one processor connected to the communication interface; and

At least one memory, which is connected to the processor and stores program instructions, and when the program instructions are executed by the at least one processor, the at least one processor executes the program described in any one of claims 1-11. A LiDAR point cloud clustering method.
A vehicle, characterized in that it comprises:

Lidar, used to collect point cloud data;

A processor, configured to cluster the point cloud data collected by the laser radar by using the laser radar point cloud clustering method described in any one of claims 1-11.
A computer-readable storage medium on which program instructions are stored, wherein when the program instructions are executed by a computer, the computer executes the laser radar point described in any one of claims 1-11. cloud clustering method.
A computer program product, characterized in that, when the computer program product is run on a computing device, the computing device is made to execute the lidar point cloud clustering method according to any one of claims 1-11.