WO2022067534A1

WO2022067534A1 - Occupancy grid map generation method and device

Info

Publication number: WO2022067534A1
Application number: PCT/CN2020/118950
Authority: WO
Inventors: 孙翔雨
Original assignee: 华为技术有限公司
Priority date: 2020-09-29
Filing date: 2020-09-29
Publication date: 2022-04-07
Also published as: CN112543938A; CN112543938B

Abstract

An occupancy grid map generation method and device. The method comprises: acquiring point clouds of a surrounding environment collected by a point cloud sensor (S801); acquiring an obstacle point cloud and a ground point cloud from the point clouds of the surrounding environment (S802); fitting a curved driving pavement according to features of the ground point cloud and the ground, the ground being the ground of a road on which a mobile platform is located, and the point cloud sensor being carried on the mobile platform (S803); determining, from the curved driving pavement, a pavement area ready for driving on, the pavement area ready for driving on being divided into a plurality of grids (S804); determining, according to the obstacle point cloud, the probability that each grid is occupied (S805); and generating a curved occupancy grid map according to the probability that each grid is occupied and the pavement area ready for driving on (S806).

Description

Method and device for generating occupancy grid map

technical field

The present application relates to computer technology, and in particular, to a method and apparatus for generating an occupancy grid map.

Background technique

There are many classification methods for robot maps, including scale maps, topological maps, and semantic maps. Among them, the occupancy grid map (OGM) in the scale map is the most widely used. Among them, the detection area can be divided into grids of a certain number and size, and the probability of each grid being occupied can be determined according to the detection results of the detector, and the occupied probability of each grid can be reflected to the corresponding detection area. On the grid, the occupancy grid map can be obtained, in which the detector can be a point cloud sensor. Therefore, the occupancy grid map can reflect the obstacle information in the detection area.

However, the current method for obtaining the occupancy grid map is not accurate enough to accurately reflect the obstacle information in the detection area.

SUMMARY OF THE INVENTION

The present application provides a method and device for generating an occupancy grid map, which can obtain an accurate occupancy grid map.

In a first aspect, an embodiment of the present application provides a method for generating an occupancy grid map, including: acquiring a point cloud of a surrounding environment collected by a point cloud sensor; acquiring an obstacle point cloud and a ground surface from the point cloud of the surrounding environment point cloud; according to the ground point cloud and the characteristics of the ground, fit a curved driving road surface, the ground is the ground of the road where the mobile platform is located, and the point cloud sensor is mounted on the mobile platform; driving from the curved surface The road surface area to be driven is determined in the road surface, and the road surface area to be driven is divided into a plurality of grids; according to the obstacle point cloud, the probability of each grid being occupied is determined; according to the occupied probability of each grid Probability and the area of the road to be traveled, generate a surface occupancy grid map.

In this scheme, the curved road surface is fitted based on the characteristics of the ground, and the road surface area to be driven is determined from the curved road surface. , to generate a surface occupied grid map, that is, the ground point cloud is no longer plane fitted, but the curved road surface that conforms to the actual characteristics of the ground is obtained by surface fitting, then the occupancy probability of each grid in the obtained road surface area is Accuracy increases, increasing the accuracy of the resulting occupancy raster map.

In an optional implementation manner, according to the ground point cloud and the characteristics of the ground, fitting the curved road surface includes: acquiring the centerline of the road where the ground is located; The vertical distance of the center line is less than or equal to the first point of the preset distance; using polynomial curve interpolation to fit each of the first points to obtain the road center fitting curve; fitting the curve according to the road center to obtain the curved road surface, The road center fitting curve is the center line of the curved road surface.

This scheme presents a specific realization of fitting curved road surface.

In an optional implementation manner, the determining the probability that each of the occupied grids is occupied according to the obstacle point cloud includes: extracting the distance between the obstacle point cloud and the curved road surface Each second point whose height difference is less than or equal to the maximum height of the vehicle; according to each of the second points, determine the probability that each of the occupied grids is occupied.

This solution can avoid misjudging the hanging objects on the road surface or the tops of bridge holes or tunnels as obstacles, and improves the accuracy of determining the probability that each grid in the road surface area to be driven is occupied, thereby improving the generated occupancy grid. accuracy of the grid map.

In an optional implementation manner, the determining the probability that each of the occupied grids is occupied according to the second point includes: for any one of the second points, determining the second point The first grid occupied by the point, and the influence probability of the second point on the first grid; add the influence probability of the second point occupying the same grid on the grid to obtain each of the grids The preselected occupancy probability of the grid; for any grid in each of the grids, the occupancy probability of the grid is obtained according to the first preselected occupancy probability of the grid and the occupancy probability of the grid at the previous moment.

In an optional implementation manner, determining the area of the road to be driven from the curved road surface includes: determining the area to be driven from the curved road surface according to the field of view of the point cloud sensor and the curved road surface. driving road area.

Optionally, the determining the area of the road to be driven from the curved road surface according to the field of view of the point cloud sensor and the curved road surface includes: according to the field of view of the point cloud sensor and the curved road surface , determine a first length, the first length is less than or equal to the length of the curved road surface; from the curved road surface, the length is determined as the first length, and the width is the first width of the road surface area to be driven , the first width is the width of the curved road surface.

Optionally, the determining the area of the road to be driven from the curved road surface according to the field of view of the point cloud sensor and the curved road surface includes: according to the field of view of the point cloud sensor and the curved road surface , determine a first length and a second width, the first length is less than or equal to the length of the curved driving surface, and the second width is less than or equal to the width of the curved driving surface; determine from the curved driving surface The length is the first length, and the width is the to-be-running road surface area of the second width.

In this solution, since the length of the road surface area to be driven is determined based on the detection range of the sensor, the determined road surface area to be driven on is more reasonable and accurate.

In an optional implementation manner, one side of the road surface area to be driven is coincident with the side of the curved driving surface area close to the mobile platform. This scheme determines the road area to be driven more reasonably and accurately.

In a second aspect, an embodiment of the present application provides an apparatus for generating an occupancy grid map, including: an acquisition module for acquiring a point cloud of the surrounding environment collected by a point cloud sensor; and a processing module for: from the surrounding environment Obtain the obstacle point cloud and the ground point cloud from the point cloud of the mobile platform; and fit the curved driving road surface according to the characteristics of the ground point cloud and the ground, the ground is the ground of the road where the mobile platform is located, and the point cloud sensor is equipped with on the mobile platform; and determining a road surface area to travel from the curved road surface, the road surface area to travel is divided into a plurality of grids; and determining each grid according to the obstacle point cloud occupied probability; and generating a surface occupied grid map according to the occupied probability of each grid and the to-be-traveled road surface area.

In an optional implementation manner, the processing module is specifically configured to: acquire the centerline of the road where the ground is located; acquire the first point cloud whose vertical distance from the centerline is less than or equal to a preset distance One point; use polynomial curve interpolation to fit each of the first points to obtain the road center fitting curve; according to the road center fitting curve, obtain the curved driving surface, and the road center fitting curve is the curved driving road surface the centerline.

In an optional implementation manner, the processing module is specifically configured to: extract each second point whose height difference between the obstacle point cloud and the curved road surface is less than or equal to the maximum height of the vehicle; According to each of the second points, the probability that each of the occupied grids is occupied is determined.

In an optional implementation manner, the processing module is specifically configured to: determine the road surface area to be driven from the curved driving road surface according to the field of view of the point cloud sensor and the curved driving road surface.

In an optional implementation manner, the processing module is specifically configured to: determine a first length according to the field of view of the point cloud sensor and the curved road surface, where the first length is less than or equal to the curved surface The length of the driving road surface; the length of the curved driving road surface is determined as the first length and the width is the first width of the road surface area to be driven, and the first width is the width of the curved driving road surface.

In an optional implementation manner, the processing module is specifically configured to: determine a first length and a second width according to the field of view of the point cloud sensor and the curved road surface, where the first length is less than or is equal to the length of the curved driving surface, and the second width is less than or equal to the width of the curved driving surface; the length determined from the curved driving surface is the first length, and the width is all the second width. Describe the road area to be driven.

In an optional implementation manner, one side of the road surface area to be driven is coincident with the side of the curved driving surface area close to the mobile platform.

In a third aspect, embodiments of the present application provide a point cloud sensor, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores data that can be used by the at least one processor Instructions to be executed, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of the first aspect or any possible implementation of the first aspect.

In a fourth aspect, an embodiment of the present application provides a mobile platform on which the point cloud sensor described in the third aspect is mounted.

In a fifth aspect, an embodiment of the present application provides a mobile platform, including a point cloud sensor and a processor; the point cloud sensor is configured to collect a point cloud of a surrounding environment, and send the point cloud of the surrounding environment to the processing the processor; the processor is configured to receive the point cloud of the surrounding environment, obtain the obstacle point cloud and the ground point cloud from the point cloud of the surrounding environment; and, according to the characteristics of the ground point cloud and the ground, fit a curved driving road surface, the ground is the ground of the road where the mobile platform is located, and the point cloud sensor is mounted on the mobile platform; is divided into a plurality of grids; and, according to the obstacle point cloud, determine the probability that each grid is occupied; and, according to the probability that each grid is occupied and the road area to be driven, generate The surface occupies the raster map.

In an optional implementation manner, the processor is further configured to execute the method described in any possible implementation manner of the first aspect.

In a sixth aspect, an embodiment of the present application provides a storage medium, wherein the storage medium includes a computer program, and the computer program is used to implement the first aspect or any possible implementation manner of the first aspect. method.

Description of drawings

1 is a schematic diagram of a current plane occupying a grid map;

FIG. 2 is a schematic diagram 1 of a scene for obtaining an occupied grid map of a road surface provided by an embodiment of the present application;

FIG. 3 is a schematic diagram 2 of a scene for obtaining a grid map of road occupancy provided by an embodiment of the present application;

FIG. 4 is a schematic diagram 3 of a scenario for obtaining a grid map of road occupancy provided by an embodiment of the present application;

FIG. 5 is a schematic diagram 4 of a scenario for acquiring a grid map of road occupancy provided by an embodiment of the present application;

FIG. 6 is a schematic diagram 5 of a scenario for obtaining an occupied grid map of a road surface provided by an embodiment of the present application;

FIG. 7 is a schematic diagram 6 of a scenario for obtaining an occupied grid map of a road surface provided by an embodiment of the present application;

8 is a flowchart of a method for generating an occupancy grid map provided by an embodiment of the present application;

9 is a schematic diagram of a road centerline provided by an embodiment of the present application;

10 is a schematic diagram of a road surface area to be driven according to an embodiment of the application;

11 is a schematic diagram of a second point in an obstacle point cloud provided by an embodiment of the present application;

12 is a schematic diagram of a curved surface occupying a grid map provided by an embodiment of the present application;

13 is a schematic block diagram of an apparatus for generating an occupancy grid map provided by an embodiment of the present application;

14 is a schematic block diagram of a mobile platform provided by an embodiment of the present application;

FIG. 15 is a schematic block diagram of an electronic device according to an embodiment of the present application.

Detailed ways

First, the elements involved in the present application will be described.

The flat occupancy grid map (FOGM) is based on the assumption that the detection area is a plane, and divides the plane detection area into multiple grids. According to the detection data of the surrounding environment by the detector (such as radar and camera), Determine the probability that each grid is occupied by obstacles, reflect the probability that each grid is occupied by obstacles to the corresponding grids in the detection area, and obtain a plane occupied grid map. A schematic diagram of a plane occupied grid map can be shown in Figure 1. The darker the color filled in the grid, the greater the probability that the grid is occupied.

Curved occupancy grid map (COGM) refers to dividing the surface detection area into multiple grids. According to the detection data of the surrounding environment by the detector, the probability that each grid is occupied by obstacles is determined. The probability that each grid is occupied by obstacles is reflected to the corresponding grid in the surface detection area, and the grid map of surface occupancy is obtained.

For a better understanding of the present application, the existing problems are described below.

Figures 2 to 7 are schematic diagrams of several scenarios for obtaining the occupancy grid map of the road surface. Refer to Figures 2 to 7. The vehicles in Figures 2 to 7 are equipped with detectors. The road in Figure 2 is a flat road. The road in Fig. 3 is a curved road with an upward slope, the road in Fig. 4 is a curved road with a downward slope, the road in Fig. 5 is a curved road with unevenness, the road in Fig. The road is a road with bridge holes and tunnels.

At present, no matter what kind of scene it is in, the road is assumed to be a plane road, and the plane road is fitted based on the ground point cloud obtained by the detector, and a drivable area is determined from the plane road according to the field of view of the detector. The drivable area is divided into multiple grids, and the probability that each grid is occupied by obstacles is determined according to the obstacle point cloud obtained by the detector, and the probability that each grid is occupied by obstacles is reflected to the corresponding drivable area. on the grid to get the plane occupancy grid map.

It can be understood that, in the current method of obtaining a plane occupied grid map, in the scenarios shown in Figures 3 to 5, since the actual road surface is not a plane, the plane occupied grid map obtained by fitting the road surface to a plane road and Inaccurate. Further, in the scenario shown in FIG. 3 , there may be a situation where the uphill road is misjudged as an obstacle, and in the scenario shown in FIG. 4 , the downhill road may be misjudged as free to drive. In the scenario shown in Figure 5, there may be cases where the uphill road is misjudged as an obstacle, and the downhill road is misjudged as a freely drivable flat area. . In the scenarios shown in Figures 6 and 7, there are cases where the hanging objects are misjudged as obstacles, and in the scene in Figure 7, there are cases where bridge holes and the roof of the tunnel are misjudged as obstacles. That is to say, the current method for obtaining a grid map of plane occupancy, because the road surface is assumed to be a plane road surface, and/or, all the obstacle point cloud data are used in determining the probability of each grid being occupied by obstacles, so that The resulting plane occupancy raster map is not accurate enough.

In order to solve the above technical problems, in the present application, the curved road surface is fitted based on the actual characteristics of the ground, which can improve the accuracy of the obtained occupancy grid map.

The method for generating an occupancy grid map of the present application will be described below using specific embodiments.

FIG. 8 illustrates a method for generating an occupancy grid map provided by the implementation of the present application. The execution subject of this embodiment may be a generating device for occupying a grid map. Referring to FIG. 8 , the method of this embodiment includes:

Step S801 , acquiring the point cloud of the surrounding environment collected by the point cloud sensor.

The point cloud sensor in this embodiment may be a time of flight (TOF) sensor, or a radar or a camera. The radar can be a lidar, and the lidar can be a rotating lidar or a solid-state lidar. The point cloud sensor can be mounted on the mobile platform to obtain the point cloud of the surrounding environment of the mobile platform. The mobile platform can be a vehicle, such as an autonomous vehicle.

The device for generating the occupancy grid map in this embodiment may be all or part of the point cloud sensor, or all or part of the mobile platform equipped with the point cloud sensor, or may be connected to the point cloud sensor or the mobile platform for communication All or part of the server or end device of the relationship.

Step S802: Obtain the obstacle point cloud and the ground point cloud from the point cloud of the surrounding environment.

It can be understood that the point cloud of the surrounding environment includes the obstacle point cloud and the ground point cloud. The ground point cloud can be extracted from the point cloud of the surrounding environment first, and the remaining point cloud is the obstacle point cloud.

Optionally, the ground point cloud fast segmentation algorithm can be used to extract the ground point cloud from the point cloud of the surrounding environment.

That is to say, the method of this embodiment can accurately extract the ground point cloud from the point cloud of the surrounding environment. For example, in the scene shown in FIG. 3 , the method of this embodiment to extract the ground point cloud from the point cloud of the surrounding environment The point cloud corresponding to the uphill will not be misjudged as the obstacle point cloud.

Step S803: Fitting a curved driving road surface according to the ground point cloud and the characteristics of the ground. The ground is the ground of the road where the mobile platform is located, and the point cloud sensor is mounted on the mobile platform.

After the ground point cloud is obtained, the surface driving road can be fitted according to the ground point cloud.

In a specific implementation, according to the ground point cloud and the characteristics of the ground, the curved road surface is fitted, including the following a1 to a4:

a1. Obtain the centerline of the road where the ground is located.

It can be understood that the road in this embodiment is a road traveled by a mobile platform equipped with a point cloud sensor, wherein the centerline of the road where the ground is located may indicate the characteristics of the ground.

In a specific implementation, the range of the centerline can be calculated according to the boundary in the width direction of the road, and then the equation of the centerline of the road where the ground is located can be obtained, which is used to indicate the centerline of the road where the ground is located. The center line of the road where the ground is located is parallel to the extending direction of the road.

For example, referring to FIG. 9 , 901 shown in FIG. 9 is the center line of the road where the ground is located.

a2. Obtain the first point in the ground point cloud whose vertical distance from the center line is less than or equal to the preset distance.

For example, the equation of the center line of the road where the mobile platform where the point cloud sensor is located is a ₁ x ⁱ +b ₁ y ^j =0, then calculate the vertical distance between each point in the point cloud and the center line according to this equation , the point in the ground point cloud whose vertical distance from the center line is less than or equal to the preset distance is the first point. It can be understood that the number of the first points is plural.

Optionally, acquiring the first point in the ground point cloud whose vertical distance from the center line is less than or equal to a preset distance includes: filtering the ground point cloud, obtaining a filtered ground point cloud, and determining the filtered ground point The point in the cloud whose vertical distance from the center line is less than or equal to the preset distance is the first point.

a3. Use polynomial curve interpolation to fit each first point to obtain the road center fitting curve.

After a plurality of first points are obtained, polynomial curve interpolation is used to fit each first point to obtain a road center fitting curve. The polynomial curve interpolation may be L-nomial curve interpolation, where L is an integer greater than or equal to 2, for example, L=2 or 3 or 4 or 5 or 6.

a4. According to the fitting curve of the road center, the curved road surface is obtained, and the road center fitting curve is the center line of the curved road surface.

According to the formation principle of a straight surface, moving a line segment with the same width of the road along the center fitting curve can form a straight surface, that is, a curved road surface.

Specifically, the first end of the fitted curve through the center of the road, the width of the road on which the mobile platform where the point cloud sensor is located, and the first line segment perpendicular to the extending direction of the road can be obtained, and the first line segment can be obtained. Moving from the first end of the road center fitting curve to the second end of the road center fitting curve along the road center fitting curve, the obtained surface with the road center fitting curve as the center line is the curved driving surface. It can be understood that the first line segment is always perpendicular to the extending direction of the road during the moving process.

In other words, the curved road surface is equivalent to a curved surface obtained by moving the first line segment from the first end of the road center fitting curve to the second end of the road center fitting curve along the road center fitting curve.

Step S804: Determine the road surface area to be driven from the curved road surface, and the road surface area to be driven is divided into a plurality of grids.

Among them, the side of the road surface area to be driven close to the mobile platform can be overlapped with the side of the curved road surface close to the mobile platform, and the point cloud sensor is mounted on the mobile platform.

In a specific implementation, the road surface area to be driven is determined from the curved road surface, which may specifically include the following b1 to b2:

b1. Obtain a preset length.

Wherein, the preset length may be stored in the generating device occupying the grid map.

b2. Determine the length from the curved driving road surface as the preset length and the width as the road surface area to be driven with the first width, and the first width is the width of the curved driving road surface.

That is to say, the area of the road surface to be driven determined in this specific implementation is an area in the curved road surface where the length is the preset length and the width is the first width. In other words, the road surface area to be traveled can be abstracted as a plane formed by a straight line perpendicular to the advancing direction of the movable platform along the arc curve in the center of the road from moving a preset length.

After the road surface area to be driven is obtained, the road surface area to be driven is divided into a plurality of grids of the same size, such as M×N grids of the same size, wherein M and N are both positive integers.

In this specific implementation, since the length of the road surface area to be driven is a preset length, the efficiency of determining the road surface area to be driven is relatively high.

In another specific implementation, the area to be driven on the road surface may be determined from the curved road surface according to the detection range of the sensor and the curved road surface, which may specifically include the following c1 to c2:

c1. Determine the first length according to the field of view of the point cloud sensor and the curved road surface, where the first length is less than or equal to the length of the curved road surface.

Among them, when the farthest distance that the point cloud sensor can detect is less than the length of the curved road surface, the first length is the farthest distance that the point cloud sensor can detect, and when the farthest distance that the point cloud sensor can detect is greater than or When equal to the length of the curved road surface, the first length is equal to the length of the curved road surface.

c2. Determine the length as the first length and the width as the to-be-running road surface area of the first width from the curved driving road surface, and the first width is the width of the curved driving road surface.

That is to say, the road surface area to be driven is an area of the curved road surface where the length is the first length and the width is the first width.

After obtaining the road surface area to be driven, the road surface area to be driven is divided into a plurality of grids of the same size, as shown in FIG. 10 . Referring to FIG. 10 , the road area to be driven in FIG. 10 may be the road area to be driven obtained in the scenario shown in FIG. 3 . It can be understood that there are grids of different sizes in the road area to be driven because the grids are visually different. The driving road surface is fitted to a curved surface. In fact, the size of each grid included in the road surface area to be driven is the same.

In this specific implementation, since the length of the road surface area to be driven is determined based on the detection range of the sensor, the determined road surface area to be driven on is more reasonable and accurate.

In yet another specific implementation, the area to be driven on the road surface may be determined from the curved road surface according to the detection range of the sensor, which may specifically include the following d1 to d2:

d1. Determine a first length and a second width according to the field of view of the point cloud sensor and the curved road surface. The first length is less than or equal to the length of the curved road surface, and the second width is less than or equal to the width of the curved road surface.

The method for determining the first length is the same as above, and details are not repeated here. For the second width, when the maximum width that can be detected by the point cloud sensor is smaller than the first width of the curved road surface, the second width is the maximum width that the point cloud sensor can detect. When greater than or equal to the first width, the second width is equal to the first width.

d2. Determine, from the curved driving road surface, the road surface area to be driven, the length being the first length and the width being the second width.

Step S805: Determine the probability that each grid is occupied according to the obstacle point cloud.

The method for determining the probability that each grid in the road surface area to be driven is occupied according to the obstacle point cloud may refer to the current general method, which will not be repeated here.

In an optional manner, in order not to misjudge the suspension on the road surface or the top of a bridge hole or tunnel as an obstacle, in this step, "according to the obstacle point cloud, each The probability that the grid is occupied" can include the following e1~e2:

e1, extracting each second point whose height difference between the obstacle point cloud and the curved road surface is less than or equal to the maximum height of the vehicle.

That is, the vertical height difference between each second point and the curved road surface is less than or equal to the maximum height of the vehicle.

In one way, the equation of the curved road surface is: ax ^m +by ⁿ +cz ^k =0; for each point in the obstacle point cloud, the vertical distance between the point and the curved road surface is obtained. The point is the second point if the distance is less than or equal to the maximum height of the vehicle. Among them, a, b, c are constants, m is an integer greater than or equal to 2, n is a positive integer, such as 1 or 2 or 3, and k is a positive integer, such as 1 or 2 or 3.

Exemplarily, referring to FIG. 11 , 111 is a side view of a curved driving road surface, and the points between the curve 112 and the curve 111 are the second points.

e2. According to each second point, determine the probability that each grid in the road surface area to be driven is occupied.

Wherein, according to each second point, the specific implementation of determining the probability that each grid in the road surface area to be driven is occupied may be as follows:

e21. For any one of the second points, determine the first grid occupied by the second point and the influence probability of the second point on the first grid.

Wherein, the method for determining the influence probability of the second point on the first grid may refer to the current general method, which will not be repeated here.

e22. Add the influence probabilities of the second point occupying the same grid to the grid to obtain the preselected occupancy probability of each grid;

e23. For any first grid in each grid, obtain the occupancy probability of the first grid according to the first preselected occupancy probability of the first grid and the occupancy probability of the first grid at the previous moment.

Among them, the initial occupancy probability of each grid is 0.

According to the first pre-selected occupancy probability of the first grid and the occupancy probability of the first grid at the previous moment, the method for obtaining the occupancy probability of the first grid may refer to the current general method, which will not be repeated here.

The method e1-e2 determines the probability that each grid in the road area to be driven is occupied can avoid misjudging the hanging objects on the road surface or the top of a bridge or tunnel as an obstacle, which improves the determination of the probability of being occupied in the road area to be driven. The accuracy of the probability that each grid is occupied, thereby improving the accuracy of the generated occupied grid map.

Step S806 , according to the occupied probability of each grid in the road surface area to be driven and the road surface area to be driven, generate a grid map occupied by the curved surface.

That is, by reflecting the occupied probability of each grid in the road surface area to be driven to the corresponding grid, the surface occupied grid map can be generated. Specifically, as shown in Figure 12, the darker the color, the greater the probability that the grid is occupied.

Through the above steps S801 to S806, the method for obtaining the occupied grid map at time t according to the surrounding environment point cloud collected by the point cloud sensor at time t is described. It can be understood that the occupancy grid map at any time can be obtained according to the same method as above.

In this embodiment, the curved road surface is fitted based on the characteristics of the ground, and the road surface area to be driven is determined from the curved road surface. area, and generate a surface occupied grid map, that is, the ground point cloud is no longer flatly fitted, but the surface is fitted to obtain a curved driving road surface that conforms to the actual characteristics of the ground, then the occupancy probability of each grid in the road area to be driven is obtained. will increase the accuracy of the resulting occupancy raster map.

The method according to the present application has been described above, and the device according to the present application is described below.

FIG. 13 is a schematic block diagram of an apparatus for generating an occupancy grid map provided by an embodiment of the present application. Referring to FIG. 13 , the apparatus in this embodiment includes an acquisition module 1301 and a processing module 1302 .

The acquisition module 1301 is used to acquire the point cloud of the surrounding environment collected by the point cloud sensor;

The processing module 1302 is used for:

obtaining an obstacle point cloud and a ground point cloud from the point cloud of the surrounding environment; and

According to the ground point cloud and the characteristics of the ground, fit a curved driving surface, the ground is the ground of the road where the mobile platform is located, and the point cloud sensor is mounted on the mobile platform; and

determining a road surface area to travel from the curved road surface area, the road surface area to travel is divided into a plurality of grids; and

determining a probability that each of the grids is occupied according to the obstacle point cloud; and

According to the occupied probability of each grid and the area of the road surface to be driven, a grid map occupied by a curved surface is generated.

Optionally, the processing module 1302 is specifically used for:

Get the centerline of the road where the ground is located;

Obtain the first point whose vertical distance from the centerline is less than or equal to a preset distance in the ground point cloud;

Use polynomial curve interpolation to fit each of the first points to obtain a road center fitting curve;

According to the road center fitting curve, the curved driving surface is obtained, and the road center fitting curve is the center line of the curved driving road surface.

Optionally, the processing module 1302 is specifically used for:

extracting each second point whose height difference between the obstacle point cloud and the curved road surface is less than or equal to the maximum height of the vehicle;

According to each of the second points, the probability that each of the occupied grids is occupied is determined.

Optionally, the processing module 1302 is specifically used for:

According to the field of view of the point cloud sensor and the curved road surface, a road surface area to be driven is determined from the curved road surface.

Optionally, the processing module 1302 is specifically used for:

Determine a first length according to the field of view of the point cloud sensor and the curved road surface, where the first length is less than or equal to the length of the curved road surface;

From the curved driving surface, the length of the road surface area to be driven is determined as the first length and the width is the first width, and the first width is the width of the curved driving surface.

Optionally, the processing module 1302 is specifically used for:

According to the field of view of the point cloud sensor and the curved road surface, a first length and a second width are determined, the first length is less than or equal to the length of the curved road surface, and the second width is less than or equal to the the width of the surface driving surface;

A length of the first length and a width of the to-be-run surface area of the second width are determined from the curved running surface.

Optionally, one side of the road surface area to be driven is coincident with the side of the curved road surface area close to the mobile platform.

The apparatus in this embodiment can be used to execute the technical solutions in the foregoing method embodiments, and the implementation principles and technical effects thereof are similar, and details are not described herein again.

FIG. 14 is a schematic block diagram of a mobile platform provided by an embodiment of the present application. Referring to FIG. 14 , the apparatus of this embodiment includes: a point cloud sensor 1401 and a processor 1402;

The point cloud sensor is used for collecting the point cloud of the surrounding environment, and sending the point cloud of the surrounding environment to the processor;

The processor is configured to receive the point cloud of the surrounding environment, obtain the obstacle point cloud and the ground point cloud from the point cloud of the surrounding environment; and, according to the characteristics of the ground point cloud and the ground, fit a curved surface for driving a road surface, where the ground is the ground of the road where the mobile platform is located, and the point cloud sensor is mounted on the mobile platform; and a road surface area to be driven is determined from the curved road surface, and the road surface area to be driven is divided is a plurality of grids; and, according to the obstacle point cloud, determine the probability that each grid is occupied; and, according to the probability that each grid is occupied and the road area to be driven, generate a curved surface occupied Raster map.

Optionally, the processor 1402 is specifically configured to:

Get the centerline of the road where the ground is located;

Optionally, the processor 1402 is specifically configured to:

Optionally, the processor 1402 is specifically configured to: determine the road surface area to be driven from the curved driving road surface according to the field of view of the point cloud sensor and the curved driving road surface.

Optionally, the processor 1402:

Optionally, the processor 1402 is specifically configured to:

The mobile platform in this embodiment can be used to execute the technical solutions in the foregoing method embodiments, and the implementation principles and technical effects thereof are similar, and details are not described herein again.

Embodiments of the present application further provide a mobile platform, where a point cloud sensor is mounted on the mobile platform, and the point cloud sensor can execute the methods in the foregoing method embodiments.

FIG. 15 is a schematic block diagram of an electronic device according to an embodiment of the present application. The electronic device in this embodiment may be a mobile platform, or a chip, a chip system, or a processor that supports the mobile platform to implement the above method; the electronic device may be a point cloud sensor, or a point cloud sensor that supports the implementation of the above method chip, system-on-chip, or processor. The electronic device in this embodiment can be used to implement the method described in the foregoing method embodiment, and for details, reference may be made to the description in the foregoing method embodiment.

The electronic device may include one or more processors 1501, and the processors 1501 may also be referred to as processing units, which may implement certain control functions. The processor 1501 may be a general-purpose processor or a special-purpose processor, or the like.

In an optional design, the processor 1501 may also store instructions and/or data 1503, and the instructions and/or data 1503 may be executed by the processor, so that the electronic device executes the above method embodiments method described.

In another optional design, the processor 1501 may include a transceiver unit for implementing receiving and transmitting functions. For example, the transceiver unit may be a transceiver circuit, or an interface, or an interface circuit. Transceiver circuits, interfaces or interface circuits used to implement receiving and transmitting functions may be separate or integrated. The above-mentioned transceiver circuit, interface or interface circuit can be used for reading and writing code/data, or the above-mentioned transceiver circuit, interface or interface circuit can be used for signal transmission or transmission.

Optionally, the electronic device may include one or more memories 1502 on which instructions 1504 may be stored, and the instructions may be executed on the processor, so that the electronic device executes the above method embodiments method described. Optionally, data may also be stored in the memory. Optionally, instructions and/or data may also be stored in the processor. The processor and the memory can be provided separately or integrated together. For example, the corresponding relationship described in the above method embodiments may be stored in a memory or in a processor.

Optionally, the electronic device may further include a transceiver 1505 and/or an antenna 1506 . The processor 1501 may be referred to as a processing unit, and controls the electronic device. The transceiver 1505 may be referred to as a transceiver unit, a transceiver, a transceiver circuit or a transceiver, etc., and is used to implement a transceiver function.

An embodiment of the present application further provides a storage medium, characterized in that, the storage medium includes a computer program, and the computer program is used to implement the method in the foregoing method embodiment.

The processors and transceivers described in the embodiments of this application can be fabricated using various IC process technologies, such as complementary metal oxide semiconductor (CMOS), nMetal-oxide-semiconductor (nMetal-oxide-semiconductor, NMOS), P-type metal oxide semiconductor (positive channel metal oxide semiconductor, PMOS), bipolar junction transistor (Bipolar Junction Transistor, BJT), bipolar CMOS (BiCMOS), silicon germanium (SiGe), gallium arsenide (GaAs) )Wait.

It should be understood that the processor in this embodiment of the present application may be an integrated circuit chip, which has a signal processing capability. In the implementation process, each step of the above method embodiments may be completed by a hardware integrated logic circuit in a processor or an instruction in the form of software. The above-mentioned processor may be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other possible Programming logic devices, discrete gate or transistor logic devices, discrete hardware components.

It can be understood that the memory in this embodiment of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically programmable Erase programmable read-only memory (electrically EPROM, EEPROM) or flash memory. Volatile memory may be random access memory (RAM), which acts as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (synchlink DRAM, SLDRAM) ) and direct memory bus random access memory (direct rambus RAM, DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but not be limited to, these and any other suitable types of memory.

When the above-described embodiments are implemented using software, they may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server, or data center Transmission to another website site, computer, server, or data center by wire (eg, coaxial cable, optical fiber, digital subscriber line, DSL) or wireless (eg, infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that includes an integration of one or more available media. The available media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, high-density digital video discs (DVDs)), or semiconductor media (eg, solid state disks, SSD)) etc.

In the several embodiments provided in this application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus 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 may be combined or Can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.

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

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

The functions, if implemented in the form of software functional units and sold or used as independent products, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be 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 above are only specific embodiments of the present application, but the protection scope of the present application is not limited to this. should be covered within the scope of protection of this application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

It is to be understood that reference throughout the specification to an "embodiment" means that a particular feature, structure, or characteristic associated with the embodiment is included in at least one embodiment of the present application. Thus, various embodiments throughout this specification are not necessarily necessarily referring to the same embodiment. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner in one or more embodiments. It should be understood that, in various embodiments of the present application, the size of the sequence numbers of the above-mentioned processes does not mean the sequence of execution, and the execution sequence of each process should be determined by its functions and internal logic, and should not be dealt with in the embodiments of the present application. implementation constitutes any limitation.

It should also be understood that, in this application, "when", "if" and "if" all mean that the electronic device in this application will perform corresponding processing under certain objective circumstances, not a limited time, and also It is not required that the electronic device must have a judging action when it is implemented, nor does it mean that there are other limitations.

References in this application to elements in the singular are intended to mean "one or more" rather than "one and only one" unless specifically stated otherwise. In this application, unless otherwise specified, "at least one" is intended to mean "one or more", and "plurality" is intended to mean "two or more".

In addition, the term "and/or" in this article is only an association relationship to describe the associated objects, indicating that there can be three kinds of relationships, for example, A and/or B, it can mean that A exists alone, A and B exist at the same time, There are three cases of B alone, where A can be singular or plural, and B can be singular or plural.

The character "/" generally indicates that the associated objects are an "or" relationship.

The terms "at least one of" or "at least one of" herein mean all or any combination of the listed items, eg, "at least one of A, B, and C", It can be expressed as: A alone exists, B alone exists, C alone exists, A and B exist simultaneously, B and C exist simultaneously, and A, B and C exist simultaneously, where A can be singular or plural, and B can be Singular or plural, C can be singular or plural.

It should be understood that, in various embodiments of the present application, "B corresponding to A" means that B is associated with A, and B can be determined according to A. However, it should also be understood that determining B according to A does not mean that B is only determined according to A, and B may also be determined according to A and/or other information.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention. should be included within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

A method for generating an occupancy grid map, comprising:

Obtain the point cloud of the surrounding environment collected by the point cloud sensor;

Obtain the obstacle point cloud and the ground point cloud from the point cloud of the surrounding environment;

Fitting a curved road surface according to the ground point cloud and the characteristics of the ground, the ground is the ground of the road where the mobile platform is located, and the point cloud sensor is mounted on the mobile platform;

determining a road surface area to be driven from the curved road surface, the road surface area to be driven is divided into a plurality of grids;

According to the obstacle point cloud, determine the probability that each of the grids is occupied;

According to the occupied probability of each grid and the area of the road surface to be driven, a grid map occupied by a curved surface is generated.
The method according to claim 1, wherein the fitting of the curved road surface according to the ground point cloud and the characteristics of the ground comprises:

Obtain the centerline of the road where the ground is located;

Obtain the first point whose vertical distance from the centerline is less than or equal to a preset distance in the ground point cloud;

Use polynomial curve interpolation to fit each of the first points to obtain a road center fitting curve;

According to the road center fitting curve, the curved driving surface is obtained, and the road center fitting curve is the center line of the curved driving road surface.
The method according to claim 1 or 2, wherein the determining the probability that each of the occupied grids is occupied according to the obstacle point cloud comprises:

extracting each second point whose height difference between the obstacle point cloud and the curved road surface is less than or equal to the maximum height of the vehicle;

According to each of the second points, the probability that each of the occupied grids is occupied is determined.
The method according to claim 3, wherein the determining the probability that each of the occupied grids is occupied according to the second point comprises:

For any one of the second points, determine the first grid occupied by the second point and the influence probability of the second point on the first grid;

Adding the influence probabilities of the second point occupying the same grid to the grid to obtain the preselected occupancy probability of each of the grids;

For any one of the grids, the occupancy probability of the grid is obtained according to the first preselected occupancy probability of the grid and the occupancy probability at the previous moment.
The method according to any one of claims 1 to 4, wherein determining the road surface area to be driven from the curved road surface includes:

According to the field of view of the point cloud sensor and the curved road surface, a road surface area to be driven is determined from the curved road surface.
The method according to claim 5, wherein the determining, according to the field of view of the point cloud sensor, the area to be driven on the road surface from the curved road surface comprises:

Determine a first length according to the field of view of the point cloud sensor and the curved road surface, where the first length is less than or equal to the length of the curved road surface;

From the curved driving surface, the length of the road surface area to be driven is determined as the first length and the width is the first width, and the first width is the width of the curved driving surface.
The method according to claim 5, wherein the determining, according to the field of view of the point cloud sensor and the curved driving road surface, determines the road surface area to be driven from the curved driving road surface, comprising:

According to the field of view of the point cloud sensor and the curved road surface, a first length and a second width are determined, the first length is less than or equal to the length of the curved road surface, and the second width is less than or equal to the the width of the surface driving surface;

A length of the first length and a width of the to-be-run surface area of the second width are determined from the curved running surface.
The method according to any one of claims 1 to 7, characterized in that, one side of the road surface area to be driven is coincident with a side of the curved road surface area close to the moving platform.
A device for generating an occupied grid map, comprising:

The acquisition module is used to acquire the point cloud of the surrounding environment collected by the point cloud sensor;

Processing module for:

obtaining an obstacle point cloud and a ground point cloud from the point cloud of the surrounding environment; and

According to the ground point cloud and the characteristics of the ground, fit a curved driving surface, the ground is the ground of the road where the mobile platform is located, and the point cloud sensor is mounted on the mobile platform; and

determining a road surface area to travel from the curved road surface area, the road surface area to travel is divided into a plurality of grids; and

determining a probability that each of the grids is occupied according to the obstacle point cloud; and

According to the occupied probability of each grid and the area of the road surface to be driven, a grid map occupied by a curved surface is generated.
The device according to claim 9, wherein the point cloud sensor is mounted on a mobile platform, and the processing module is specifically used for:

Get the centerline of the road where the ground is located;

Obtain the first point whose vertical distance from the centerline is less than or equal to a preset distance in the ground point cloud;

Use polynomial curve interpolation to fit each of the first points to obtain a road center fitting curve;

According to the road center fitting curve, the curved driving surface is obtained, and the road center fitting curve is the center line of the curved driving road surface.
The device according to claim 9 or 10, wherein the processing module is specifically configured to:

extracting each second point whose height difference between the obstacle point cloud and the curved road surface is less than or equal to the maximum height of the vehicle;

According to each of the second points, the probability that each of the occupied grids is occupied is determined.
The device according to any one of claims 9 to 11, wherein the processing module is specifically configured to:

According to the field of view of the point cloud sensor and the curved road surface, a road surface area to be driven is determined from the curved road surface.
The device according to claim 12, wherein the processing module is specifically configured to:

Determine a first length according to the field of view of the point cloud sensor and the curved road surface, where the first length is less than or equal to the length of the curved road surface;

From the curved driving surface, the length of the road surface area to be driven is determined as the first length and the width is the first width, and the first width is the width of the curved driving surface.
The device according to claim 12, wherein the processing module is specifically configured to:

According to the field of view of the point cloud sensor and the curved road surface, a first length and a second width are determined, the first length is less than or equal to the length of the curved road surface, and the second width is less than or equal to the the width of the surface driving surface;

A length of the first length and a width of the to-be-run surface area of the second width are determined from the curved running surface.
The device according to any one of claims 9 to 14, wherein a side of the road surface area to be driven is coincident with a side of the curved road surface that is close to the moving platform.
A point cloud sensor, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the execution of any one of claims 1 to 8 Methods.
A mobile platform, characterized in that the mobile platform is equipped with the point cloud sensor of claim 16 .
A mobile platform, characterized in that it includes a point cloud sensor and a processor;

The point cloud sensor is used for collecting the point cloud of the surrounding environment, and sending the point cloud of the surrounding environment to the processor;

The processor is configured to receive a point cloud of the surrounding environment, and obtain an obstacle point cloud and a ground point cloud from the point cloud of the surrounding environment; and,

According to the ground point cloud and the characteristics of the ground, fit a curved driving surface, the ground is the ground of the road where the mobile platform is located, and the point cloud sensor is mounted on the mobile platform; and,

Determining a road surface area to travel on from the curved road surface, the road surface area to travel on is divided into a plurality of grids; and,

determining a probability that each of the grids is occupied according to the obstacle point cloud; and,

According to the occupied probability of each grid and the area of the road surface to be driven, a grid map occupied by a curved surface is generated.
The mobile platform according to claim 18, wherein the processor is further configured to execute the method of any one of claims 2-8.
A storage medium, characterized in that the storage medium comprises a computer program, and the computer program is used to implement the method according to any one of claims 1 to 8.