CN115683142A

CN115683142A - Method and device for determining region of interest

Info

Publication number: CN115683142A
Application number: CN202211309047.7A
Authority: CN
Inventors: 李文宽; 杜胜武; 万国强; 薛俊亮
Original assignee: Tianjin Jingwei Hengrun Technology Co ltd
Current assignee: Tianjin Jingwei Hengrun Technology Co ltd
Priority date: 2022-10-25
Filing date: 2022-10-25
Publication date: 2023-02-03

Abstract

The embodiment of the invention discloses a method and a device for determining an interested area, which can screen the interested area of a vehicle according to the connection relation between road sections and the distance between the boundary of each road section and the vehicle after positioning a target road section where the vehicle is located currently, namely, can screen partial road sections as the interested area in all road sections in a sensing range, so that only sensing information of the interested area can be calculated during environment sensing.

Description

Region-of-interest determination method and device

Technical Field

The invention relates to the technical field of automatic driving, in particular to a method and a device for determining an area of interest.

Background

In the automatic driving technology of a closed scene, the existing environment sensing method mainly uses a camera and a laser radar as sensing equipment. The camera can provide abundant visual information, and then utilizes the principle of binocular stereovision to judge information such as distance, degree of depth, can also be used to 3D and rebuild simultaneously, object identification etc.. The laser radar can provide abundant point cloud data and has the advantages of high precision, good robustness and the like. The closed scene may include a limited area scene such as a port, a mine, an airport, and a logistics park, for example.

In particular, in a known camera and radar-based environment sensing method, all information in an acquired sensing range is generally used for decision and navigation of automatic driving.

However, the above method has a problem that the camera and the radar can provide a wide sensing range and rich environmental information, but not all information within their sensing range can be used for automatic driving decision and navigation. That is, there are some redundant information that is not related to automatic driving, such as road information that is far from the vehicle, and information other than the road, and the like. When the redundant information is used, the calculation complexity is increased, the difficulty of environment sensing is increased, and further, the error of the environment sensing result is possibly large.

Disclosure of Invention

The invention provides a method and a device for determining an interested area, which are used for reducing the difficulty of environmental perception and improving the accuracy of an environmental perception result. The specific technical scheme is as follows.

In a first aspect, an embodiment of the present invention provides a method for determining a region of interest, where the method includes:

determining the current position of the vehicle in a map coordinate system;

loading a high-precision map, wherein the high-precision map at least comprises position information of each road section and a connection relation between each road section;

determining a current target road section of the vehicle according to the current position of the vehicle and the position information of each road section in the high-precision map;

searching each road section connected with the target road section according to the connection relation among the road sections, and adding the target road section and the searched road sections into a road section array to be extracted;

detecting whether an unextracted road section exists in the road section array to be extracted or not;

when the road section array to be extracted has the road section which is not extracted, selecting a current road section which is not extracted from the road section array to be extracted, and screening target boundary points of the current road section which is not extracted from the boundary points of the current road section which is not extracted according to the distance between each boundary point and the vehicle to be used as the interesting area of the vehicle; adding the unextracted road sections connected with the current unextracted road sections into the road section array to be extracted when the interesting regions are screened out, and returning to the step of detecting whether the unextracted road sections exist in the road section array to be extracted;

and when no road section which is not extracted exists in the road section array to be extracted, obtaining each interested area of the vehicle.

Optionally, the high-precision map further includes types of each road segment, where the types of each road segment include a road or an intersection;

in the boundary points of the currently unextracted road section, screening the target boundary points of the currently unextracted road section according to the distance between each boundary point and the vehicle, wherein the step of taking the target boundary points as the interested areas of the vehicle comprises the following steps:

when the current unextracted road section is a road, acquiring edge lines on two sides of the current unextracted road section, and respectively determining two end points of the distance between the two end points and the vehicle within a preset searching radius in the points included by each edge line to be used as the interested area of the vehicle;

and when the current unextracted road section is a road junction, acquiring all boundary points of the current unextracted road section, and determining the interested area of the vehicle according to the distance between each boundary point and the vehicle.

Optionally, the determining, among the points included in each edge line, two end points whose distance from the vehicle is within a preset search radius, as the region of interest of the vehicle, includes:

regarding a point included in any edge line, when the point satisfies any one of the following conditions, the point is taken as the region of interest of the vehicle: the current distance between the point and the vehicle is smaller than or equal to a preset retrieval radius, and the point is an endpoint of the edge line; the current distance between the point and the vehicle is smaller than or equal to a preset retrieval radius, and the current distance between the adjacent point with the point and the vehicle is larger than the preset retrieval radius.

Optionally, when the current unextracted road segment is a road, the obtaining of the edge lines on the two sides of the current unextracted road segment includes:

when the current unextracted road section is a road, judging whether the current unextracted road section is a straight road or not;

when the current unextracted road section is a straight line road, acquiring edge lines on two sides of the straight line road from the high-precision map as the edge lines on two sides of the current unextracted road section;

when the current unextracted road section is a curved road, acquiring edge lines on two sides of the curved road from the high-precision map, dividing the edge line on each side into a plurality of straight line segments, and taking the obtained plurality of straight line segments as the edge line on the side of the current unextracted road.

Optionally, the determining the region of interest of the vehicle according to the distance between each boundary point and the vehicle includes:

calculating the distance between each boundary point and the vehicle;

and when the distance between a boundary point and the vehicle is smaller than or equal to the preset retrieval radius, taking all the boundary points as the interested areas of the vehicle.

Optionally, the method further includes:

and constructing and sending data comprising each region-of-interest information of the vehicle, wherein each region-of-interest information at least comprises: the number of interested areas, the current position of the vehicle, the number of boundary points of each interested area and the position information of each boundary point.

In a second aspect, an embodiment of the present invention provides an apparatus for determining a region of interest, where the apparatus includes:

the current position determining module is used for determining the current position of the vehicle in a map coordinate system;

the high-precision map loading module is used for loading a high-precision map, and the high-precision map at least comprises position information of each road section and a connection relation between each road section;

the target road section determining module is used for determining a target road section where the vehicle is located currently according to the current position of the vehicle and the position information of each road section in the high-precision map;

the connected road section searching module is used for searching each road section connected with the target road section according to the connection relation between the road sections, and adding the target road section and the searched road sections into a road section array to be extracted;

the non-extraction road section detection module is used for detecting whether a non-extraction road section exists in the road section array to be extracted;

the road section detection module is used for selecting a current unextracted road section from the road section array to be extracted when the unextracted road section detection module determines that the unextracted road section exists in the road section array to be extracted, and screening a target boundary point of the current unextracted road section as an interested area of the vehicle in the boundary point of the current unextracted road section according to the distance between each boundary point and the vehicle; adding the un-extracted road section connected with the current un-extracted road section into the array of road sections to be extracted when the region of interest is screened out, and triggering the un-extracted road section detection module;

and the interested region determining module is used for obtaining each interested region of the vehicle when the unextracted road section detecting module determines that the unextracted road section does not exist in the road section array to be extracted.

Optionally, the high-precision map further includes types of each road segment, where the types of each road segment include a road or an intersection; the section detection module includes:

the road detection submodule is used for acquiring edge lines on two sides of the current unextracted road section when the current unextracted road section is a road, and determining two end points of which the distance from the vehicle is within a preset search radius in points included in each edge line to be used as an interested area of the vehicle;

and the intersection detection submodule is used for acquiring all boundary points of the current unextracted road section when the current unextracted road section is an intersection, and determining the region of interest of the vehicle according to the distance between each boundary point and the vehicle.

Optionally, the road detection sub-module is specifically configured to: regarding a point included in any edge line, when the point satisfies any one of the following conditions, the point is taken as the region of interest of the vehicle: the current distance between the point and the vehicle is smaller than or equal to a preset retrieval radius, and the point is an endpoint of the edge line; the current distance between the point and the vehicle is smaller than or equal to a preset retrieval radius, and the current distance between the adjacent point with the point and the vehicle is larger than the preset retrieval radius.

Optionally, the road detection sub-module is specifically configured to:

when the current unextracted road section is a road, judging whether the current unextracted road section is a straight line road or not;

when the current unextracted road section is a straight road, acquiring edge lines on two sides of the straight road from the high-precision map, and using the edge lines as the edge lines on two sides of the current unextracted road section;

Optionally, the intersection detection submodule is specifically configured to:

calculating the distance between each boundary point and the vehicle;

Optionally, the apparatus further comprises:

an information building module, configured to build and send data including information of each region of interest of the vehicle, where the information of each region of interest at least includes: the number of interested areas, the current position of the vehicle, the number of boundary points of each interested area and the position information of each boundary point.

As can be seen from the above, the method and the device for determining an area of interest provided in the embodiments of the present invention can screen out an area of interest of a vehicle according to a connection relationship between road sections and a distance between a boundary of each road section and the vehicle after positioning a target road section where the vehicle is currently located, that is, can screen out a part of road sections as the area of interest in all road sections within a sensing range, so that when performing environment sensing, only sensing information of the area of interest can be calculated.

The innovation points of the embodiment of the invention comprise:

1. after the target road section where the vehicle is located currently is located, the interested region of the vehicle is screened out according to the connection relation between the road sections and the distance between the boundary of each road section and the vehicle, that is, part of the road sections can be screened out in all the road sections in the sensing range to serve as the interested region, so that when environment sensing is conducted, only the sensing information of the interested region can be calculated, and compared with the case that all the sensing information is calculated, the calculation complexity can be reduced, the difficulty of environment sensing can be reduced, and the accuracy of an environment sensing result is improved.

2. The road condition of the road type road section is usually simpler, so that the region of interest can be determined according to the edge lines on the two sides of the road type road section only, so that the calculation resources are saved; the road conditions of the road sections of the intersection type are usually complex, so that the region of interest can be determined according to all boundary points of the road sections of the intersection type to acquire sufficient road information, and the accuracy of the environment sensing result is improved.

3. After the curved road is divided into a plurality of straight line sections, the region of interest of the curved road can be determined by adopting a region of interest determination method of the straight line sections, so that the difficulty in determining the region of interest of the curved road can be reduced, and the accuracy of determining the region of interest can be improved.

4. For the road section of the intersection type, when the boundary points located in the vehicle retrieval radius exist, all the boundary points are used as the interested areas, namely, the information of the whole intersection can be used as useful perception information to be processed, the accuracy of environment perception can be improved, and therefore the safety of vehicle driving is improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It is to be understood that the drawings in the following description are of some embodiments of the invention only. For a person skilled in the art, without inventive effort, further figures can be obtained from these figures.

Fig. 1 is a schematic flowchart of a method for determining a region of interest according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of information stored in a high-precision map according to an embodiment of the present invention;

fig. 3 is another schematic flow chart of a region of interest determination method according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a boundary point determination result of a road type road segment;

FIG. 5 is a diagram illustrating the result of determining the boundary points of the type road segments of the intersection;

FIG. 6 is a diagram illustrating a region of interest extraction result;

FIG. 7 is a diagram illustrating another region of interest extraction result;

FIG. 8 is a diagram of information constructed in an embodiment of the present invention;

fig. 9 is a schematic structural diagram of a region-of-interest determining apparatus according to an embodiment of the present invention.

Detailed Description

The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It is to be understood that the described embodiments are merely a few embodiments of the invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention.

It is to be noted that the terms "comprises" and "comprising" and any variations thereof in the embodiments and drawings of the present invention are intended to cover non-exclusive inclusions. A process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements but may alternatively include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

The embodiment of the invention discloses a method and a device for determining an area of interest, which can reduce the difficulty of environment perception and improve the accuracy of an environment perception result. The following provides a detailed description of embodiments of the invention.

Fig. 1 is a schematic flowchart of a method for determining a region of interest according to an embodiment of the present invention. The method can be applied to electronic equipment, and specifically can be applied to a processor with a data processing function in an autonomous vehicle, or any processor establishing connection with the autonomous vehicle, which is not limited in the embodiment of the present invention. The method specifically comprises the following steps.

S110: the current position of the vehicle in the map coordinate system is determined.

In the embodiment of the present invention, the electronic device may perform a positioning operation on the vehicle in real time, for example, the current position of the vehicle may be determined according to a set period, such as 1 millisecond, 3 milliseconds, 5 milliseconds, and the like.

In one implementation, the vehicle may be located by a Global Positioning System (GPS), so as to obtain a position of the vehicle in a GPS coordinate System, that is, longitude, latitude, and altitude coordinates of the vehicle may be obtained. In the embodiment of the invention, in order to determine the road section where the vehicle is located in the high-precision map and match the coordinates of each element in the high-precision map, such as a lane line, a lane edge and the like, an area of interest around the vehicle, namely an area which plays an important role in vehicle decision making, is determined. After the position of the vehicle in the GPS coordinate system is obtained, it can be converted into a map coordinate system. The region of interest may also be referred to as a geofence, and refers to a region around the vehicle where road conditions play an important role in vehicle decision making.

For example, the coordinates (x, y, z) of the vehicle in the map coordinate system can be obtained by the following formula:

where lon is the longitude of the vehicle in the GPS coordinate system, lat is the latitude of the vehicle in the GPS coordinate system, height is the height of the vehicle in the GPS coordinate system, and a and e are known constants.

S120: and loading a high-precision map, wherein the high-precision map at least comprises position information of each road section and a connection relation between each road section.

In the embodiment of the invention, the high-precision map of the closed scene can be constructed and stored in advance. As shown in fig. 2, the information stored in the high-precision map is shown in the figure, and taking the link 2 (segment 2) as an example, the following information can be stored in the high-precision map: the road section 2 comprises a road 1 and a road 2, wherein the road 1 consists of a left lane line, a right lane line and a road center line; the road 2 is also composed of a left lane line, a right lane line and a road center line; the links connected to the link 2 are an upper link 1 (segment 1) and a lower link 3 (segment 3). The specific information of the road segment1 and the road segment3 is similar to the road segment2, which is not described in detail in the embodiment of the present invention.

S130: and determining the current target road section of the vehicle according to the current position of the vehicle and the position information of each road section in the high-precision map.

For example, the current position of the vehicle may be entered into a high-precision map through an interface of the high-precision map. The current position of the vehicle is known, and the high-precision map stores the position information of each road section, namely the road section number where the vehicle is located can be determined, and the road section number can be called as a target road section.

S140: and searching each road section connected with the target road section according to the connection relation among the road sections, and adding the target road section and the searched road sections into the road section array to be extracted.

It can be understood that the perception information which plays a key role in vehicle decision making is information around the vehicle, and the road information far away from the vehicle has little influence on the decision making of the vehicle. Therefore, in the embodiment of the present invention, after the target road segment where the vehicle is currently located is determined, the region of interest of the vehicle may be determined in the road segment near the target road segment.

Specifically, the electronic device may search each road segment connected to the target road segment according to the connection relationship between the road segments, and add the target road segment and the searched road segments to the road segment array to be extracted. The road section array to be extracted is a data set, in which all road section information that needs to be extracted in the region of interest is stored, such as road section numbers, information of each lane line included in the road sections, and the like. For example, the information stored in the high-precision map of the target link and the searched links may be directly copied and stored in the link array to be extracted.

For example, taking fig. 2 as an example, when it is determined that the current target road segment of the vehicle is road segment2, the road segments connected to road segment2 are road segment1 and road segment3 according to the connection relationship stored in the high-precision map, that is, the related information of road segment2, road segment1 and road segment3 stored in the high-precision map may be added to the road segment array to be extracted.

S150: and detecting whether the road section to be extracted exists in the road section array to be extracted or not.

For example, each link information progress mark may be marked, for example, an unextracted link may be marked as 0, and an extracted link may be marked as 1. In this case, when detecting whether there is an unextracted link in the link array to be extracted, the electronic device may detect whether there is a link marked as 0 in the link array to be extracted, and if so, may determine that there is an unextracted link.

It is to be understood that, it is possible to mark each road segment in other manners, such as marking with different letters, numbers, or symbols, as long as the extracted road segment and the unextracted road segment can be distinguished, and this is not specifically limited in the embodiment of the present invention.

S160: when the road section array to be extracted has the road section which is not extracted, selecting a current road section which is not extracted from the road section array to be extracted, and screening target boundary points of the current road section which is not extracted from the boundary points of the current road section which is not extracted according to the distance between each boundary point and a vehicle to be used as the interested area of the vehicle; and adding the unextracted road sections connected with the current unextracted road sections into the road section array to be extracted when the region of interest is screened out, and returning to execute the step S150.

When the unextracted road section exists in the road section array to be extracted, a current unextracted road section can be selected from the road section array to be extracted, for example, the earliest stored road section can be selected as the current unextracted road section according to the storage sequence of each road section; or, according to the numbering sequence of each road section, selecting the road section with the minimum number as the current unextracted road section; or the selection of the currently unextracted road segment may be performed according to other rules, which is all possible and is not limited in the embodiment of the present invention.

It will be appreciated that an entire road segment is typically a larger area, wherein each smaller area has different attributes, such as different road conditions, different distances from the vehicle, and the same contribution to vehicle decision making. Therefore, in the unextracted section, a region which plays an important role in vehicle decision, called a region of interest, can be determined.

Specifically, the target boundary points of the currently unextracted road section can be screened as the interested area of the vehicle according to the distance between each boundary point and the vehicle in the boundary points of the currently unextracted road section. The boundary points of each road section can be directly obtained from the information stored in the high-precision map. As shown in fig. 2, in the road segment2, the boundary points are the left lane line of the road 1 and the right lane line of the road 2, i.e. two edge lines of the road segment 2.

For example, the distance between each boundary point and the vehicle may be calculated, and the boundary points with the distance less than or equal to the preset distance are all used as the target boundary points, and a region may be defined by a plurality of target boundary points as the region of interest of the vehicle. The preset distance may be set according to an actual situation, for example, the preset distance may be determined according to the size and the type of a closed scene, or may be determined according to the decision accuracy of an autonomous vehicle, which is not limited in the embodiment of the present invention.

After the current unextracted road section is extracted, the current unextracted road section mark can be modified into the extracted one according to the preset mark rule.

When the region of interest is screened out on the current unextracted road section, the fact that a region close to the vehicle exists in the current unextracted road section is shown, and a region meeting the distance requirement possibly exists in the unextracted road section connected with the current unextracted road section. In this case, the unextracted road segment connected to the current unextracted road segment may be added to the road segment array to be extracted to extract the newly added road segment, and a sufficient region of interest satisfying the distance requirement may be extracted to improve the accuracy of determining the region of interest.

And circularly executing the step S150 and the step S160 until all the road segments in the road segment array to be extracted are subjected to the region of interest extraction.

S170: and when the road sections which are not extracted do not exist in the road section array to be extracted, obtaining each interested area of the vehicle.

When there is no unextracted road segment in the road segment array to be extracted, it indicates that all road segments around the vehicle have been extracted, and the region of interest extraction is finished, and all the regions of interest extracted in the foregoing step S160 may be regarded as the regions of interest of the vehicle.

As can be seen from the above, in this embodiment, after the target road segment where the vehicle is currently located is located, the region of interest of the vehicle is screened out according to the connection relationship between the road segments and the distance between the boundary point of each road segment and the vehicle, that is, part of the road segments can be screened out as the region of interest in all the road segments within the sensing range, so that when environment sensing is performed, only the sensing information of the region of interest can be calculated.

It will be appreciated that in practical applications, different road segment types will affect vehicle decisions differently. For example, road type road segments are generally under relatively simple conditions, while intersection type road segments are generally under relatively complex conditions. The road type is a type in which the vehicle can travel only in one direction, and the intersection type is other than a road type, such as a junction, a crossroad, a roundabout, and the like. The intersection type of the complex road condition has great influence on the vehicle decision. Therefore, in the embodiment of the invention, different methods can be adopted to determine the region of interest aiming at different types of road sections, so as to improve the accuracy of determining the region of interest and further improve the accuracy of environment perception.

As an implementation manner of the embodiment of the present invention, the high-precision map further includes types of each road segment, and each type of each road segment includes a road or an intersection. Fig. 3 is another schematic flow chart of a method for determining a region of interest according to an embodiment of the present invention, where the method includes the following steps:

s310: the current position of the vehicle in the map coordinate system is determined.

S320: and loading a high-precision map, wherein the high-precision map at least comprises position information of each road section and a connection relation between each road section.

S330: and determining the current target road section of the vehicle according to the current position of the vehicle and the position information of each road section in the high-precision map.

S340: and searching each road section connected with the target road section according to the connection relation among the road sections, and adding the searched road sections into the road section array to be extracted.

S350: and detecting whether the road section to be extracted exists in the road section array to be extracted or not.

S360: and when the road section not extracted exists in the road section array to be extracted, selecting a current road section not extracted from the road section array to be extracted.

S370: when the current unextracted road section is a road, acquiring edge lines on two sides of the current unextracted road section, and determining two end points of which the distance from the vehicle is within a preset search radius in points included by each edge line respectively to be used as an interested area of the vehicle; and adding the road section not extracted connected with the road section not extracted at present into the road section array to be extracted when the region of interest is screened out, and returning to execute the step S350.

In one implementation, the end point of each edge line may be retrieved by, for a point included in any one of the edge lines, regarding the point as a region of interest of the vehicle when the point satisfies any one of the following conditions: the current distance between the point and the vehicle is smaller than or equal to a preset retrieval radius, and the point is an endpoint of an edge line; the current distance between the point and the vehicle is smaller than or equal to the preset retrieval radius, and the current distance between the adjacent point with the point and the vehicle is larger than the preset retrieval radius.

Specifically, the coordinates of each point included in any edge line may be acquired, the distance between the first point and the vehicle is calculated, and whether the distance is smaller than or equal to the search radius is determined, if the distance is smaller than or equal to the search radius, the distance is directly used as the endpoint of the edge line, and if the distance is larger than the search radius, the distance is not used as the endpoint, and the determination is continued for the next point.

Judging each intermediate point according to the following steps, calculating the distance between the current intermediate point and the vehicle, if the distance between the current intermediate point and the vehicle is larger than the retrieval radius, not taking the distance as an end point, and continuing to judge the next intermediate point; if the distance between the current intermediate point and the vehicle is smaller than or equal to the retrieval radius, the distance between the previous point and the vehicle is obtained, if the distance between the previous point and the vehicle is larger than the retrieval radius, the current intermediate point is used as an end point, if the distance between the previous point and the vehicle is smaller than or equal to the retrieval radius, the distance between the next point and the vehicle is calculated, if the distance between the next point and the vehicle is smaller than or equal to the retrieval radius, the current intermediate point is not used as the end point, and if the distance between the next point and the vehicle is larger than the retrieval radius, the current intermediate point is used as the end point.

For the last point, the distance between the last point and the vehicle can be calculated, whether the distance is smaller than or equal to the retrieval radius is judged, if so, the distance is directly used as the endpoint of the edge line, and if so, the distance is not used as the endpoint.

As shown in fig. 4, the circle in the figure is a circle drawn according to the search radius with the vehicle as the center of the circle, and in the edge line on one side of the road where the vehicle is located, the point 410 is the first point of the current edge line, and is located within the search radius, and is reserved as the end point of the edge line; point 420 is within the search radius and the next point to that point is not within the search radius, then that point is retained as the other end point of the edge line. Points 430 are within distance, but they are neither the first or last point, nor the next point is outside the retrieved distance, and therefore not an end point; the point 440 is out of range and discarded. The end points 410 and 420 of the side edge lines are finally obtained. Using a similar method for the other side edge line, the

edge lines

450, 460 can be obtained. That is, for this segment, the resulting region of interest is the

end points

410, 420, 450, 460.

It will be appreciated that in practical applications, the road segments of the road type may include straight roads and may also include curved roads. In the embodiment of the invention, in order to simplify the algorithm complexity and improve the algorithm precision, when the current unextracted road section is a road and the edge lines on two sides of the current unextracted road section are obtained, whether the current unextracted road section is a straight line road can be judged; when the current unextracted road section is a straight line road, acquiring edge lines on two sides of the straight line road from the high-precision map as the edge lines on two sides of the current unextracted road section; when the current unextracted road section is a curved road, acquiring edge lines on two sides of the curved road from the high-precision map, dividing the edge line on each side into a plurality of straight line segments, and taking the obtained plurality of straight line segments as the edge line on the side of the current unextracted road.

In one implementation, whether it is a straight road or a curved road may be determined according to the curvature of the currently unextracted link. For example, a road having a curvature smaller than a certain value may be determined as a straight road, and a road having a curvature larger than a certain value may be determined as a curved road. The above-mentioned fixed value can be set according to the requirement, and the embodiment of the present invention is not limited to this specifically. In dividing a curved road into a plurality of straight line segments, for example, a segment of the road having a curvature within a certain range may be divided into one straight line segment.

After the curved road is divided into a plurality of straight line sections, the region of interest of the curved road can be determined by adopting a region of interest determination method of the straight line sections, so that the difficulty of determining the region of interest of the curved road can be reduced, and the accuracy of determining the region of interest can be improved. In addition, for the curved road section with larger curvature, the road condition can be simplified by dividing the curved road section into the straight line sections, and the difficulty of determining the region of interest is further reduced.

S380: when the current unextracted road section is the intersection, acquiring all boundary points of the current unextracted road section, and determining the region of interest of the vehicle according to the distance between each boundary point and the vehicle; and adding the unextracted road sections connected with the current unextracted road sections into the road section array to be extracted when the region of interest is screened out, and returning to execute the step S350.

For example, the distance between each boundary point included in the intersection and the vehicle may be calculated, and when the distances between the boundary points satisfying a certain proportion and the vehicle are all smaller than or equal to the retrieval radius, all the boundary points included in the intersection are determined as the region of interest of the vehicle. The above ratio may include 40%, 50%, 60%, and the like, which is not limited in the embodiment of the present invention.

In one implementation, the electronic device may calculate a distance between each boundary point and the vehicle; and when the distance between one boundary point and the vehicle is smaller than or equal to the preset retrieval radius, all the boundary points are used as the interested areas of the vehicle.

That is, as long as one boundary point is located within the preset retrieval radius, the intersection can be used as the interested area of the vehicle, so as to improve the accuracy of vehicle decision.

For the road section of the intersection type, when the boundary points positioned in the vehicle retrieval radius exist, all the boundary points are used as interested areas, namely, the information of the whole intersection can be used as useful perception information to be processed, so that the accuracy of environment perception can be improved, and the driving safety of the vehicle is improved.

As shown in fig. 5, the vehicle is located at the link 2 (segment 2), the link 5 (segment 5) is a link connected to the link 2, the link 5 is an intersection, and the existing boundary points are located within the radius, so that all the boundary points of the link 5 can be regarded as the region of interest of the vehicle.

S390: and when the road sections which are not extracted do not exist in the road section array to be extracted, obtaining each interested area of the vehicle.

The road condition of the road section of the road type is usually simpler, so that the interested area can be determined according to the edge lines on the two sides of the road section of the road type only so as to save computing resources; the road conditions of the road sections of the intersection type are usually complex, so that the region of interest can be determined according to all boundary points of the road sections of the intersection type to acquire sufficient road information, and the accuracy of the environment sensing result is improved.

Fig. 6 is a schematic diagram of a result of extracting an interesting region, where a circle is a circle drawn according to a retrieval radius and with a vehicle as a center. The vehicle is located in the section 2 (segment 2), and the sections 5 (segment 5) and 6 (segment 6) are sections connected to the section 2.

A region of interest ROI-2 is retrieved in the road segment2, consisting of 4 asterisk boundary points. All boundary points of the road segment6 are located outside the retrieval radius, so no region of interest of the vehicle is retrieved for the road segment 6. Since the road section 5 is an intersection and the existing boundary points are located within the radius, all the boundary points of the road section 5 can be regarded as the region of interest of the vehicle, i.e., ROI-5, and are composed of a plurality of circular boundary points. After the interesting region of the road segment5 is determined, the

road segments

1, 3 and 4 (segment 1, segment3 and segment 4) connected to the road segment5 may be added to the road segment array to be extracted, and the interesting regions of the

road segments

1, 3 and 4 may be sequentially extracted.

For example, after the road segment1 is extracted, the region of interest ROI-1 is obtained, and is composed of 4 asterisk boundary points, the road segment 7 (not shown in the figure) connected to the road segment1 may also be added to the road segment array to be extracted, and when the road segment 7 is extracted, all the boundary points are located outside the search radius, so that the region of interest is not extracted from the road segment 7. The same operation was performed for the

links

3, 4, and the results were the same.

The regions of interest obtained finally are ROI-2, ROI-5, ROI-1, ROI-3 and ROI-4. Wherein, ROI-1, ROI-2, ROI-3 and ROI-4 are all composed of 4 asterisk boundary points, and ROI-5 is composed of a plurality of circular boundary points.

Fig. 7 is a schematic diagram of another region of interest extraction result, in which the circle is a circle drawn according to the retrieval radius with the vehicle as the center. The vehicles are located on the link 1 (segment 1), and the link 5 (segment 5) and the link 2 (segment 2) are links connected to the link 1.

A region of interest ROI-1 is retrieved in the road segment1, consisting of 4 asterisk boundary points. All boundary points of the road segment5 are located outside the retrieval radius, so the region of interest of the vehicle is not retrieved for the road segment 5. Since the road segment2 is an intersection and the boundary points are located within the radius, all the boundary points of the road segment2 can be regarded as the region of interest of the vehicle, i.e., ROI-2, and are formed by a plurality of circular boundary points. After the region of interest of the road segment2 is determined, the road segments 3 and 4 (segment 3 and segment 4) connected to the road segment2 may be added to the road segment array to be extracted, and the region of interest of the

road segments

3 and 4 may be sequentially extracted.

For example, the region of interest ROI-3 obtained by extracting the road segment3 is composed of 4 asterisk boundary points, and the region of interest ROI-4 obtained by extracting the road segment4 is composed of 4 asterisk boundary points.

The regions of interest finally extracted are ROI-1, ROI-2, ROI-3 and ROI-4. Wherein, ROI-1, ROI-3 and ROI-4 are all composed of 4 asterisk boundary points, and ROI-2 is composed of a plurality of circular boundary points.

As an implementation manner of the embodiment of the present invention, after the region of interest of the vehicle is determined, the electronic device may further send the obtained information to a module such as a decision module and the like that needs to use the perception information, so that the module that receives the information may screen the information perceived by the sensor according to the region of interest of the vehicle, and only make a decision according to the perception information in the region of interest, thereby reducing the calculation amount and improving the accuracy of the decision.

In one implementation, the electronic device may construct and transmit data including region-of-interest information of the vehicle, where the region-of-interest information includes at least: the number of the interested areas, the current position of the vehicle, the number of boundary points of each interested area and the position information of each boundary point.

After extracting the respective regions of interest ROI-1-ROI-5 of the vehicle, as shown in fig. 8, the information as shown in the figure can be constructed and sent to the decision module. The constructed information includes: the information head and each region of interest, wherein the information head includes: the number of ROIs and vehicle coordinates; the specific contents in each ROI include the number of points and the coordinates of each point. The number of points included in the region of interest is the number of boundary points.

Specifically, the number of the interested regions included in the information header is 5; the number of points included in ROI-1 was 4, the number of points included in ROI-2 was 4, the number of points included in ROI-3 was 4, the number of points included in ROI-4 was 4, the number of points included in ROI-5 was 24.

An embodiment of the present invention provides a device for determining a region of interest, as shown in fig. 9, the device includes:

a current position determining module 910, configured to determine a current position of the vehicle in a map coordinate system;

the high-precision map loading module 920 is configured to load a high-precision map, where the high-precision map at least includes location information of each road segment and a connection relationship between the road segments;

a target road section determining module 930, configured to determine a target road section where the vehicle is currently located according to the current position of the vehicle and position information of each road section included in the high-precision map;

a connected road section searching module 940, configured to search, according to the connection relationship between the road sections, each road section connected to the target road section, and add the target road section and each searched road section to the road section array to be extracted;

an unextracted road section detection module 950 for detecting whether an unextracted road section exists in the road section array to be extracted;

the road section detection module 960 is configured to, when the unextracted road section detection module 850 determines that an unextracted road section exists in the road section array to be extracted, select a current unextracted road section in the road section array to be extracted, and screen, in a boundary point of the current unextracted road section, a target boundary point of the current unextracted road section as an area of interest of the vehicle according to a distance between each boundary point and the vehicle; adding the unextracted road sections connected with the current unextracted road sections into the road section array to be extracted when the interesting regions are screened out, and triggering the unextracted road section detection module;

the region of interest determining module 970 is configured to obtain each region of interest of the vehicle when the unextracted road segment detecting module 850 determines that the unextracted road segment does not exist in the road segment array to be extracted.

Optionally, the high-precision map further includes types of each road segment, where the types of each road segment include a road or an intersection; the section detection module 960 includes:

Optionally, the road detection submodule is specifically configured to: regarding a point included in any edge line, when the point satisfies any one of the following conditions, the point is taken as the region of interest of the vehicle: the current distance between the point and the vehicle is smaller than or equal to a preset retrieval radius, and the point is an endpoint of the edge line; the current distance between the point and the vehicle is smaller than or equal to a preset retrieval radius, and the current distance between the adjacent point with the point and the vehicle is larger than the preset retrieval radius.

Optionally, the road detection sub-module is specifically configured to:

and when the current unextracted road section is a curved road, acquiring edge lines on two sides of the curved road from the high-precision map, dividing the edge line on each side into a plurality of straight line sections, and taking the obtained plurality of straight line sections as the edge line on the side of the current unextracted road.

Optionally, the intersection detection submodule is specifically configured to:

calculating the distance between each boundary point and the vehicle;

Optionally, the apparatus further comprises:

The device for determining the region of interest provided by the embodiment of the invention can screen the region of interest of the vehicle according to the connection relation between road sections and the distance between the boundary of each road section and the vehicle after positioning the current target road section of the vehicle, that is, can screen part of road sections in all road sections in the sensing range as the region of interest, so that only the sensing information of the region of interest can be calculated during environment sensing, and compared with the calculation of all sensing information, the device can reduce the complexity of calculation, namely reduce the difficulty of environment sensing and improve the accuracy of an environment sensing result.

The above device embodiment corresponds to the method embodiment, and has the same technical effect as the method embodiment, and for the specific description, refer to the method embodiment. The device embodiment is obtained based on the method embodiment, and for specific description, reference may be made to the method embodiment section, which is not described herein again.

Those of ordinary skill in the art will understand that: the figures are merely schematic representations of one embodiment, and the blocks or flow diagrams in the figures are not necessarily required to practice the present invention.

Those of ordinary skill in the art will understand that: modules in the devices in the embodiments may be distributed in the devices in the embodiments according to the description of the embodiments, or may be located in one or more devices different from the embodiments with corresponding changes. The modules of the above embodiments may be combined into one module, or further split into multiple sub-modules.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A method of region of interest determination, the method comprising:

determining the current position of the vehicle in a map coordinate system;

and when the road section array to be extracted does not have the road section which is not extracted, obtaining each interested area of the vehicle.

2. The method of claim 1, wherein the high-precision map further comprises a type of each road segment, the type of each road segment comprising a road or an intersection;

and when the current unextracted road section is the intersection, acquiring all boundary points of the current unextracted road section, and determining the region of interest of the vehicle according to the distance between each boundary point and the vehicle.

3. The method according to claim 2, wherein the determining, as the region of interest of the vehicle, two end points whose distance from the vehicle is within a preset search radius, among the points included in each edge line, respectively, comprises:

4. The method according to claim 2, wherein when the current unextracted segment is a road, obtaining edge lines on both sides of the current unextracted segment comprises:

5. The method of claim 2, wherein determining the region of interest of the vehicle based on the distance of each boundary point from the vehicle comprises:

calculating the distance between each boundary point and the vehicle;

6. The method according to any one of claims 1-5, further comprising:

7. An apparatus for region of interest determination, the apparatus comprising:

the unextracted road section detection module is used for detecting whether the unextracted road section exists in the road section array to be extracted;

the road section detection module is used for selecting a current unextracted road section from the road section array to be extracted when the unextracted road section detection module determines that the unextracted road section exists in the road section array to be extracted, and screening target boundary points of the current unextracted road section as the interested area of the vehicle in the boundary points of the current unextracted road section according to the distance between each boundary point and the vehicle; adding the un-extracted road section connected with the current un-extracted road section into the array of road sections to be extracted when the region of interest is screened out, and triggering the un-extracted road section detection module;

and the interesting region determining module is used for obtaining each interesting region of the vehicle when the unextracted road section detecting module determines that the unextracted road section does not exist in the road section array to be extracted.

8. The apparatus of claim 7, wherein the high-precision map further comprises a type of each road segment, the type of each road segment comprising a road or an intersection; the section detection module includes:

9. The apparatus of claim 8, wherein the road detection sub-module is specifically configured to: regarding a point included in any edge line, when the point satisfies any one of the following conditions, the point is taken as the region of interest of the vehicle: the current distance between the point and the vehicle is smaller than or equal to a preset retrieval radius, and the point is an endpoint of the edge line; the current distance between the point and the vehicle is smaller than or equal to a preset retrieval radius, and the current distance between the adjacent point with the point and the vehicle is larger than the preset retrieval radius.

10. The apparatus of claim 8, wherein the road detection sub-module is specifically configured to: