WO2023216651A1

WO2023216651A1 - Lane positioning method, computer device, computer-readable storage medium and vehicle

Info

Publication number: WO2023216651A1
Application number: PCT/CN2023/073864
Authority: WO
Inventors: 余红松; 郝立良; 王健; 杨东升; 易世春; 袁弘渊
Original assignee: 安徽蔚来智驾科技有限公司
Priority date: 2022-05-09
Filing date: 2023-01-30
Publication date: 2023-11-16

Abstract

A lane positioning method, comprising: positioning a vehicle by means of a two-dimensional map, and determining a position of the vehicle (S101); acquiring a communication relationship between lanes around the position of the vehicle (S102); and according to the communication relationship between the lanes around the position of the vehicle, determining a lane ID of the lane where the vehicle is currently located (S103). By means of the method, even if a vehicle travels in a road network having road layers of different heights, which lane the vehicle travels on can be accurately determined by means of a two-dimensional map and according to a communication relationship between lanes. Further provided are a computer device, a computer-readable storage medium and a vehicle.

Description

Lane positioning method, computer equipment, computer-readable storage medium and vehicle

The present invention has priority in the application number 202210507812. "Apparatus, Storage Media and Vehicles", the entire contents of which are incorporated herein by reference.

Technical field

The present invention relates to the field of positioning technology, and specifically provides a lane positioning method, computer equipment, computer-readable storage media and vehicles.

Background technique

Vehicles are usually equipped with on-board maps, which are used for positioning and navigation during driving. When a vehicle is driving in a road network with road layers of different heights, if there are multiple roads belonging to road layers of different heights at the same location, these roads will be cross-displayed on the on-board map. When the vehicle is driving at this location, it will It is impossible to accurately locate which road the vehicle is driving on through the on-board map. Furthermore, when each layer of roads contains multiple lanes, it is even more difficult to locate which lane the vehicle is driving on.

Accordingly, a new technical solution is needed in this field to solve the above problems.

Contents of the invention

In order to overcome the above shortcomings, the present invention is proposed to provide a lane positioning method, a computer device, and a computer readable method that solve or at least partially solve the technical problem of how to accurately locate the current driving lane of a vehicle in a road network with road layers of different heights. Storage media and vehicles.

In a first aspect, the present invention provides a lane positioning method, which method includes: locating a vehicle through a two-dimensional map and determining the vehicle position; obtaining the lanes around the vehicle position and the connectivity between the lanes; Determine the lane ID of the lane where the vehicle is currently located based on the lanes surrounding the vehicle position and the connectivity between the lanes.

In one technical solution of the above lane positioning method, the step of "determining the lane ID of the lane where the vehicle is currently located based on the lanes surrounding the vehicle position and the connectivity relationship between the lanes" specifically includes: determining the lane ID of the vehicle based on the last determined Whether the last lane ID can be determined; if it can be determined, determine the lane ID of the lane where the vehicle is currently located based on the map range of the lane corresponding to the last lane ID on the two-dimensional map and the current vehicle position; If it cannot be determined, the lane ID of the lane where the vehicle is currently located is determined based on the map range of the lane located within the preset range centered on the current vehicle position and falling on the two-dimensional map and the current vehicle position.

In one technical solution of the above lane positioning method, the step of "determining the lane ID of the lane where the vehicle is currently located based on the map range of the lane corresponding to the last lane ID on the two-dimensional map and the current vehicle position" specifically includes: Determine whether the current vehicle position falls within the map range corresponding to the last lane ID and the two-dimensional map only contains one lane within the map range; if so, use the last lane ID as the vehicle's current The lane ID of the lane where it is located; if not, obtain other lanes that belong to the same road layer as the lane corresponding to the last lane ID and have a connectivity relationship with the lane corresponding to the last lane ID, and locate according to the other lanes. Based on the map range and the current vehicle position on the two-dimensional map, determine the lane ID of the lane where the vehicle is currently located.

In one technical solution of the above lane positioning method, the step of "determining the lane ID of the lane where the vehicle is currently located based on the map range of the other lanes falling on the two-dimensional map and the current vehicle position" specifically includes: determining whether the current vehicle position is It only falls within the map range corresponding to one of the other lanes; if so, the lane ID of the other lane is used as the lane ID of the lane where the vehicle is currently located; if not, the lane ID of the other lane is used as the lane ID centered on the current vehicle position. The lanes within the preset range fall within the map range on the two-dimensional map and the current vehicle position, and the lane ID of the lane where the vehicle is currently located is determined.

In a technical solution of the above lane positioning method, "the lane ID of the lane where the vehicle is currently located is determined based on the map range of the lane located within the preset range centered on the current vehicle position and falling on the two-dimensional map and the current vehicle position." The steps specifically include: if the current vehicle position only falls within the map range corresponding to one lane within the preset range, then use the lane ID of the one lane within the preset range as the lane of the lane where the vehicle is currently located. ID; if the current vehicle position falls within the map range corresponding to multiple lanes within the preset range, then use the multiple lanes within the preset range as candidate lanes, based on the last lane The ID corresponds to a lane with connected lanes and/or the pose of the current vehicle, and the lane ID of the lane where the vehicle is currently located is filtered out from the lane IDs of the candidate lanes.

In one technical solution of the above lane positioning method, "select vehicles from the lane IDs of the candidate lanes based on the lanes that have a connected relationship with the lane corresponding to the last lane ID and/or the position and orientation of the current vehicle. The steps of "lane ID of the current lane" specifically include: determining whether the candidate lane is a virtual lane; when it is determined that it is a virtual lane, obtaining a virtual lane that matches the pose of the current vehicle, if the number of matching virtual lanes is 1, then the lane ID of the matched virtual lane is used as the lane ID of the lane where the vehicle is currently located; if the number of matched virtual lanes is not 1, initialization processing is performed based on the current vehicle position, and the starting lane of the vehicle is redetermined. , use the lane ID of the starting lane as the lane ID of the lane where the vehicle is currently located; when it is determined that it is not a virtual lane, obtain the lane that has a connected relationship with the lane corresponding to the last lane ID from the candidate lane; if If the number of lanes with a connected relationship is 1, then the lane ID of the lane with a connected relationship is used as the lane ID of the lane where the vehicle is currently located; if the number of lanes with a connected relationship is not 1, then the lane ID of the lane with a connected relationship is used as the lane ID of the lane where the vehicle is currently located; if the number of lanes with a connected relationship is not 1, then the lane ID of the lane with a connected relationship is used The position is initialized, the starting lane of the vehicle is re-determined, and the lane ID of the starting lane is used as the lane ID of the lane where the vehicle is currently located.

In a technical solution of the above lane positioning method, the method further includes: based on the current vehicle position and Carry out initialization processing through the following steps to determine the starting lane of the vehicle: obtain the lane located within the preset range centered on the current vehicle position; determine whether the current vehicle position falls within the map range corresponding to a lane within the preset range within; when it falls within the map range corresponding to only one lane within the preset range, obtain the distance between the current vehicle position and the boundary of the lane within the preset range; if the distance is greater than is equal to the preset distance threshold, then the lane within the preset range is used as the starting lane of the vehicle; if the distance is less than the preset distance threshold, lane positioning abnormal prompt information is output; when there is no When a lane within the preset range is within the map range corresponding to the lane, lane positioning abnormality prompt information is output.

In a technical solution of the above lane positioning method, the step of "locating the vehicle through a two-dimensional map and determining the vehicle position" specifically includes: positioning the current driving road of the vehicle through a two-dimensional map and determining where the road point in the current driving road is. Two-dimensional coordinates of the two-dimensional rectangular coordinate system; obtain the relative height of the road point relative to the vehicle, and determine the origin, x-axis, and y-axis of the world coordinate system based on the origin, x-axis, and y-axis of the two-dimensional rectangular coordinate system , use the relative height as the z-axis coordinate of the road point in the world coordinate system, and back-project the road point from the world coordinate system to the image coordinate system according to the two-dimensional coordinates of the road point and the relative height. ; Determine the road elements located around the vehicle, obtain the back-projection position of the road point at the location of the road element to the image coordinate system and obtain the position of the road element determined by the image acquisition device of the vehicle. The road point is at the collection position of the image coordinate system; the back-projection position of the road point at the location of the road element is position matched with the collection position, and the position of the vehicle in the world coordinate system is determined based on the position matching result.

In a technical solution of the above lane positioning method, the step of "obtaining the relative height of the road point relative to the vehicle" specifically includes: obtaining the plane vector of the plane where the vehicle is located in the vehicle body coordinate system and obtaining the plane vector according to the plane vector. The unit normal vector of the plane; according to the two-dimensional coordinates of the road point and by solving the following equations, it is obtained that the origin, x-axis and y-axis of the two-dimensional rectangular coordinate system are the origin, x-axis and The coordinate zh of the z-axis in the world coordinate system of the y-axis: (xh-x1)×nx+(yh-y1)×ny+(zh-z1)×nz=0, where x1, y1 and z1 represent the vehicle body coordinates The coordinates of the x-axis, y-axis and z-axis in the system, nx, ny and nz represent the coordinates of the unit normal vector on the x-axis, y-axis and z-axis in the vehicle body coordinate system, xh and yh represent the coordinates of the road point The coordinates of the x-axis and the y-axis in the two-dimensional coordinates; the relative height of the road point relative to the vehicle is determined based on the coordinates zh.

In a technical solution of the above lane positioning method, the method further includes displaying road connections on the two-dimensional map of the road currently traveled by the vehicle and the roads it has traveled before.

In one technical solution of the above lane positioning method, the step of "matching the back-projection position of the road point at the location of the road element with the collection position, and determining the position of the vehicle in the world coordinate system based on the result of the position matching" Specifically, the method includes: position matching the back-projected position of the road point at the position of the road element with the acquisition position to obtain pose parameters for conversion between the device coordinate system of the image acquisition device and the world coordinate system; and obtaining the location of the vehicle. The road point at the location is back-projected to the back-projection position of the image coordinate system; according to the back-projection position of the road point at the location of the vehicle and the pose parameter, determine where the vehicle is sitting in the world. The position of the standard system.

In one technical solution of the above lane positioning method, "the back-projection position of the road point at the position of the road element is matched with the acquisition position, and the device coordinate system of the image acquisition device and the world coordinate system are converted. The step of "pose parameter" specifically includes: establishing a distance error equation shown in the following formula based on the back-projection position and collection position of the road point at the position of the road element: loss=d(f(A,O),cd ), where f(A,O) represents the back-projected position of the current road point, A represents the position of the current road point in the world coordinate system, O represents the pose parameter for conversion between the device coordinate system and the world coordinate system, and cd represents the The collection positions of the two road points closest to the back-projection position of the current road point are the line segments formed by the endpoints. d represents the distance calculation function from the back-projection position f (A, O) to the line segment cd, and loss represents the distance calculation function. d calculated distance; with the goal of making the distance loss less than the preset distance threshold, the Levenberg-Marquardt algorithm is used to perform the pose parameter O in the distance error equation. Iteratively optimize, and obtain the pose parameter O when the distance loss is less than the preset distance threshold, and use the pose parameter O as the final position for conversion between the device coordinate system of the image acquisition device and the world coordinate system. posture parameters.

In a technical solution of the above lane positioning method, in "determining the road elements located around the vehicle, obtaining the back-projected position of the road point at the location of the road element to the image coordinate system and obtaining the image acquisition device passing through the vehicle Before determining that the road point at the position of the road element is at the acquisition position of the image coordinate system, the method further includes: using at least two image acquisition devices to collect the road point at the position of the road element. Image collection is performed to obtain a road point image of the road point at the location of the road element; and based on the road point image, the collection position of the road point at the location of the road element in the image coordinate system is determined.

In a second aspect, a computer device is provided. The computer device includes a processor and a storage device. The storage device is adapted to store a plurality of program codes. The program codes are adapted to be loaded and run by the processor to execute the above. The lane positioning method described in any one of the technical solutions of the lane positioning method.

In a third aspect, a computer-readable storage medium is provided, which stores a plurality of program codes, and the program codes are suitable for being loaded and run by a processor to perform the technical solution of the above lane positioning method. The lane positioning method described in any technical solution.

In a fourth aspect, a vehicle is provided, which vehicle includes the computer device described in the above computer device technical solution.

One or more of the above technical solutions of the present invention have at least one or more of the following beneficial effects:

In implementing the technical solution of the present invention, the vehicle can be positioned first through a two-dimensional map to determine the vehicle position (the vehicle position is two-dimensional information, for example, the vehicle position can include the coordinates of the x-axis and y-axis of the vehicle in the world coordinate system) , and then obtain the lanes around the vehicle position and the connectivity between the lanes (if you drive from one lane to another normally, the two lanes have a connectivity, otherwise the two lanes do not have a connectivity), according to the obtained The arriving lane and the connection relationship determine the lane ID of the lane where the vehicle is currently located. Through the above technical solution, even if the vehicle is driving in a road network with road layers of different heights, it can pass the two-dimensional map and according to the lane The connection relationship between them accurately determines which lane the vehicle is driving on.

Furthermore, in some technical solutions for implementing the present invention, in order to reduce the consumption of computing resources, when determining the lane ID of the lane where the vehicle is currently located based on the lanes around the vehicle position and the connectivity between lanes, the lane ID of the last time can be determined in turn. The map range where the corresponding lane falls on the two-dimensional map, the map range where other lanes that belong to the same road layer as the lane corresponding to the last lane ID and have a connectivity relationship with the lane corresponding to the last lane ID fall on the two-dimensional map, and The lanes located within the preset range centered on the current vehicle position fall within the map range on the two-dimensional map, and are combined with the current vehicle position to determine the lane ID of the lane where the vehicle is currently located. In the above process, as long as the lane ID of the lane where the vehicle is currently located is determined based on a certain map range and the current vehicle position, there is no need to continue subsequent analysis, which can greatly reduce the consumption of computing resources and improve the efficiency of lane positioning. .

Furthermore, in some technical solutions for implementing the present invention, when lane positioning is just started, initialization processing can be performed based on the vehicle position to determine the starting lane of the vehicle, and the lane ID of the starting lane is used as the first lane ID of the lane where the vehicle is located. At the same time, in the process of determining the lane ID of the lane where the vehicle is currently located based on the lanes around the vehicle position and the connectivity between lanes, if the lane ID of the lane where the vehicle is currently located cannot be determined in the end, initialization processing can also be performed based on the vehicle location. , determine the starting lane of the vehicle, and use the lane ID of the starting lane as the lane ID of the lane where the vehicle is located.

Further, in some technical solutions for implementing the present invention, the current driving road of the vehicle can be positioned through a two-dimensional map, and the two-dimensional coordinates of the road points on the current driving road in the two-dimensional rectangular coordinate system (including the two-dimensional rectangular coordinate system of the vehicle) can be determined. coordinates of the x-axis and y-axis in the system), and at the same time obtain the relative height of the road point relative to the vehicle, and determine the origin, x-axis and y-axis of the world coordinate system based on the origin, x-axis and y-axis of the two-dimensional rectangular coordinate system, that is, The coordinates of the x-axis and y-axis in the above two-dimensional coordinates are used as the coordinates of the x-axis and y-axis of the road point in the world coordinate system (three-dimensional rectangular coordinate system), and then the above-mentioned relative height is used as the coordinates of the road point on the z-axis in the world coordinate system. coordinates, and then back-project the road point from the world coordinate system to the image coordinate system according to the above-mentioned two-dimensional coordinates of the road point and the above-mentioned relative height. After the road points are back-projected to the image coordinate system, the road points at the positions of the road elements around the vehicle can be back-projected to the back-projected positions of the image coordinate system and the road elements determined by the image acquisition device of the vehicle The road point at the location is position matched at the acquisition position of the image coordinate system, and then the position of the vehicle in the world coordinate system is determined based on the position matching results.

Through the above method, even if the vehicle's on-board map is a two-dimensional map, and the global navigation satellite system cannot be used for positioning, the vehicle's position in the world coordinate system can still be obtained through the two-dimensional map, and it can also be adjusted according to the density of road points. The positioning accuracy of the vehicle (the higher the density, the higher the positioning accuracy, the lower the density, the lower the positioning accuracy), achieving centimeter-level positioning of the vehicle, and no positioning drift will occur after long-term positioning.

Furthermore, in the technical solution for implementing the present invention, when a vehicle is traveling on a road network that includes multiple different road layers, the road connections that the vehicle is currently traveling on and the roads it has traveled before can be displayed on a two-dimensional map. After the roads are connected, it is possible to exclude road pairs in other road layers using 2D The impact of vehicle positioning display on the map can more clearly show which road and where the vehicle is located.

Description of the drawings

The disclosure of the present invention will become more understandable with reference to the accompanying drawings. Those skilled in the art can easily understand that these drawings are for illustrative purposes only and are not intended to limit the scope of the present invention. Additionally, like numbers in the figures identify similar parts, where:

Figure 1 is a schematic flowchart of the main steps of a lane positioning method according to an embodiment of the present invention;

Figure 2 is a schematic diagram of a road with road layers at different heights according to an embodiment of the present invention;

Figure 3 is a schematic flowchart of the main steps of a vehicle positioning method according to an embodiment of the present invention;

Figure 4 is a schematic diagram of the plane vector a and the unit normal vector n of the plane where the vehicle is located in the vehicle body coordinate system according to an embodiment of the present invention;

Figure 5 is a schematic diagram of a line segment formed by the back-projection position a of a road point and the collection positions c and d of the two road points closest to the back-projection position a as endpoints according to an embodiment of the present invention;

Figure 6 is a schematic diagram of a road network including multiple different road layers according to an embodiment of the present invention.

Figure 7 is a schematic diagram of a virtual lane according to an embodiment of the present invention;

Figure 8 is a schematic diagram 2 of a virtual lane according to an embodiment of the present invention;

Figure 9 is a schematic diagram three of a virtual lane according to an embodiment of the present invention;

Figure 10 is a schematic flowchart of the main steps of a starting lane determination method according to an embodiment of the present invention.

Detailed ways

Some embodiments of the invention are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of the present invention.

In the description of the present invention, "processor" may include hardware, software, or a combination of both. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. The processor has data and/or signal processing functions. The processor can be implemented in software, hardware, or a combination of both. Computer-readable storage media includes any suitable medium that can store program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and/or B" means all possible combinations of A and B, such as just A, just B, or A and B. The terms "at least one A or B" or "at least one of A and B" have a similar meaning to "A and/or B" and may include just A, just B or A and B. The singular forms "a," "the" and "the" may also include the plural form.

Here we first explain some terms involved in the present invention.

A two-dimensional map is a map represented on a plane in a two-dimensional form. In the embodiment of the present invention, by positioning the target object on the two-dimensional map, the two-dimensional coordinates of the target object in the two-dimensional rectangular coordinate system can be obtained, that is, the target The coordinates of the object on the x-axis and y-axis in the two-dimensional Cartesian coordinate system.

The world coordinate system refers to a three-dimensional rectangular coordinate system with a point in three-dimensional space as the origin.

The image coordinate system is a two-dimensional coordinate system with the center of the image plane formed as the origin when the image acquisition device performs image acquisition. Cartesian coordinate system.

The device coordinate system of the image acquisition device is a three-dimensional rectangular coordinate system established with the focus center of the image acquisition device as the origin and the optical axis as the z-axis. It should be noted that the device coordinate system of the image acquisition device has the same meaning as the camera coordinate system in the field of computer vision technology, and the device coordinate system will not be described again here.

The pose parameters (intrinsic parameters) converted by the image coordinate system and the device coordinate system can project the points in the image coordinate system to the device coordinate system, and can also back-project the points in the device coordinate system to the image coordinate system. The pose parameters (external parameters) converted by the device coordinate system and the world coordinate system can project the points in the device coordinate system to the world coordinate system, and can also back-project the points in the world coordinate system to the device coordinate system. Further, If you use the above internal and external parameters at the same time, you can back-project the points in the world coordinate system to the image coordinate system. It should be noted that the above-mentioned internal parameters and external parameters have the same meaning as the camera internal parameters and camera external parameters in the field of computer vision technology. Both include rotation matrices and translation vectors, which will not be described again here.

The vehicle body coordinate system refers to a three-dimensional coordinate system. Its origin is at the point where the center of mass of the carrier is fixedly connected to the carrier. The x-axis points to the right along the axial direction of the carrier, the y-axis points forward, and the z-axis, x-axis, and y-axis satisfy the right-hand rule and point toward the sky. , the vehicle body coordinate system can also be called the "right-front-up (R-F-U)" coordinate system.

Referring to Figure 1, Figure 1 is a schematic flowchart of the main steps of a lane positioning method according to an embodiment of the present invention. As shown in Figure 1, the lane positioning method in the embodiment of the present invention mainly includes the following steps S101 to S103.

Step S101: Position the vehicle through the two-dimensional map and determine the initial vehicle position.

A two-dimensional map is a map represented on a plane in a two-dimensional form. By positioning a vehicle on a two-dimensional map, the two-dimensional coordinates of the vehicle in the two-dimensional rectangular coordinate system corresponding to the two-dimensional map can be obtained. The two-dimensional coordinates Including the vehicle's x-axis and y-axis coordinates in the two-dimensional Cartesian coordinate system.

Step S102: Obtain the lanes around the vehicle position and the connectivity between lanes.

If a vehicle can exit one lane and enter another, then the two lanes are connected, that is, the two lanes are connected.

In the embodiment of the present invention, the position of each lane in the road network and the connectivity relationship between different lanes can be determined in advance. After the vehicle position is determined, the lanes surrounding it can be queried based on the vehicle position, and then the number of lanes among these lanes can be queried. connectivity between. In addition, each lane has its own unique lane ID to easily distinguish different lanes.

Step S103: Determine the lane ID of the lane where the vehicle is currently located based on the lanes around the vehicle position and the connectivity between lanes.

When a vehicle travels in a road network that contains multiple road layers with different heights, the heights of different roads cannot be displayed because the two-dimensional map lacks the coordinates of the z-axis in the world coordinate system. If there are multiple roads belonging to road layers of different heights at the same location, these roads will be displayed crosswise on the 2D map, and when the vehicle is driving at this location, it will be impossible to determine which road the vehicle is driving on through the 2D map. on the driveway. However, even if a plurality of roads belonging to road layers of different heights are cross-displayed at the same position on the two-dimensional map, A vehicle must only drive on a lane with a connected relationship, and will not drive from the current lane into another lane that has no connected relationship with it. Therefore, the embodiment of the present invention obtains the lane with a connected relationship based on the connected relationship between the lanes. According to These connected lanes determine which lane the vehicle is driving on, and then obtain the lane ID of the lane where the vehicle is currently located. As shown in Figure 2, the vehicle is driving on the lane of this layer of road, and there is an upper layer of road above this layer of road. On the two-dimensional map, the current layer of road and the upper layer of road will cross and display at a certain position in front of the vehicle. Since the lanes of the road on this layer are not connected to the lanes of the road on the upper layer, even if the vehicle travels to the above position, it can be determined that the vehicle will not drive into the upper road and will continue to drive on the lane of the road on this layer.

Through the method described in steps S101 to S103 above, even if the vehicle is driving in a road network with road layers of different heights, it can accurately determine where the vehicle is driving through the two-dimensional map and based on the connectivity between lanes. On a driveway.

The above steps S101 and S103 will be further described below respectively.

First, the above-mentioned step S101 will be further explained.

Referring to FIG. 3 , in some implementations of the above step S101 , the vehicle can be positioned through a two-dimensional map and according to the following steps S1011 to S1015 to determine the initial vehicle position.

Step S1011: Locate the current driving road of the vehicle through the two-dimensional map, and determine the two-dimensional coordinates of the road points on the current driving road in the two-dimensional rectangular coordinate system.

A two-dimensional map is a map represented in a two-dimensional form on a plane. By locating the current driving road of the vehicle on the two-dimensional map, the two-dimensional rectangular coordinate system corresponding to the two-dimensional map of each road point on the current driving road can be obtained. The two-dimensional coordinates in include the coordinates of the x-axis and y-axis of the road point in the two-dimensional Cartesian coordinate system.

Step S1012: Obtain the relative height of the road point relative to the vehicle.

The relative height refers to the relative height of the road point relative to the vehicle in the world coordinate system determined by the origin, x-axis and y-axis of the two-dimensional rectangular coordinate system in step S1011, along the z-axis direction of the world coordinate system. .

Step S1013: Determine the origin, x-axis, and y-axis of the world coordinate system based on the origin, x-axis, and y-axis of the two-dimensional rectangular coordinate system, and use the relative height as the coordinate of the z-axis of the road point in the world coordinate system. According to the Two-dimensional coordinates and relative heights back-project road points from the world coordinate system to the image coordinate system.

In this embodiment, an image acquisition device is provided on the vehicle. The device parameters of the image acquisition device include internal parameters and external parameters. The internal parameters refer to the posture parameters converted from the image coordinate system and the device coordinate system. The external parameters refer to the device coordinate system and the device coordinate system. The pose parameters for transformation into the world coordinate system. After obtaining the two-dimensional coordinates and relative height of the road point, the internal and external parameters of the image acquisition device can be used to back-project the road point from the world coordinate system to the image coordinate system.

In addition, in some embodiments, an image acquisition device whose image coordinate system is not a two-dimensional rectangular coordinate system can also be used. After image acquisition by such an image acquisition device to obtain a corresponding image, the image acquisition device itself can use the image The coordinate system conversion relationship between the coordinate system and the two-dimensional rectangular coordinate system is to convert the above image into the two-dimensional rectangular coordinate system, and then execute the method described in step S1013. For example, image collection The image coordinate system of the collection device can be a three-dimensional rectangular coordinate system, etc., and the image acquisition device can be a three-dimensional camera, radar, etc. In the embodiment of the present invention, radar includes but is not limited to millimeter-wave radar (Millimeter-wave Radar) and laser radar (Laser Radar). In a preferred embodiment, the radar may be laser radar.

Step S1014: Determine the road elements located around the vehicle, obtain the back-projected position of the road point at the position of the road element to the image coordinate system, and obtain the road point at the position of the road element determined by the image acquisition device of the vehicle. The acquisition position in the image coordinate system.

Road elements at least include traffic signs on the road and/or other objects that can play a marking role. Traffic signs at least include lane lines, stop lines, road signs (such as left turn arrows), traffic lights and traffic signs, etc. , other objects that can play a marking role include at least rod-shaped objects.

It should be noted that in the embodiment of the present invention, an image recognition method may be used to identify road elements and determine the location of the road elements. The embodiment of the present invention does not specifically limit the identification method of road elements, as long as the road elements can be identified and the location of the road elements can be obtained by identifying images (two-dimensional maps and road images collected by the vehicle's image acquisition device). Can. For example, a road element recognition model based on a neural network can be used to identify road elements in images.

After the road point is back-projected from the world coordinate system to the image coordinate system, a back-projection point corresponding to the road point will be formed in the image coordinate system. The position of the back-projection point in the image coordinate system is the back-projection position of the road point.

The image acquisition device of the vehicle collects images of road points located at the positions of road elements around the vehicle to obtain road point images of road points located at the positions of road elements around the vehicle. The road can be determined based on these road point images. The road point at the location of the element is at the collection position of the image coordinate system.

Furthermore, in some embodiments, in order to prevent the image acquisition device from malfunctioning or being blocked, resulting in the inability to collect images of road points at the locations of road elements around the vehicle, and thus from being unable to determine the position of the vehicle, at least two image acquisition methods can be used. The device collects images of road points at the locations of road elements around the vehicle to improve the reliability of image collection and vehicle positioning. Specifically, in these embodiments, images of road points located at positions of road elements around the vehicle can be collected through at least two image acquisition devices to obtain road point images of road points located at positions of road elements around the vehicle, Then, based on the road point image, the collection position of the road point located at the position of the road element around the vehicle in the image coordinate system is determined. Before collecting images of road points through at least two image collecting devices, these image collecting devices need to be calibrated in order to unify the device coordinate systems of these image collecting devices. In the embodiment of the present invention, a conventional calibration method of multiple image acquisition devices can be used for calibration, and the embodiment of the present invention does not specifically limit this.

Step S1015: Perform position matching on the back-projection position of the road point at the location of the road element and the collection position, and determine the position of the vehicle in the world coordinate system based on the position matching result.

The internal parameters of the image acquisition device are inherent parameters of the image acquisition device, which are determined after the production and deployment of the image acquisition device is completed, while the external parameters of the image acquisition device can be adjusted. Therefore, whether the external parameters of the image acquisition device are accurate will greatly affect the conversion between the device coordinate system of the image acquisition device and the world coordinate system. accuracy. In the embodiment of the present invention, the purpose of position matching the back-projected position of the road point located at the location of the road element around the vehicle and the acquisition position is to obtain accurate external parameters. After obtaining the accurate external parameters, the image acquisition device can be used at the same time The internal parameters and external parameters project the back-projection position of the road point in the image coordinate system from the image coordinate system to the world coordinate system (project the road point from the image coordinate system to the device coordinate system according to the internal parameters, and then project the road point from the device coordinate system according to the external parameters). coordinate system projected to the world coordinate system), and further select the position of the road point at the vehicle's location in the world coordinate system as the vehicle's position in the world coordinate system. Specifically, in some embodiments, the back-projection position of the road point located at the position of the road element around the vehicle can be position matched with the acquisition position to obtain the position of the device coordinate system of the image acquisition device and the world coordinate system. pose parameters; and then obtain the back-projection position of the road point at the vehicle's location to the image coordinate system, and determine the vehicle's position in the world coordinate system based on the back-projection position and pose parameters of the road point at the vehicle's location.

Through the method described in the above steps S1011 to S1015, the position of the vehicle in the world coordinate system can be obtained through a two-dimensional map. Therefore, even if the on-board map of the vehicle is a two-dimensional map, the global navigation satellite system cannot be used for positioning. The position of the vehicle in the world coordinate system can also be obtained. In addition, the density of road points is different, the distance between adjacent road points in the world coordinate system is also different, and the number of road points at the location of the vehicle is also different. The spacing and number of road points will affect the accuracy of the vehicle's position in the world coordinate system (positioning accuracy), and the higher the density, the higher the positioning accuracy, and the lower the density, the lower the positioning accuracy. Therefore, in this embodiment, the positioning accuracy of the vehicle in the world coordinate system can also be adjusted by adjusting the density of road points, thereby achieving centimeter-level positioning of the vehicle when the global navigation satellite system cannot be used for positioning. Positioning drift will not occur after long-term positioning.

The above steps S1012 and S1015 will be further described below.

First, the above step S1012 will be further described.

In some implementations according to the above step S1012, since the posture of the vehicle changes with the change of the driving road (for example, the plane where the driving road is located has a certain angle with the preset level, then the vehicle has a certain angle with the preset level. (the same angle), therefore, you can first establish a vehicle body coordinate system based on the vehicle (the vehicle body coordinate system is a local coordinate system relative to the world coordinate system), and use the vector formed by the road point and the origin of the vehicle body coordinate system The relative height of the road point relative to the vehicle is calculated based on the principle that the normal vector of the plane where the vehicle is located in the vehicle body coordinate system is perpendicular to each other and the vector product is zero. Specifically, in this embodiment, the relative height of the road point relative to the vehicle can be obtained through the following steps S10121 to S10123.

Step S10121: Obtain the plane vector of the plane where the vehicle is located in the vehicle body coordinate system and obtain the unit normal vector of the plane based on the plane vector.

In the vehicle body coordinate system, the plane where the vehicle is located refers to the plane formed by the x-axis and y-axis in the vehicle body coordinate system. As shown in Figure 4, x ₁ and z ₁ represent the x-axis and z-axis in the vehicle body coordinate system respectively, a represents the plane vector of the plane where the vehicle is located, and n represents the unit normal vector of the plane where the vehicle is located. x ₂ and z ₂ respectively represent the x axis and z axis in the "world coordinate system with the origin, x axis and y axis of the two-dimensional rectangular coordinate system as the origin, x axis and y axis respectively".

If the vehicle is equipped with devices that rely on data from the vehicle body coordinate system, these devices can be used directly to obtain the plane vector of the plane where the vehicle is located. For example, the plane vector of the plane where the vehicle is located in the vehicle body coordinate system can be directly obtained through the accelerometer.

After obtaining the plane vector of the plane where the vehicle is located, conventional calculation methods for normal vectors in the field of mathematics and technology can be used to calculate the unit normal vector of the plane where the vehicle is located based on the plane vector of the plane where the vehicle is located, which will not be described in detail here.

Step S10122: According to the two-dimensional coordinates of the road point and by solving the following equations, the z-axis of the road point in the world coordinate system with the origin, x-axis and y-axis of the two-dimensional rectangular coordinate system as the origin, x-axis and y-axis respectively Coordinates zh.

(xh-x1)×nx+(yh-y1)×ny+(zh-z1)×nz=0

In the above equation, x1, y1 and z1 represent the coordinates of the vehicle on the x-axis, y-axis and z-axis in the vehicle body coordinate system, nx, ny and nz represent the unit normal vectors on the x-axis, y-axis and The coordinates of the z-axis, xh and yh represent the coordinates of the x-axis and y-axis in the two-dimensional coordinates of the road point.

(xh-x1, yh-y1, zh-z1) actually represents the vector formed by the two points of the vehicle and the road point in the vehicle body coordinate system, (nx, ny, nz) represents the vehicle in the vehicle body coordinate system The unit normal vector of the plane, the vector (xh-x1, yh-y1, zh-z1) and the vector (nx, ny, nz) are perpendicular, so the vector product of these two vectors is equal to zero. Among them, xh and yh can be obtained directly from the two-dimensional coordinates of the road point, and the value of the coordinate zh can be obtained by solving the above equation.

Step S10123: Determine the relative height of the road point relative to the vehicle according to the coordinate zh.

Through the method described in the above steps S10121 to S10123, the relative height of the road point relative to the vehicle can be accurately obtained, which is conducive to accurately converting the road point from the world coordinate system to the image coordinate system.

Step S1015 will be further described below.

In some implementations according to the above step S1015, position matching is performed on the back-projection position of the road point located at the position of the road element around the vehicle and the acquisition position to obtain the device coordinate system of the image acquisition device and the world coordinate system for conversion. When the pose parameter is used, this pose parameter can be used as an equation parameter, and a distance error equation is established based on the back-projected position of the road point located at the position of the road element around the vehicle and the acquisition position (the equation parameters of the distance error equation include the above pose parameters ), and then use a nonlinear iterative optimization algorithm to iteratively optimize the pose parameters in the distance error equation with the goal that the distance error equation satisfies the preset convergence conditions, and obtain the optimized pose parameters as the final image acquisition device The pose parameters for conversion between the device coordinate system and the world coordinate system. Specifically, in this embodiment, the following steps S10151 to S10152 can be used to perform position matching on the back-projected position of the road point located at the location of the road element around the vehicle and the collection position:

Step S10151: Establish a distance error equation shown in the following formula based on the back-projection position and collection position of the road point located at the position of the road element around the vehicle:

loss=d(f(A,O),cd)

In the above distance error equation, f(A,O) represents the back-projection position of the current road point, A represents the position of the current road point in the world coordinate system, and O represents the pose parameter for conversion between the device coordinate system and the world coordinate system. cd represents the line segment formed by taking the collection positions of the two road points closest to the back-projected position of the current road point as endpoints, d represents the distance calculation function from the back-projected position f (A, O) to the line segment cd, and loss represents the distance calculation The distance calculated by function d. As shown in Figure 5, point a in Figure 5 represents the back-projection point when the road point at the location of the road element is back-projected to the image coordinate system. Points c and d respectively represent the location of the road element determined through the image acquisition device. The road points at are the two closest points to the back-projection point a in the image coordinate system.

Step S10152: With the goal of making the distance loss smaller than the preset distance threshold, use the Levenberg-Marquardt algorithm to iteratively optimize the pose parameter O in the distance error equation, and obtain the distance The pose parameter O when the loss is less than the preset distance threshold, the pose parameter O is used as the pose parameter for conversion between the device coordinate system of the final image acquisition device and the world coordinate system.

It should be noted that the Levenberg-Marquardt algorithm is a conventional nonlinear least squares algorithm in the field of mathematical technology, and its algorithm principle and calculation process will not be described in detail here.

The above is a further description of step S1015.

When a vehicle travels in a road network containing multiple different road layers, the height of different roads cannot be displayed because the two-dimensional map lacks the coordinates of the z-axis in the world coordinate system. As shown in Figure 6, when there are multiple roads belonging to different road layers at the same location, these roads will be displayed crosswise on the two-dimensional map, making it impossible to clearly show which road the vehicle is on. In a vehicle positioning method according to another embodiment of the present invention, in addition to the steps S1011 to S1015 in the foregoing method embodiment, the vehicle positioning method may also map the vehicle on a two-dimensional map before or after each step. The current driving road and the previously traveled road are displayed for road connectivity, so that the influence of roads in other road layers on the vehicle positioning display using a two-dimensional map can be eliminated, and the vehicle can be more clearly displayed on which road and which location. superior. In a preferred embodiment, before back-projecting the road point from the world coordinate system to the image coordinate system according to the two-dimensional coordinates and relative height of the road point, the vehicle's current driving road and the previously traveled road can be mapped on the two-dimensional map. Display road connections.

The above is the description of the above-mentioned step S101. Next, step S103 shown in FIG. 1 will be further described.

There may be multiple lanes around the vehicle. If the connectivity relationship and location matching are performed on each lane separately, it will significantly increase the consumption of computing resources. Therefore, in order to reduce the consumption of computing resources, you can first locate the lane corresponding to the last lane ID. The map range on the two-dimensional map and the current position of the vehicle are used to determine the lane ID of the lane where the vehicle is currently located. If the lane ID cannot be determined based on the map range of the last lane ID corresponding to the lane on the two-dimensional map, then continue to other lanes. Perform connectivity relationship judgment and location matching.

Specifically, in some implementations of the above-mentioned step S103, the lane ID of the lane where the vehicle is currently located may be determined through the following steps S1031 to S1033 and based on the lanes around the vehicle position and the connectivity between lanes.

Step S1031: Determine whether the last lane ID can be determined based on the last determined vehicle position; if the last lane ID can be determined, go to step S1032; if the last lane ID cannot be determined, go to Step S1033.

Step S1032: Determine the lane ID of the lane where the vehicle is currently located based on the map range where the lane corresponding to the last lane ID falls on the two-dimensional map and the current vehicle position.

The map range where the lane falls on the two-dimensional map refers to the map range of the lane on the two-dimensional map determined based on the actual position of the lane according to the map scale of the two-dimensional map.

If the current vehicle position is in the map range where the lane corresponding to the last lane ID falls on the two-dimensional map, it means that the vehicle is very likely to continue driving in the lane corresponding to the last lane ID, and the lane ID of the lane where the vehicle is currently located is still the same as the last time. lane ID; if the current vehicle position is not located in the map range where the lane corresponding to the last lane ID falls on the two-dimensional map, it means that the vehicle has left the lane corresponding to the last lane ID. At this time, it can be obtained from other locations around the vehicle position. The lane ID in the lane determines the lane the vehicle is currently in.

Further, in some preferred implementations of step S1032, the following steps 11 to 13 can be used, and based on the map range of the last lane ID corresponding to the lane falling on the two-dimensional map and the current vehicle position, the location of the lane where the vehicle is currently located can be determined. Lane ID.

Step 11: Determine whether the current vehicle position falls within the map range corresponding to the last lane ID and the two-dimensional map only contains one lane within the map range; if the current vehicle position falls within the map range corresponding to the last lane ID and If the 2D map contains only one lane within the map range, go to step 12; otherwise, go to step 13.

Step 12: Use the last lane ID as the lane ID of the current lane where the vehicle is located.

If the current vehicle position is within the map range where the lane corresponding to the last lane ID falls on the two-dimensional map, it means that the vehicle is very likely to continue driving in the lane corresponding to the last lane ID. Furthermore, if the two-dimensional map contains only one lane within the map range, it means that the vehicle must continue driving in the lane corresponding to the last lane ID. Therefore, the last lane ID can be directly used as the lane ID of the current lane of the vehicle.

Step 13: Obtain other lanes that belong to the same road layer as the lane corresponding to the last lane ID and have a connectivity relationship with the lane corresponding to the last lane ID, and based on the map range of the other lanes falling on the two-dimensional map and the current vehicle Position, determine the lane ID of the lane the vehicle is currently in.

If the current vehicle position is not located within the map range corresponding to the last lane ID, it may be located in the same road layer and other nearby lanes fall within the map range on the two-dimensional map. Therefore, these other lanes can be located within the two-dimensional map range. The map range on the map and the current vehicle position determine the lane ID of the lane where the vehicle is currently located.

If the current vehicle position is within the map range corresponding to the last lane ID, but the two-dimensional map within this map range contains multiple lanes, it means that the vehicle may be located in an intersection area of road layers of different heights (such as an overpass). If the connectivity relationship and position matching are performed on the lanes of each road layer, it will significantly increase the consumption of computing resources. Therefore, in order to reduce the consumption of computing resources, you can first check the "vehicles corresponding to the last lane ID". "Other lanes that belong to the same road layer and have a connected relationship with the lane corresponding to the last lane ID" are used for location matching. If the current vehicle position is within the map range corresponding to the above-mentioned other lanes, it means that the vehicle is very likely to be on the above-mentioned other lanes. Driving, the lane ID of the lane where the vehicle is currently located can be determined.

By matching the "map range corresponding to the last lane ID" with the current vehicle position through the method described in steps 11 to 13 above, it can be accurately determined whether the vehicle continues to drive in the lane corresponding to the last lane ID.

Further, in some preferred implementations of step 13 above, the lane ID of the lane where the vehicle is currently located can be determined through the following steps 131 to 133 and based on the map range of the other lanes falling on the two-dimensional map and the current vehicle position. .

Step 131: Determine whether the current vehicle position only falls within the map range corresponding to one of the other lanes; if it only falls within the map range corresponding to one of the other lanes, go to step 132; otherwise, go to step 133.

Step 132: Use the lane ID of the other lane as the lane ID of the lane where the vehicle is currently located.

If the current vehicle position is within the map range where the other lane falls on the two-dimensional map, it means that the vehicle is very likely to be traveling on the other lane. Furthermore, if the current vehicle position is only within the map range corresponding to one of the other lanes, it means that the vehicle must be driving on this other lane. Therefore, the lane ID of this other lane can be directly used as the lane ID of the lane where the vehicle is currently located. .

Step 133: Determine the lane ID of the lane where the vehicle is currently located based on the map range where the lane located within the preset range centered on the current vehicle position falls on the two-dimensional map and the current vehicle position.

When the current vehicle position is not located within the map range where the other lanes fall on the two-dimensional map, the search range of the lane is expanded by acquiring lanes located within the preset range centered on the current vehicle position, and the vehicle position is increased. The number of surrounding roads from which the vehicle's current lane can be determined.

Through the method described in steps 131 to 133 above, "other lanes that belong to the same road layer as the lane corresponding to the last lane ID and have a connectivity relationship with the lane corresponding to the last lane ID" are matched with the current vehicle position, It can accurately determine whether the vehicle is driving on the other lanes mentioned above.

Further, in some preferred implementations of the above step 133, after determining that the lane located within the preset range centered on the current vehicle position falls within the map range on the two-dimensional map, it is possible to obtain how many lanes the current vehicle position falls within. Within the map range of lanes, different methods are used to determine the lane ID of the lane where the vehicle is currently located based on the number of lanes.

Specifically, if the current vehicle position only falls within the map range corresponding to a lane within a preset range, it means that the vehicle must be driving on this lane. At this time, the lane ID of this lane can be used as the lane ID of the vehicle's current lane. Lane ID.

If the current vehicle position falls within the map range corresponding to multiple lanes within the preset range, these lanes can be used as candidate lanes at this time, and the lanes and lanes that have a connectivity relationship with the lane corresponding to the last lane ID are used. /or the pose of the current vehicle, filter out the lane ID of the lane where the vehicle is currently located from the lane ID of the candidate lane. Specifically, the candidate lane among the candidate lanes that has a connected relationship with the lane corresponding to the last lane ID can be used as the lane where the vehicle is currently located. At the same time, a candidate lane that matches the pose of the current vehicle can also be selected from the candidate lanes as the lane where the vehicle is currently located. In some cases, a candidate lane that has a connected relationship with the lane corresponding to the last lane ID and matches the pose of the current vehicle can also be used as the lane where the vehicle is currently located. Among them, the pose of the vehicle refers to the position and attitude (Pose) of the vehicle. For example, if two candidate lanes facing left and right are determined based on the position of the vehicle, and the posture of the vehicle is turning left, then it can be determined that the candidate lane facing left matches the posture of the current vehicle. The candidate lane facing left can be regarded as the current lane of the vehicle.

Furthermore, in some preferred embodiments, the following steps 1331 to 1335 can be used to filter the lane IDs of the candidate lanes based on the lanes that have a connected relationship with the lane corresponding to the last lane ID and/or the pose of the current vehicle. Lane ID of the lane where the vehicle is currently located.

Step 1331: Determine whether the candidate lane is a virtual lane; if it is a virtual lane, go to step 1332; if it is not a virtual lane, go to step 1334.

Virtual lanes refer to lanes that are planned when navigating a vehicle through a two-dimensional map and have no actual lane lines on the actual road.

As shown in Figure 7, the vehicle will pass through a T-shaped intersection when driving forward. There is no specific lane delineated in the T-shaped intersection, that is, there is no lane line. When the vehicle is navigated through a two-dimensional map, two virtual lanes will be generated at this T-intersection (shown as a dotted line in Figure 7). The vehicle can drive from south to north along one virtual lane (through virtual lane) into the north lane. Vehicles can drive from south to east along another virtual lane (right-turn virtual lane) into the east lane.

Step 1332: Obtain the virtual lane that matches the current vehicle's pose.

Referring to Figures 7 and 8, when the vehicle travels from the position shown in Figure 7 to the position shown in Figure 8, the vehicle position falls within the map range corresponding to the two virtual lanes at the same time. However, only the straight virtual lane is a virtual lane that matches the pose of the current vehicle.

Referring to Figures 7 and 9, when the vehicle travels from the position shown in Figure 7 to the position shown in Figure 9, the vehicle position falls within the map range corresponding to the two virtual lanes at the same time. However, only the right-turn virtual lane is a virtual lane that matches the pose of the current vehicle.

Step 1333: Determine whether the number of virtual lanes matching the current vehicle's pose is 1; if it is 1, then use the lane ID of this matching virtual lane as the lane ID of the lane where the vehicle is currently located; if not 1, then Initialization processing is performed based on the current vehicle position, the starting lane of the vehicle is re-determined, and the lane ID of the starting lane is used as the lane ID of the lane where the vehicle is currently located.

Step 1334: Obtain the lane that has a connected relationship with the lane corresponding to the last lane ID from the candidate lanes.

Step 1335: Determine whether the number of lanes that have a connected relationship with the lane corresponding to the last lane ID is 1; if it is 1, use the lane ID of this connected lane as the vehicle in the lane where the vehicle is currently located. Lane ID; if it is not 1, initialization processing is performed based on the current vehicle position, the starting lane of the vehicle is re-determined, and the lane ID of the starting lane is used as the lane ID of the lane where the vehicle is currently located.

Through the methods described in steps 1331 to 1335 above, when the current vehicle position falls within the map range corresponding to lanes within multiple preset ranges, it is possible to determine the location of the vehicle according to the connected relationship with the lane corresponding to the last lane ID. The lane and the current vehicle's pose can accurately determine the lane ID of the lane where the vehicle is currently located.

The above is the specific description of step S1032, and the description of step S1033 will continue below.

Step S1033: Determine the lane ID of the lane where the vehicle is currently located based on the map range of the lane located within the preset range centered on the current vehicle position on the two-dimensional map and the current vehicle position.

Without determining the last lane ID, the lane search range is expanded by acquiring lanes located within a preset range centered on the current vehicle position, and the number of roads surrounding the vehicle position is increased so that these roads can be Determine the lane the vehicle is currently in.

It should be noted that the method of "determining the lane ID of the lane where the vehicle is currently located based on the map range and the current vehicle position of the lane located within the preset range centered on the current vehicle position on the two-dimensional map" in step S1033 is the same as The method described in step 133 in step S1032 is the same and will not be described again.

Referring to Figure 10, in one embodiment according to the present invention, when the vehicle has just started the lane positioning function or otherwise needs to re-determine the starting lane of the vehicle (such as the situation described in steps 1333 and 1335), the following step S201 can be performed Go to step S206 to perform initialization processing and determine the starting lane of the vehicle.

Step S201: Obtain the lane located within the preset range centered on the current vehicle position.

Step S202: Determine whether the current vehicle position falls within a map range corresponding to a lane within a preset range; if it falls within a map range corresponding to a lane within a preset range, go to step S203; otherwise, go to step S203. S206.

Step S203: Obtain the distance between the current vehicle position and the boundary of the lane within the preset range.

Step S204: If the distance is greater than or equal to the preset distance threshold; if yes, go to step S205; if not, go to step S206.

Step S205: Use the lane within the preset range as the starting lane of the vehicle.

When locating a vehicle to determine its location through a two-dimensional map, the vehicle's location may fluctuate within a certain range, such as within a range of 5cm to 30cm. If the vehicle position is too close to the boundary of the lane, it may not be possible to accurately determine whether the vehicle is in the lane. In order to avoid the influence of the above fluctuations on the judgment, a distance threshold can be set. If the distance between the current vehicle position and the boundary of the lane is greater than or equal to this distance threshold, it will be judged to be driving in this lane. At this time, this lane can be As the starting lane for vehicles.

Step S206: Output lane positioning abnormality prompt information.

If the starting lane of the vehicle cannot be determined through the initialization process, manual intervention or other means are required to solve the problem. Therefore, in this case, the lane positioning abnormal prompt information can be output to remind the vehicle user or other personnel to handle the abnormal situation. deal with.

It should be pointed out that although the various steps are described in a specific sequence in the above embodiments, However, those skilled in the art can understand that in order to achieve the effects of the present invention, different steps do not have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders. These changes are within the scope of the present invention. within.

Those skilled in the art can understand that the present invention can implement all or part of the process in the method of the above-mentioned embodiment, and can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable file. In the storage medium, when the computer program is executed by the processor, the steps of each of the above method embodiments can be implemented. Wherein, the computer program includes computer program code, which may be in the form of source code, object code, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, media, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier wave signals, telecommunications signals, and software distribution media, etc. It should be noted that the content contained in the computer-readable storage medium can be appropriately added or deleted according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media Storage media does not include electrical carrier signals and telecommunications signals.

Furthermore, the present invention also provides a computer device. In one embodiment of the computer equipment according to the present invention, the computer equipment includes a processor and a storage device. The storage device may be configured to store a program for executing the lane positioning method of the above method embodiment. The processor may be configured to execute the storage A program in the device, which program includes but is not limited to a program for executing the lane positioning method of the above method embodiment. For ease of explanation, only the parts related to the embodiments of the present invention are shown. If specific technical details are not disclosed, please refer to the method part of the embodiments of the present invention. The computer device may be a device formed including various electronic devices.

Furthermore, the present invention also provides a computer-readable storage medium. In one embodiment of a computer-readable storage medium according to the present invention, the computer-readable storage medium may be configured to store a program for executing the lane positioning method of the above method embodiment, and the program may be loaded and run by a processor to implement the above lane positioning method. Positioning method. For ease of explanation, only the parts related to the embodiments of the present invention are shown. If specific technical details are not disclosed, please refer to the method part of the embodiments of the present invention. The computer-readable storage medium may be a storage device formed by various electronic devices. Optionally, in the embodiment of the present invention, the computer-readable storage medium is a non-transitory computer-readable storage medium.

Furthermore, the present invention also provides a vehicle. In a vehicle embodiment according to the present invention, the vehicle may include a computer device as described in the above computer device embodiment. In this embodiment, the vehicle may be an autonomous vehicle, an unmanned vehicle, or other vehicles. In addition, according to the type of power source, the vehicle in this embodiment may be a fuel vehicle, an electric vehicle, a hybrid vehicle that mixes electric energy with fuel, or a vehicle that uses other new energy sources.

So far, the technical solution of the present invention has been described with reference to the preferred embodiments shown in the drawings. However, those skilled in the art can easily understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to relevant technical features, and technical solutions after these modifications or substitutions will fall within the protection scope of the present invention.

Claims

A lane positioning method, characterized in that the method includes:

Locate the vehicle through a two-dimensional map and determine its location;

Obtain the lanes around the vehicle position and the connectivity between lanes;

The lane ID of the lane where the vehicle is currently located is determined based on the lanes surrounding the vehicle position and the connectivity relationship between the lanes.
The lane positioning method according to claim 1, wherein the step of "determining the lane ID of the lane where the vehicle is currently located based on the lanes around the vehicle position and the connectivity between the lanes" specifically includes:

Determine whether the last lane ID can be determined based on the last determined vehicle position;

If it can be determined, determine the lane ID of the lane where the vehicle is currently located based on the map range where the lane corresponding to the last lane ID falls on the two-dimensional map and the current vehicle position;

If it cannot be determined, the lane ID of the lane where the vehicle is currently located is determined based on the map range of the lane located within the preset range centered on the current vehicle position and falling on the two-dimensional map and the current vehicle position.
The lane positioning method according to claim 2, characterized in that "determine the lane ID of the lane where the vehicle is currently located based on the map range of the lane corresponding to the last lane ID falling on the two-dimensional map and the current vehicle position" The specific steps include:

Determine whether the current vehicle position falls within the map range corresponding to the last lane ID and the two-dimensional map only contains one lane within the map range;

If so, use the last lane ID as the lane ID of the lane where the vehicle is currently located;

If not, obtain other lanes that belong to the same road layer as the lane corresponding to the last lane ID and have a connected relationship with the lane corresponding to the last lane ID, and based on where the other lanes fall on the two-dimensional map The map range and the current vehicle position determine the lane ID of the lane where the vehicle is currently located.
The lane positioning method according to claim 3, wherein the step of "determining the lane ID of the lane where the vehicle is currently located based on the map range of the other lanes falling on the two-dimensional map and the current vehicle position" specifically includes:

Determine whether the current vehicle position only falls within the map range corresponding to one of the other lanes;

If so, use the lane ID of one of the other lanes as the lane ID of the lane where the vehicle is currently located;

If not, the lane ID of the lane where the vehicle is currently located is determined based on the map range of the lane located within the preset range centered on the current vehicle position falling on the two-dimensional map and the current vehicle position.
The lane positioning method according to any one of claims 2 to 4, characterized in that "according to the map range of the lane located within the preset range centered on the current vehicle position falling on the two-dimensional map and the current vehicle position , the steps to determine the lane ID of the lane where the vehicle is currently located specifically include:

If the current vehicle position only falls within the map range corresponding to one lane within the preset range, then the lane ID of the lane within the preset range is used as the lane ID of the lane where the vehicle is currently located;

If the current vehicle position falls within the map range corresponding to multiple lanes within the preset range, then the multiple lanes within the preset range are used as candidate lanes, based on the corresponding lane ID of the last time The lanes have connected lanes and/or the position and orientation of the current vehicle, and the lane ID of the lane where the vehicle is currently located is filtered out from the lane IDs of the candidate lanes.
The lane positioning method according to claim 5, characterized in that, "According to the lane that has a connected relationship with the lane corresponding to the last lane ID and/or the position and posture of the current vehicle, from the lane ID of the candidate lane The steps to filter out the lane ID of the lane where the vehicle is currently located include:

Determine whether the candidate lane is a virtual lane;

When it is determined that it is a virtual lane, obtain a virtual lane that matches the position and orientation of the current vehicle. If the number of matching virtual lanes is 1, use the lane ID of the matching virtual lane as the lane ID of the lane where the vehicle is currently located. ; If the number of matching virtual lanes is not 1, initialization processing is performed based on the current vehicle position, the starting lane of the vehicle is re-determined, and the lane ID of the starting lane is used as the lane ID of the lane where the vehicle is currently located;

When it is determined that it is not a virtual lane, obtain a lane that has a connectivity relationship with the lane corresponding to the last lane ID from the candidate lane;

If the number of lanes with a connected relationship is 1, then the lane ID of the lane with a connected relationship is used as the lane ID of the lane where the vehicle is currently located; if the number of lanes with a connected relationship is not 1, then the lane ID of the lane with a connected relationship is used as the lane ID of the lane where the vehicle is currently located. The vehicle position is initialized, the starting lane of the vehicle is redetermined, and the lane ID of the starting lane is used as the lane ID of the lane where the vehicle is currently located.
The lane positioning method according to claim 2, 3, 4 or 6, characterized in that the method also includes performing an initialization process according to the current vehicle position and through the following steps to determine the starting lane of the vehicle:

Get lanes located within a preset range centered on the current vehicle position;

Determine whether the current vehicle position falls within a map range corresponding to a lane within the preset range;

When it falls within the map range corresponding to only one lane within the preset range, obtain the distance between the current vehicle position and the boundary of the lane within the preset range; if the distance is greater than or equal to the preset range, If the distance threshold is set, the lane within the preset range will be used as the starting lane of the vehicle; if the distance is less than the preset distance threshold, lane positioning abnormality prompt information will be output;

When it does not fall within the map range corresponding to only one lane within the preset range, lane positioning abnormality prompt information is output.
The lane positioning method according to claim 1, characterized in that the step of "locating the vehicle through a two-dimensional map and determining the vehicle position" specifically includes:

Locate the vehicle's current driving road through a two-dimensional map and determine the two-dimensional coordinates of the road points on the current driving road in the two-dimensional Cartesian coordinate system;

Obtain the relative height of the road point relative to the vehicle, determine the origin, x-axis, and y-axis of the world coordinate system based on the origin, x-axis, and y-axis of the two-dimensional rectangular coordinate system, and use the relative height as the road point in the world The coordinates of the z-axis in the coordinate system, back-projecting the road point from the world coordinate system to the image coordinate system according to the two-dimensional coordinates of the road point and the relative height;

Determine the road elements located around the vehicle, obtain the back-projection position of the road point at the location of the road element to the image coordinate system, and obtain the road at the location of the road element determined by the image acquisition device of the vehicle The acquisition position of the point in the image coordinate system;

Position matching is performed on the back-projected position of the road point at the location of the road element and the collection position, and the position of the vehicle in the world coordinate system is determined based on the position matching result.
The lane positioning method according to claim 8, wherein the step of "obtaining the relative height of the road point relative to the vehicle" specifically includes:

Obtain the plane vector of the plane where the vehicle is located in the vehicle body coordinate system and obtain the unit normal vector of the plane according to the plane vector;

According to the two-dimensional coordinates of the road point and by solving the following equations, it is obtained that the road point is in the world coordinate system with the origin, x-axis and y-axis of the two-dimensional rectangular coordinate system as the origin, x-axis and y-axis respectively. The coordinate of the z-axis zh:

(xh-x1)×nx+(yh-y1)×ny+(zh-z1)×nz=0

Among them, x1, y1 and z1 represent the coordinates of the vehicle on the x-axis, y-axis and z-axis in the vehicle body coordinate system, nx, ny and nz represent the coordinates of the unit normal vector on the x-axis, y-axis and z-axis in the vehicle body coordinate system. The coordinates of the axes, xh and yh represent the coordinates of the x-axis and y-axis in the two-dimensional coordinates of the road point;

The relative height of the road point relative to the vehicle is determined based on the coordinate zh.
The lane positioning method according to claim 8, characterized in that the method further includes displaying road connections on the two-dimensional map of the road currently traveled by the vehicle and the roads it has traveled before.
The lane positioning method according to claim 8, characterized in that "the back-projected position of the road point at the position of the road element is position matched with the collection position, and the position of the vehicle in the world coordinate system is determined according to the position matching result. The steps for "location" specifically include:

Perform position matching between the back-projection position of the road point at the location of the road element and the acquisition position to obtain pose parameters for conversion between the device coordinate system of the image acquisition device and the world coordinate system;

Obtain the back-projection position of the road point at the location of the vehicle to the image coordinate system;

The position of the vehicle in the world coordinate system is determined based on the back-projection position of the road point at the location of the vehicle and the pose parameter.
The lane positioning method according to claim 11, characterized in that "the back-projection position of the road point at the position of the road element is matched with the collection position to obtain the device coordinate system of the image collection device and the world The steps of "Pose and Orientation Parameters for Coordinate System Transformation" specifically include:

According to the back-projection position and collection position of the road point at the location of the road element, a distance error equation shown in the following formula is established:

loss=d(f(A,O),cd)

Among them, f(A,O) represents the back-projection position of the current road point, A represents the position of the current road point in the world coordinate system, O represents the pose parameter for conversion between the device coordinate system and the world coordinate system, and cd represents the distance from the current road point. The collection positions of the two closest road points to the back-projection position of the road point are the line segments formed by the endpoints. d represents the distance calculation function from the back-projection position f (A, O) to the line segment cd. Loss represents the distance calculation function d calculation. the distance obtained;

With the goal of making the distance loss smaller than the preset distance threshold, the Levenberg-Marquardt algorithm is used to iteratively optimize the pose parameter O in the distance error equation, and obtain the The pose parameter O when the distance loss is less than the preset distance threshold is used as the final pose parameter for conversion between the device coordinate system of the image acquisition device and the world coordinate system.
The lane positioning method according to claim 8, characterized in that in "determining the road elements located around the vehicle, obtaining the back-projected position of the road point at the location of the road element to the image coordinate system and obtaining the back-projected position of the vehicle passing by Before the step of "the image acquisition device determines that the road point at the position of the road element is at the acquisition position of the image coordinate system," the method further includes:

Use at least two image acquisition devices to collect images of the road points at the locations of the road elements to obtain road point images of the road points at the locations of the road elements;

According to the road point image, the collection position of the road point at the location of the road element in the image coordinate system is determined.
A computer device, comprising a processor and a storage device, the storage device being adapted to store a plurality of program codes, characterized in that the program codes are adapted to be loaded and run by the processor to execute claims 1 to 13 The lane positioning method described in any one of the above.
A computer-readable storage medium in which a plurality of program codes are stored, characterized in that the program codes are adapted to be loaded and run by the processor to perform the lane positioning described in any one of claims 1 to 13 method.
A vehicle, characterized in that the vehicle includes the computer device of claim 14.