CN114966651A - Drivable region detection method, computer device, storage medium, and vehicle - Google Patents
Drivable region detection method, computer device, storage medium, and vehicle Download PDFInfo
- Publication number
- CN114966651A CN114966651A CN202210557993.7A CN202210557993A CN114966651A CN 114966651 A CN114966651 A CN 114966651A CN 202210557993 A CN202210557993 A CN 202210557993A CN 114966651 A CN114966651 A CN 114966651A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- ground
- point
- grid
- dynamic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/06—Systems determining position data of a target
- G01S13/42—Simultaneous measurement of distance and other co-ordinates
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/86—Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
- G01S13/867—Combination of radar systems with cameras
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Electromagnetism (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of vehicles, in particular to a travelable region detection method, computer equipment, a storage medium and a vehicle, and aims to solve the problem of accurately detecting travelable regions around the vehicle. The method comprises the steps of obtaining a three-dimensional point cloud in a vehicle driving environment; establishing a polar coordinate system corresponding to a three-dimensional coordinate system of the three-dimensional point cloud, generating a polar coordinate grid diagram, and projecting the three-dimensional point cloud to the polar coordinate grid diagram; performing dynamic and static detection on non-ground point clouds in the three-dimensional point clouds to obtain the point cloud category of each non-ground point cloud; determining non-collision points according to the non-ground point clouds projected into each grid; determining the boundary of a drivable area in the driving environment of the vehicle according to the position of the non-collision point, and determining the attribute of the boundary according to the point cloud type of the non-collision point and the relative height of the ground. By the method, not only can more accurate boundary positions be determined, but also the boundary attributes can be determined, and determination of safer and more reliable driving behaviors is facilitated.
Description
Technical Field
The invention relates to the technical field of vehicles, and particularly provides a travelable region detection method, computer equipment, a storage medium and a vehicle.
Background
At present, a conventional vehicle travelable area detection method mainly acquires a two-dimensional image around a vehicle through a vision sensor, converts environmental features of environmental points in the two-dimensional image into a three-dimensional space, and determines a travelable area around the vehicle according to the position of the environmental features in the three-dimensional space. Since the vision sensor cannot directly acquire the distance information between the vehicle and the surrounding points, the method cannot acquire a more accurate travelable area.
Accordingly, there is a need in the art for a new travelable area detection scheme that addresses the above-mentioned problems.
Disclosure of Invention
In order to overcome the above-mentioned drawbacks, the present invention has been made to provide a travelable region detection method, a computer device, a storage medium, and a vehicle that solve or at least partially solve the technical problem of how to more accurately detect a travelable region around the vehicle.
In a first aspect, the present invention provides a travelable region detection method, the method further comprising:
acquiring three-dimensional point cloud in a vehicle driving environment;
establishing a polar coordinate system corresponding to the three-dimensional coordinate system of the three-dimensional point cloud, generating a polar coordinate grid diagram according to the preset grid number, and projecting the three-dimensional point cloud to the polar coordinate grid diagram;
performing dynamic and static detection on non-ground point clouds in the three-dimensional point clouds to obtain point cloud categories of each non-ground point cloud, wherein the point cloud categories comprise dynamic point clouds and static point clouds;
determining non-collision points according to the non-ground point clouds projected into each grid;
determining the boundary of a drivable area in the vehicle driving environment according to the position of the non-collision point, and determining the attribute of the boundary according to the point cloud type of the non-collision point and the ground relative height.
In one embodiment of the method for detecting a travelable area, after the step of "performing dynamic and static detection on the non-ground point clouds in the three-dimensional point cloud to obtain the point cloud type of each non-ground point cloud", the method further includes correcting the point cloud types of the non-ground point clouds in the following manner:
clustering the non-ground point clouds according to the distance between the three-dimensional point clouds to obtain a plurality of point cloud clusters;
acquiring the dynamic and static types of each point cloud cluster;
correcting the point cloud category of the non-ground point cloud in the corresponding point cloud cluster according to the dynamic and static types;
and/or the step of determining non-crashable points from the non-ground point clouds projected into each grid specifically comprises:
and taking the non-ground point cloud corresponding to the non-ground point cloud projection point with the minimum pole diameter in each grid as a non-collision point.
In a technical solution of the above method for detecting a drivable area, "acquiring the moving and static types of each point cloud cluster" specifically includes:
counting the proportion of the dynamic point clouds in each point cloud cluster;
comparing the proportion of the dynamic point cloud with a preset proportion threshold;
if the proportion of the dynamic point cloud in the current point cloud cluster is larger than or equal to a preset proportion threshold value, the dynamic and static types of the current point cloud cluster are the dynamic point cloud cluster;
and if the proportion of the dynamic point cloud in the current point cloud cluster is smaller than a preset proportion threshold value, the dynamic and static types of the current point cloud cluster are static point cloud clusters.
In one technical scheme of the method for detecting the travelable area, the step of correcting the point cloud type of the non-ground point cloud in the corresponding point cloud cluster according to the dynamic and static types specifically comprises the following steps:
if the dynamic and static types of the point cloud cluster are dynamic point cloud clusters, correcting the static point cloud in the point cloud cluster into dynamic point cloud;
and if the dynamic and static types of the point cloud cluster are static point cloud clusters, not correcting the point cloud category of the non-ground point cloud in the point cloud cluster.
In one aspect of the above travelable region detection method, the method further includes acquiring a ground relative height of the non-collision point by:
clustering the non-ground point clouds according to the space between the three-dimensional point clouds to obtain a plurality of point cloud clusters;
and acquiring the ground relative height of each non-ground point cloud in the point cloud cluster to which the non-collision point belongs, and taking the maximum ground relative height as the ground relative height of the non-collision point.
In one aspect of the above drivable area detection method, the method further comprises determining a ground height of a projected grid coverage area of each non-ground point cloud by:
if the grid projected by the current non-ground point cloud contains ground point cloud projection points, taking the average value of the point cloud heights of all three-dimensional point clouds projected into the grid as the ground height of the coverage area of the current grid, and marking the grid projected by the current non-ground point cloud as a ground point grid;
and if the projected grid of the current non-ground point cloud does not contain the ground point cloud projected point, determining the ground height of the coverage area of the projected grid of the current non-ground point cloud according to the ground heights of the coverage areas of other grids around the grid.
In one embodiment of the above method for detecting a drivable area, "determining the ground height of a grid coverage area projected by a current non-ground point cloud according to the ground height of other grid coverage areas around the grid" specifically includes:
respectively judging whether each other grid contains ground point cloud projection points;
if at least one other grid comprises a ground point cloud projection point, acquiring the ground height of the coverage area of the other grid comprising the ground point cloud projection point, carrying out interpolation calculation on the ground height, taking the height value obtained by the interpolation calculation as the ground height of the coverage area of the current non-ground point cloud projected grid, and marking the current non-ground point cloud projected grid into a ground point grid;
and if each other grid does not contain the ground point cloud projection point, taking the average value of the ground heights of all the grid coverage areas of the ground points as the ground height of the grid coverage area projected by the current non-ground point cloud.
In a second aspect, there is provided a computer apparatus comprising a processor and a storage device adapted to store a plurality of program codes adapted to be loaded and run by the processor to perform the travelable region detection method of any of the above-described travelable region detection methods.
In a third aspect, there is provided a computer readable storage medium having stored therein a plurality of program codes adapted to be loaded and run by a processor to execute the travelable region detection method according to any one of the above-described travelable region detection methods.
In a fourth aspect, a vehicle is provided, comprising a computer arrangement according to the above-mentioned technical solution.
One or more technical schemes of the invention at least have one or more of the following beneficial effects:
in the technical scheme of the invention, the three-dimensional point cloud in the vehicle driving environment can be obtained, a polar coordinate system corresponding to the three-dimensional coordinate system of the three-dimensional point cloud is established, a polar coordinate grid diagram is generated according to the number of preset grids, and the three-dimensional point cloud is projected to the polar coordinate grid diagram; then performing dynamic and static detection on non-ground point clouds in the three-dimensional point clouds to obtain point cloud categories of each non-ground point cloud, wherein the point cloud categories comprise dynamic point clouds and static point clouds; determining non-collision points according to the non-ground point clouds projected into each grid; and finally, determining the boundary of a drivable area in the driving environment of the vehicle according to the position of the non-collision point, and determining the attribute of the boundary according to the point cloud category of the non-collision point and the ground relative height.
The three-dimensional point cloud is an echo signal reflected back to a vehicle by an environment point (such as obstacles such as pedestrians, motor vehicles and non-motor vehicles in the driving environment) in the driving environment after the vehicle emits electromagnetic waves to the driving environment, the position information of the three-dimensional point cloud comprises coordinates of an x axis, a y axis and a z axis in a three-dimensional coordinate system, the coordinate of the z axis represents the point cloud height of the three-dimensional point cloud, and the distance between the three-dimensional point cloud and the vehicle can be calculated through the coordinates of the x axis and the y axis. After the non-collision point is determined, the boundary of the travelable area in the vehicle traveling environment can be obtained according to the coordinates of the non-collision point on the x axis and the y axis in the three-dimensional coordinate system, and the distance between the three-dimensional point cloud and the vehicle can be accurately calculated according to the coordinates of the x axis and the y axis of the three-dimensional point cloud, so that the boundary of the travelable area obtained by the method can accurately represent the area range of the travelable area.
In addition, the point cloud type of the non-collision point can indicate whether the non-collision point is dynamic or static, the ground relative height of the non-collision point indicates the distance between the non-collision point and the ground, and the point cloud type of the non-collision point and the ground relative height are taken as the attributes of the boundary and are output along with the boundary of the drivable area, so that safer and more reliable driving behaviors can be determined, such as crossing or bypassing the area where the non-collision point is located, and the vehicle can be driven safely.
Drawings
The disclosure of the present invention will become more readily understood with reference to the accompanying drawings. As is readily understood by those skilled in the art: these drawings are for illustrative purposes only and are not intended to constitute a limitation on the scope of the present invention. Wherein:
FIG. 1 is a schematic diagram of a polar grid graph according to one embodiment of the present invention;
FIG. 2 is a schematic diagram of a polar grid graph according to another embodiment of the invention.
Fig. 3 is a flowchart illustrating main steps of a travelable region detection method according to an embodiment of the present invention;
FIG. 4 is a flow chart illustrating the main steps of a method for dynamic and static detection of point clouds according to an embodiment of the invention;
fig. 5 is a flow chart illustrating the main steps of a point cloud category correction method according to an embodiment of the present invention.
Detailed Description
Some embodiments of the invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and are not intended to limit the scope of the present invention.
In the description of the present invention, a "processor" may include hardware, software, or a combination of both. A module may comprise hardware circuitry, various suitable sensors, communication ports, memory, may comprise software components such as program code, or may be a combination of software and hardware. The processor may be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and/or signal processing functionality. The processor may be implemented in software, hardware, or a combination thereof. The computer readable storage medium includes any suitable medium that can store program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random-access memory, and the like. The term "a and/or B" denotes all possible combinations of a and B, such as a alone, B alone or a and B. The term "at least one A or B" or "at least one of A and B" means similar to "A and/or B" and may include only A, only B, or both A and B. The singular forms "a", "an" and "the" may include the plural forms as well.
Some terms to which the present invention relates are explained first.
Polar coordinates (polar coordinates) refer to a coordinate system consisting of a pole O, a polar axis Ox and a polar diameter in a plane, the polar axis Ox is a ray drawn from the pole O, the polar diameter refers to a distance from a certain point P to the pole O in the plane, the point P can be represented as P (ρ, θ), ρ represents the polar diameter of the point P, θ represents the polar angle of the point P, and the polar angle is an angle from the polar axis Ox to a line segment OP in a counterclockwise direction. In the embodiment of the present invention, the pole O of the polar coordinate system and the origin of the three-dimensional coordinate system of the three-dimensional point cloud may be the same point, and the point may be a certain point on the vehicle.
The polar coordinate grid diagram refers to a representation diagram of a polar coordinate formed by meshing a plane having a certain polar angle range in a polar coordinate system, and the plane can be divided into a plurality of grids through meshing. As shown in fig. 1, the method of mesh division in the embodiment of the present invention is to divide a plane into a plurality of grids of sector shapes from a polar axis Ox in a counterclockwise direction with a pole O as a starting point, and each grid corresponds to a certain polar angle range. For example, the polar angle range of the plane is 120 °, the number of the grids formed by dividing is 192, and the polar angle range corresponding to each grid is 0.625 °. And rotating in the counterclockwise direction, wherein the polar angle range corresponding to the first grid is 0-0.625 degrees, the polar angle range corresponding to the second grid is 0.625-1.25 degrees, and the like to obtain the angles corresponding to other grids. Further, as shown in fig. 2, in some embodiments, after the plane is divided into a plurality of grids of the sector shape starting from the polar axis Ox in the counterclockwise direction with the pole O as a starting point, the grid of each sector shape may be further subdivided into a plurality of grids in the order of the smaller to the larger pole diameters.
Next, a travelable area detection method embodiment of the present invention will be explained.
Referring to fig. 3, fig. 3 is a flow chart illustrating main steps of a travelable region detection method according to an embodiment of the present invention. As shown in fig. 3, the travelable region detection method in the embodiment of the present invention mainly includes the following steps S101 to S105.
Step S101: and acquiring the three-dimensional point cloud in the vehicle running environment.
The three-dimensional point cloud is three-dimensional data determined according to echo signals reflected back to the vehicle after the environment points in the vehicle driving environment receive electromagnetic waves sent by the vehicle, and the three-dimensional data comprises coordinates of the environment points in a three-dimensional coordinate system. In the embodiment of the present invention, the vehicle may transmit electromagnetic waves to an environment point in the driving environment through a Radar (Radar), and the like, wherein the Radar includes, but is not limited to, a Millimeter-wave Radar (Millimeter-wave Radar) and a Laser Radar (Laser Radar), and in a preferred embodiment, the Radar may be a Laser Radar.
Step S102: and establishing a polar coordinate system corresponding to the three-dimensional coordinate system of the three-dimensional point cloud, generating a polar coordinate grid diagram according to the preset number of grids, and projecting the three-dimensional point cloud to the polar coordinate grid diagram.
The polar coordinate system established by taking the origin of the three-dimensional coordinate system as the pole is the polar coordinate system corresponding to the three-dimensional coordinate system.
After the three-dimensional point clouds are projected to the polar coordinate grid diagram, the coordinates of the projection points of each three-dimensional point cloud in the polar coordinate grid diagram can be obtained according to the coordinate conversion relation between the polar coordinate system and the three-dimensional coordinate system, and the projection points falling into each grid can be obtained according to the coordinates of the projection points and the position information of each grid.
Step S103: and performing dynamic and static detection on non-ground point clouds in the three-dimensional point clouds to obtain the point cloud category of each non-ground point cloud, wherein the point cloud categories comprise dynamic point clouds and static point clouds.
When the non-ground point cloud in the three-dimensional point cloud is subjected to dynamic and static detection, a dynamic target detection model can be used for carrying out dynamic target detection on the non-ground point cloud, if a dynamic target is detected, the three-dimensional point cloud belonging to the dynamic target is the dynamic point cloud, and the three-dimensional point cloud not belonging to the dynamic target is the static point cloud. Dynamic objects include, but are not limited to, pedestrians, automobiles, and non-automobiles, among others. The dynamic target detection model may be a conventional model capable of performing target detection on the three-dimensional point cloud in the technical field of target detection, for example, a target detection model based on a neural network.
In some embodiments, the dynamic object detection model may include a feature extraction network and a feature detection network, as shown in fig. 4, in this embodiment, the non-ground point cloud in the three-dimensional point cloud may be subjected to dynamic and static detection through the following steps S201 to S204.
Step S201: and acquiring the three-dimensional point cloud.
Step S202: and carrying out voxelization processing on the three-dimensional point cloud to obtain a plurality of point cloud voxels.
In this embodiment, a conventional point cloud voxelization method may be adopted to perform voxelization on the three-dimensional point cloud to obtain a plurality of point cloud voxels, which is not described herein again.
Step S203: and respectively extracting the characteristics of each point cloud voxel by adopting a characteristic extraction network.
Step S204: and performing dynamic and static detection on the features by adopting a feature detection network to obtain the point cloud category of each point cloud.
In addition, in the embodiment of the present invention, a Principal Component Analysis (Principal Component Analysis) may be used to perform ground detection on the three-dimensional point cloud, so as to determine which three-dimensional point clouds are ground point clouds belonging to the ground and which three-dimensional point clouds are non-ground point clouds not belonging to the ground, and further perform dynamic and static detection on the non-ground point clouds. Specifically, the ground point cloud and the non-ground point cloud in the three-dimensional point cloud may be determined by the following steps S1031 to S1035.
Step S1031: and respectively determining the plane normal vector of each grid coverage area according to the three-dimensional point cloud projected into each grid by adopting a principal component analysis method.
Step S1032: and respectively judging whether the plane normal vector of each grid coverage area meets a preset normal vector constraint condition of the ground plane, wherein the normal vector constraint condition is that the angle deviation of the plane normal vector is smaller than a preset angle deviation threshold value. If the plane normal vector of the current grid coverage area meets the normal vector constraint condition, go to step S1033; if the plane normal vector of the current grid coverage area does not satisfy the normal vector constraint condition, go to step S1034.
Those skilled in the art can flexibly set a specific value of the preset angle deviation threshold, such as 15 °, according to actual requirements, which is not specifically limited in the embodiment of the present invention.
Step S1033: and judging whether the point cloud height of the three-dimensional point cloud projected into the current grid meets a preset height constraint condition of the ground plane, wherein the height constraint condition at least comprises that the average value and the variance of the point cloud height of the three-dimensional point cloud projected into the current grid are respectively smaller than the corresponding threshold values. If the height constraint condition is satisfied, go to step S1035; if the height constraint condition is not satisfied, go to step S1034.
Those skilled in the art can flexibly set the specific values of the threshold corresponding to each of the average value and the variance according to actual requirements, which is not specifically limited in the embodiment of the present invention.
Step S1034: and taking the three-dimensional point cloud projected into the current grid as non-ground point cloud.
Step S1035: and taking the three-dimensional point cloud projected into the current grid as the ground point cloud.
The above is a specific description of the method of determining the ground point cloud and the non-ground point cloud, and the following description is continued with step S104 and step S105 shown in fig. 1.
Step S104: and determining non-collision points according to the non-ground point clouds projected into each grid.
In the embodiment of the invention, the non-ground point cloud closest to the vehicle can be selected as the non-collision point, so that the area between the vehicle and the non-collision point is the area where the vehicle can run.
The position information of the three-dimensional point cloud comprises coordinates of an x axis, a y axis and a z axis in a three-dimensional coordinate system, the coordinate of the z axis represents the point cloud height of the three-dimensional point cloud, the distance between the three-dimensional point cloud and the vehicle can be calculated through the coordinates of the x axis and the y axis, and the non-collision point can be determined according to the distance.
Because the pole of the polar coordinate system is a certain point on the vehicle, the point with the minimum polar diameter is the point closest to the vehicle, and the point with the minimum polar diameter can be obtained to obtain the non-collision point. In particular, in some embodiments, the non-ground point cloud corresponding to the non-ground point cloud projection point with the smallest pole diameter within each grid may be taken as the non-collidable point.
Step S105: determining the boundary of a drivable area in the driving environment of the vehicle according to the position of the non-collision point, and determining the attribute of the boundary according to the point cloud type of the non-collision point and the relative height of the ground.
In this embodiment, the position of the non-collision point may be used as the boundary position of the travelable region, so as to determine an accurate boundary position. The point cloud type of the non-collision point can indicate whether the non-collision point is dynamic or static, the ground relative height of the non-collision point indicates the distance between the non-collision point and the ground, and the point cloud type of the non-collision point and the ground relative height are taken as the attributes of the boundary and output along with the boundary of the drivable area, so that safer and more reliable driving behaviors can be determined, such as crossing or bypassing the area where the non-collision point is located, and the vehicle can be driven safely.
Through the method from the step S101 to the step S105, the accurate boundary position of the drivable area can be obtained, and the dynamic and static information and the ground relative height at the boundary position can be obtained.
The following further describes step S103 and step S105, respectively.
In step S103, the dynamic target detection models have different detection accuracies and different degrees of influence on the result of the dynamic and static detection of the non-ground point cloud, and the higher the detection accuracy is, the more accurate the result of the dynamic and static detection is, the lower the detection accuracy is, and the less accurate the result of the dynamic and static detection is. In this regard, as shown in fig. 5, in some embodiments, after the point cloud category of the non-ground point cloud is obtained, the point cloud category of the non-ground point cloud may be further modified through the following steps S301 to S303, so as to improve the accuracy of the point cloud category.
Step S301: and clustering the non-ground point clouds according to the space between the three-dimensional point clouds to obtain a plurality of point cloud clusters.
In this embodiment, a conventional data clustering method may be adopted to cluster the non-ground point clouds according to the distance between the three-dimensional point clouds, which is not specifically limited in this embodiment.
Step S302: and acquiring the dynamic and static types of each point cloud cluster.
The dynamic and static types can comprise a dynamic point cloud cluster and a static point cloud cluster, if the point cloud cluster is the dynamic point cloud cluster, all or most of three-dimensional point clouds in the point cloud cluster are three-dimensional point clouds belonging to a dynamic target; if the point cloud cluster is a static point cloud cluster, all or most of the three-dimensional point clouds in the point cloud cluster are three-dimensional point clouds belonging to a static target.
In some embodiments, the dynamic and static types of the point cloud clusters may be determined according to the occupation ratio of the dynamic point clouds in the point cloud clusters. Specifically, the moving and static types of each point cloud cluster may be acquired through the following steps S3021 to S3022.
Step S3021: and counting the occupation ratio of the dynamic point clouds in each point cloud cluster.
Step S3022: and comparing the proportion of the dynamic point cloud with a preset proportion threshold value.
If the proportion of the dynamic point clouds in the current point cloud cluster is larger than or equal to the preset proportion threshold value, the fact that most of the three-dimensional point clouds in the current point cloud cluster are dynamic point clouds is shown, the probability that the point cloud cluster is the dynamic point cloud cluster is high, and therefore the dynamic and static types of the current point cloud cluster are determined to be the dynamic point cloud cluster.
If the proportion of the dynamic point cloud in the current point cloud cluster is smaller than a preset proportion threshold value, the fact that most of the three-dimensional point clouds in the current point cloud cluster are static point clouds is shown, the probability that the point cloud cluster is the static point cloud cluster is high, and therefore the dynamic and static types of the current point cloud cluster are determined to be the static point cloud cluster.
Step S303: and correcting the point cloud category of the non-ground point cloud in the corresponding point cloud cluster according to the dynamic and static types.
According to the dynamic and static types of the point cloud cluster, a more real point cloud state (dynamic point cloud or static point cloud) can be determined, so that the point cloud type of the non-ground point cloud is corrected according to the dynamic and static types of the point cloud cluster, and the point cloud type can be more accurate.
In some embodiments, because the dynamic target has a large influence on the driving safety of the vehicle, in order to determine the boundary attribute of the safe and reliable drivable area as much as possible, if the dynamic and static types of the point cloud cluster are dynamic point cloud clusters, the static point cloud in the point cloud cluster is also modified into the dynamic point cloud, that is, the three-dimensional point cloud in the whole point cloud cluster is the dynamic point cloud, so as to avoid missing the dynamic point cloud and threatening the safety of the vehicle. And if the dynamic and static types of the point cloud cluster are static point cloud clusters, keeping the point cloud category of each three-dimensional point cloud in the point cloud cluster unchanged, and not correcting.
The above is a further description of step S103, and step S105 is further described below.
The ground relative height of the non-collidable point has the same great influence on the driving safety of the vehicle, and in order to determine the boundary attribute of the safe and reliable driving area as much as possible, the maximum ground relative height of the target to which the non-collidable point belongs can be selected as the ground relative height of the non-collidable point. Specifically, in some embodiments of the above step S105, the ground relative height of the non-collision point may be determined by the following steps S1051 to S105.
Step S1051: and clustering the non-ground point clouds according to the space between the three-dimensional point clouds to obtain a plurality of point cloud clusters.
In this embodiment, a conventional data clustering method may be adopted to cluster the non-ground point clouds according to the distance between the three-dimensional point clouds, which is not specifically limited in this embodiment.
Step S1052: and acquiring the ground relative height of each non-ground point cloud in the point cloud cluster to which the non-collision point belongs, and taking the maximum ground relative height as the ground relative height of the non-collision point.
The point cloud cluster to which the non-collision point belongs can represent the target to which the non-collision point belongs, and the maximum ground relative height in the point cloud cluster is the maximum ground relative height of the target to which the non-collision point belongs.
Further, in some embodiments of step S105, the ground height of the projected grid coverage area of each non-ground point cloud may be determined, and then the respective ground relative height may be determined according to the ground height of the projected grid coverage area of each non-ground point cloud and the respective point cloud height. Specifically, the ground height of the projected grid coverage area of each non-ground point cloud can be determined in the present embodiment through the following steps 11 to 13.
Step 11: judging that the projection grid of the current non-ground point cloud contains ground point cloud projection points; if the ground point cloud projection point is included, turning to step 12; if the ground point cloud projection point is not included, go to step 13.
Step 12: and taking the average value of the point cloud heights of all the three-dimensional point clouds projected into the grid as the ground height of the coverage area of the current grid, and marking the grid projected by the current non-ground point cloud into a ground point grid.
According to the method for determining the ground point cloud and the non-ground point cloud in the three-dimensional point clouds described in steps S1031 to S1035 in the foregoing method embodiment, if the grid includes the ground point cloud projection point, all the three-dimensional point clouds projected into the grid are the ground point clouds, and therefore, the average value of the point cloud heights of the three-dimensional point clouds can be used as the ground height of the grid coverage area.
Step 13: and determining the ground height of the coverage area of the grid projected by the current non-ground point cloud according to the ground height of the coverage area of other grids around the grid.
Because the ground height is smooth and continuous, the ground height of the current grid coverage area and the ground height of other surrounding grid coverage areas can not be changed suddenly, and therefore the ground height of the current grid coverage area can be approximately obtained according to the ground height of other surrounding grid coverage areas.
In some embodiments, the ground height of the current grid coverage area can be approximated according to the ground heights of other surrounding grid coverage areas through the following steps 131 to 133.
Step 131: respectively judging whether each other grid (other grids around the current grid) contains ground point cloud projection points; if at least one other grid includes a ground point cloud projection point, go to step 132; if all other grids do not contain ground point cloud projection points, go to step 133.
Step 132: and acquiring other grids comprising ground point cloud projection points, acquiring the ground heights of the coverage areas of the other grids, performing interpolation calculation on the ground heights, taking the height values obtained by the interpolation calculation as the ground heights of the coverage areas of the grids projected by the current non-ground point cloud, and marking the grids projected by the current non-ground point cloud into ground point grids. The ground height of the current grid coverage area can be approximately obtained according to the ground heights of other grid coverage areas through interpolation calculation.
Step 133: and taking the average value of the ground heights of all the ground point grid coverage areas as the ground height of the grid coverage area projected by the current non-ground point cloud.
When the current grid and other grids around the current grid do not contain ground point cloud projection points, the ground height of the current grid coverage area can be approximately obtained according to the ground height of all ground point grid coverage areas in the current polar coordinate grid diagram. Wherein the average of the ground heights of all the ground point grid coverage areas may represent the global ground height of the current entire polar grid map coverage area.
The above is a further description of step S105.
It should be noted that, although the foregoing embodiments describe each step in a specific sequence, those skilled in the art will understand that, in order to achieve the effect of the present invention, different steps do not necessarily need to be executed in such a sequence, and they may be executed simultaneously (in parallel) or in other sequences, and these changes are all within the protection scope of the present invention.
It will be understood by those skilled in the art that all or part of the flow of the method according to the above-described embodiment may be implemented by a computer program, which may be stored in a computer-readable storage medium and used to implement the steps of the above-described embodiments of the method when the computer program is executed by a processor. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable storage medium may include: any entity or device capable of carrying said computer program code, media, usb disk, removable hard disk, magnetic diskette, optical disk, computer memory, read-only memory, random access memory, electrical carrier wave signals, telecommunication signals, software distribution media, etc. It should be noted that the computer readable storage medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable storage media that does not include electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
Furthermore, the invention also provides computer equipment. In one computer apparatus embodiment according to the present invention, the computer apparatus comprises a processor and a storage device, the storage device may be configured to store a program for executing the travelable region detection method of the above-described method embodiment, and the processor may be configured to execute the program in the storage device, the program including but not limited to the program for executing the travelable region detection method of the above-described method embodiment. For convenience of explanation, only the parts related to the embodiments of the present invention are shown, and details of the specific techniques are not disclosed. The computer device may be a device formed by including various electronic devices.
Further, the invention also provides a computer readable storage medium. In one computer-readable storage medium embodiment according to the present invention, a computer-readable storage medium may be configured to store a program that executes the travelable region detection method of the above-described method embodiment, which may be loaded and executed by a processor to implement the travelable region detection method described above. For convenience of explanation, only the parts related to the embodiments of the present invention are shown, and details of the specific techniques are not disclosed. The computer readable storage medium may be a storage device formed by including various electronic devices, and optionally, the computer readable storage medium is a non-transitory computer readable storage medium in the embodiment of the present invention.
Further, the invention also provides a vehicle. In an embodiment of a vehicle according to the invention, the vehicle may comprise a computer device as described above for the embodiment of the computer device. The vehicle may be an autonomous vehicle, an unmanned vehicle, or the like in the present embodiment. In addition, the vehicle in the embodiment may be a fuel vehicle, an electric vehicle, a hybrid vehicle in which electric energy and fuel are mixed, or a vehicle using other new energy, and the like, classified according to the power source type.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
Claims (10)
1. A travelable region detection method, characterized in that the method further comprises:
acquiring a three-dimensional point cloud in a vehicle driving environment;
establishing a polar coordinate system corresponding to the three-dimensional coordinate system of the three-dimensional point cloud, generating a polar coordinate grid graph according to the preset grid number, and projecting the three-dimensional point cloud to the polar coordinate grid graph;
performing dynamic and static detection on non-ground point clouds in the three-dimensional point clouds to obtain point cloud categories of each non-ground point cloud, wherein the point cloud categories comprise dynamic point clouds and static point clouds;
determining non-collision points according to the non-ground point clouds projected into each grid;
determining the boundary of a drivable area in the vehicle driving environment according to the position of the non-collision point, and determining the attribute of the boundary according to the point cloud type of the non-collision point and the ground relative height.
2. The drivable region detection method as claimed in claim 1, characterized in that, after the step of "performing dynamic and static detection on the non-ground point clouds in the three-dimensional point cloud to obtain a point cloud class of each non-ground point cloud", the method further comprises correcting the point cloud classes of the non-ground point clouds by:
clustering the non-ground point clouds according to the distance between the three-dimensional point clouds to obtain a plurality of point cloud clusters;
acquiring the dynamic and static types of each point cloud cluster;
correcting the point cloud type of the non-ground point cloud in the corresponding point cloud cluster according to the dynamic and static types;
and/or the step of determining non-crashable points from the non-ground point clouds projected into each grid specifically comprises:
and taking the non-ground point cloud corresponding to the non-ground point cloud projection point with the minimum pole diameter in each grid as a non-collision point.
3. The drivable area detection method of claim 2, wherein the step of "acquiring the moving and static types of each point cloud cluster" specifically comprises:
counting the proportion of the dynamic point clouds in each point cloud cluster;
comparing the proportion of the dynamic point cloud with a preset proportion threshold;
if the proportion of the dynamic point cloud in the current point cloud cluster is larger than or equal to a preset proportion threshold value, the dynamic and static types of the current point cloud cluster are the dynamic point cloud cluster;
and if the proportion of the dynamic point cloud in the current point cloud cluster is smaller than a preset proportion threshold value, the dynamic and static types of the current point cloud cluster are static point cloud clusters.
4. The drivable area detection method of claim 2, wherein the step of modifying the point cloud class of the non-ground point cloud in the corresponding point cloud cluster according to the dynamic and static types specifically comprises:
if the dynamic and static types of the point cloud cluster are dynamic point cloud clusters, correcting the static point cloud in the point cloud cluster into dynamic point cloud;
and if the dynamic and static types of the point cloud cluster are static point cloud clusters, not correcting the point cloud category of the non-ground point cloud in the point cloud cluster.
5. The travelable region detection method according to claim 1, characterized in that the method further comprises acquiring the ground relative height of the non-collision point by:
clustering the non-ground point clouds according to the distance between the three-dimensional point clouds to obtain a plurality of point cloud clusters;
and acquiring the ground relative height of each non-ground point cloud in the point cloud cluster to which the non-collision point belongs, and taking the maximum ground relative height as the ground relative height of the non-collision point.
6. The drivable region detection method of claim 1 further comprising determining the ground height of the projected grid coverage area of each of the non-ground point clouds in such a way that the ground relative height of each of the non-ground point clouds can be determined from the ground height and the point cloud height of the respective non-ground point cloud:
if the grid projected by the current non-ground point cloud contains ground point cloud projection points, taking the average value of the point cloud heights of all three-dimensional point clouds projected into the grid as the ground height of the coverage area of the current grid, and marking the grid projected by the current non-ground point cloud as a ground point grid;
and if the projected grid of the current non-ground point cloud does not contain the ground point cloud projected point, determining the ground height of the coverage area of the projected grid of the current non-ground point cloud according to the ground heights of the coverage areas of other grids around the grid.
7. The drivable area detection method of claim 6, wherein the step of determining the ground height of the projected grid coverage area of the current non-ground point cloud from the ground heights of the other grid coverage areas surrounding the grid specifically comprises:
respectively judging whether each other grid contains ground point cloud projection points;
if at least one other grid comprises a ground point cloud projection point, acquiring the ground height of the coverage area of the other grid comprising the ground point cloud projection point, carrying out interpolation calculation on the ground height, taking the height value obtained by the interpolation calculation as the ground height of the coverage area of the current non-ground point cloud projected grid, and marking the current non-ground point cloud projected grid into a ground point grid;
and if each other grid does not contain the ground point cloud projection point, taking the average value of the ground heights of all the grid coverage areas of the ground points as the ground height of the grid coverage area projected by the current non-ground point cloud.
8. A computer apparatus comprising a processor and a storage device adapted to store a plurality of program codes, characterized in that the program codes are adapted to be loaded and run by the processor to perform a travelable region detection method according to any of claims 1-7.
9. A computer-readable storage medium, in which a plurality of program codes are stored, characterized in that the program codes are adapted to be loaded and executed by a processor to perform a travelable region detection method according to any of claims 1 to 7.
10. A vehicle characterized in that it comprises a computer device according to claim 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210557993.7A CN114966651A (en) | 2022-05-19 | 2022-05-19 | Drivable region detection method, computer device, storage medium, and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210557993.7A CN114966651A (en) | 2022-05-19 | 2022-05-19 | Drivable region detection method, computer device, storage medium, and vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114966651A true CN114966651A (en) | 2022-08-30 |
Family
ID=82984739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210557993.7A Pending CN114966651A (en) | 2022-05-19 | 2022-05-19 | Drivable region detection method, computer device, storage medium, and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114966651A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115469292A (en) * | 2022-11-01 | 2022-12-13 | 天津卡尔狗科技有限公司 | Environment sensing method and device, electronic equipment and storage medium |
| CN116258822A (en) * | 2023-05-16 | 2023-06-13 | 山东捷瑞数字科技股份有限公司 | Three-dimensional engine boundary defining method, device and storage medium based on meta universe |
| WO2024113594A1 (en) * | 2022-12-02 | 2024-06-06 | 智道网联科技(北京)有限公司 | Method and apparatus for quickly determining effective traveling area, and electronic device and storage medium |
-
2022
- 2022-05-19 CN CN202210557993.7A patent/CN114966651A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115469292A (en) * | 2022-11-01 | 2022-12-13 | 天津卡尔狗科技有限公司 | Environment sensing method and device, electronic equipment and storage medium |
| WO2024113594A1 (en) * | 2022-12-02 | 2024-06-06 | 智道网联科技(北京)有限公司 | Method and apparatus for quickly determining effective traveling area, and electronic device and storage medium |
| CN116258822A (en) * | 2023-05-16 | 2023-06-13 | 山东捷瑞数字科技股份有限公司 | Three-dimensional engine boundary defining method, device and storage medium based on meta universe |
| CN116258822B (en) * | 2023-05-16 | 2023-08-11 | 山东捷瑞数字科技股份有限公司 | Three-dimensional engine boundary defining method, device and storage medium based on meta universe |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114966651A (en) | Drivable region detection method, computer device, storage medium, and vehicle | |
| EP3624055B1 (en) | Ground detection method, apparatus, electronic device, vehicle and storage medium | |
| CN108629231B (en) | Obstacle detection method, apparatus, device and storage medium | |
| CN111932943B (en) | Dynamic target detection method and device, storage medium and roadbed monitoring equipment | |
| CN110286389B (en) | Grid management method for obstacle identification | |
| CN110632617B (en) | Laser radar point cloud data processing method and device | |
| CN110799989A (en) | Obstacle detection method, equipment, movable platform and storage medium | |
| CN111308500B (en) | Obstacle sensing method and device based on single-line laser radar and computer terminal | |
| CN113569958B (en) | Laser point cloud data clustering method, device, equipment and medium | |
| CN109948635B (en) | Target identification method and device based on laser scanning | |
| CN113295176B (en) | Map updating method, map updating device, and computer-readable storage medium | |
| CN114255252B (en) | Obstacle contour acquisition method, device, equipment and computer-readable storage medium | |
| US20230368407A1 (en) | Drivable area detection method, computer device, storage medium, and vehicle | |
| CN115436910A (en) | Data processing method and device for carrying out target detection on laser radar point cloud | |
| CN113859228A (en) | Vehicle control method and device, electronic equipment and storage medium | |
| CN115406457A (en) | Driving region detection method, system, equipment and storage medium | |
| CN114241448A (en) | Method and device for obtaining heading angle of obstacle, electronic equipment and vehicle | |
| CN115082891B (en) | Target detection method, computer device, computer-readable storage medium, and vehicle | |
| CN116434181A (en) | Ground point detection method, device, electronic equipment and medium | |
| CN115421160A (en) | Road edge detection method, device, equipment, vehicle and storage medium | |
| CN115205846A (en) | Three-dimensional target detection method, computer device, storage medium, and vehicle | |
| CN116863325A (en) | Method for multiple target detection and related product | |
| CN114549764A (en) | Obstacle identification method, device, equipment and storage medium based on unmanned vehicle | |
| CN113421306A (en) | Positioning initialization method, system and mobile tool | |
| CN116719054B (en) | Virtual laser radar point cloud generation method, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |