CN118670378B

CN118670378B - Electronic map data verification method, electronic map construction method and device

Info

Publication number: CN118670378B
Application number: CN202411140090.4A
Authority: CN
Inventors: 雷祎东; 刘彦辉; 范旭; 尹占涛
Original assignee: Hangzhou Langge Technology Co ltd; Zhejiang Geely Holding Group Co Ltd
Current assignee: Hangzhou Langge Technology Co ltd; Zhejiang Geely Holding Group Co Ltd
Priority date: 2024-08-20
Filing date: 2024-08-20
Publication date: 2025-02-07
Anticipated expiration: 2044-08-20
Also published as: CN118670378A

Abstract

The present application provides a method for verifying electronic map data, a method for constructing electronic maps, and a device. In the method for verifying electronic map data, at least two non-point cloud data collected by a collection vehicle for a preset area are obtained, and then the at least two non-point cloud data are converted into point clouds to obtain first point cloud data corresponding to the at least two non-point cloud data. Finally, the verification result of the at least two non-point cloud data is determined based on the difference between the first point cloud data and the second point cloud data collected by a radar unit for the preset area. In this technical solution, by determining various data used for making maps in the preset area collected by the collection vehicle as point cloud data, and comparing them with the point cloud data obtained by the radar that actually fits the preset area, it is determined whether the data used for making maps is accurate. No human intervention is required in the process, thereby saving waste of human resources and improving the accuracy of data verification.

Description

Verification method of electronic map data, electronic map construction method and device

Technical Field

The present application relates to the field of electronic map technologies, and in particular, to a method for verifying electronic map data, and a method and apparatus for constructing an electronic map.

Background

In the development process of the automatic driving technology, the electronic map is a very core infrastructure, can provide road information with the most effectiveness and accuracy, and has decisive influence on the safety and the high efficiency of the automatic driving in terms of environment perception, positioning, path planning and the like, so that the accuracy verification of related data forming the electronic map is also very important.

In the prior art, the related data are generally directly generated into an electronic map, then a global positioning system (Global Positioning System, GPS) measuring instrument is used for carrying out coordinate dotting on the actual road position in a field test mode, and the measured coordinate point is compared with the electronic map in space precision to determine the accuracy of the related data.

However, the above method needs to generate the electronic map, which is complicated to implement, and has the problems of waste of human resources and insufficient verification accuracy due to human participation.

Disclosure of Invention

The application provides a verification method of electronic map data, an electronic map construction method and a device, which are used for solving the problems of resource waste, low verification accuracy, complex realization and the like in the verification of the electronic map data in the prior art.

In a first aspect, an embodiment of the present application provides a method for verifying electronic map data, including:

Acquiring at least two non-point cloud data acquired by an acquisition vehicle aiming at a preset area;

Performing point cloud conversion processing on the at least two non-point cloud data to obtain first point cloud data corresponding to the at least two non-point cloud data;

Determining verification results of the at least two non-point cloud data according to differences between the first point cloud data and second point cloud data acquired from the preset area through the radar unit, wherein the verification results indicate data quality when the at least two non-point cloud data are used for constructing an electronic map.

In one or more embodiments, the non-point cloud data includes image information acquired by a camera unit in the acquisition vehicle, positioning information acquired by a position measurement unit in the acquisition vehicle, and vehicle speed information acquired by an inertial measurement unit in the acquisition vehicle.

In one or more embodiments, the performing a point cloud conversion process on the at least two non-point cloud data to obtain first point cloud data corresponding to the at least two non-point cloud data includes:

determining the running track information of the collected vehicle according to the positioning information and the vehicle speed information;

acquiring three-dimensional point cloud information around the acquired vehicle according to the image information;

and converting the three-dimensional point cloud information into a global coordinate system according to the moving track information and the visual angle information of the camera shooting unit to obtain the first point cloud data.

In one or more embodiments, the determining the verification result of the at least two non-point cloud data according to the difference between the first point cloud data and the second point cloud data acquired by the radar unit for the preset area includes:

Determining third point cloud data corresponding to the environment to be verified from the first point cloud data;

And determining verification results of the at least two non-point cloud data according to differences of fourth point cloud data corresponding to the environment to be verified in the third point cloud data and the second point cloud data.

In one or more embodiments, the determining, according to the difference between the third point cloud data and the fourth point cloud data corresponding to the environment to be verified in the second point cloud data, the verification result of the at least two non-point cloud data includes:

Determining first truth value data corresponding to the third point cloud data and second truth value data corresponding to the fourth point cloud data;

and determining verification results of the at least two non-point cloud data according to the difference between the first truth data and the second truth data.

In one or more embodiments, the determining the verification result of the at least two non-point cloud data according to the difference between the first truth data and the second truth data includes:

Determining a first verification result according to the space position difference value between the geometric vector in the first truth value data and the geometric vector in the second truth value data and a first preset threshold value;

and/or the number of the groups of groups,

Determining a second verification result according to the attribute difference value between the elements in the first truth data and the elements in the second truth data and a second preset threshold value;

The verification results comprise the first verification result and/or the second verification result.

In one or more embodiments, the method further comprises:

And dividing the verification result into sub-verification results with different deviation dimensions according to the space position difference value and the attribute difference value.

In one or more embodiments, the method further comprises:

and adjusting parameters of the shooting unit, the position measuring unit and the inertia measuring unit according to the verification results of the at least two non-point cloud data.

In one or more embodiments, the environment to be verified includes at least one of an area to be verified, a scene to be verified, an element to be verified, and a sample magnitude to be verified in the preset area.

In a second aspect, an embodiment of the present application provides an electronic map construction method, including:

Determining a verification result of collecting at least two non-point cloud data for a preset area based on the verification method of the electronic map data of any one of the first aspect and any one of the possible designs;

if the data quality indicated by the verification result meets the map data quality requirement, constructing an electronic map corresponding to the preset area through the at least two non-point cloud data.

In a third aspect, an embodiment of the present application provides an apparatus for verifying electronic map data, including:

the acquisition module is used for acquiring at least two non-point cloud data acquired by the acquisition vehicle aiming at a preset area;

The processing module is used for carrying out point cloud conversion processing on the at least two non-point cloud data to obtain first point cloud data corresponding to the at least two non-point cloud data;

The determining module is used for determining verification results of the at least two non-point cloud data according to the difference between the first point cloud data and the second point cloud data acquired by the radar unit for the preset area, wherein the verification results indicate data quality when the at least two non-point cloud data are used for constructing an electronic map.

In one or more embodiments, the processing module is specifically configured to:

In one or more embodiments, the determining module determines a verification result of the at least two non-point cloud data according to a difference between the first point cloud data and the second point cloud data acquired by the radar unit for the preset area, and is specifically configured to:

In one or more embodiments, the determining module determines the verification result of the at least two non-point cloud data according to the difference between the third point cloud data and fourth point cloud data corresponding to the environment to be verified in the second point cloud data, where the verification result is specifically configured to:

In one or more embodiments, the determining module determines a verification result of the at least two non-point cloud data according to a difference between the first truth data and the second truth data, specifically for:

and/or the number of the groups of groups,

In one or more embodiments, the determining module is further configured to:

In a fourth aspect, an embodiment of the present application provides an electronic map construction apparatus, including:

a determining module, configured to determine a verification result of collecting at least two non-point cloud data for a preset area based on the verification method of electronic map data according to any one of the first aspect and any one of the possible designs;

and the construction module is used for constructing the electronic map corresponding to the preset area through the at least two non-point cloud data when the data quality indicated by the verification result meets the map data quality requirement.

In a fifth aspect, an embodiment of the present application provides an electronic device, including a processor, and a memory communicatively coupled to the processor;

The memory stores computer-executable instructions;

the processor executes computer-executable instructions stored by the memory to implement the method as described in the first, second or any of the above aspects.

In a sixth aspect, embodiments of the present application provide a computer-readable storage medium having stored therein computer-executable instructions that, when executed by a processor, are configured to implement the method according to the first, second or any one of the above-described aspects

In a seventh aspect, embodiments of the present application provide a computer program, where the computer program product includes a computer program stored in a computer readable storage medium, from which at least one processor can read the computer program, where the at least one processor can implement the method according to the first, second or any one of the above-mentioned aspects when executing the computer program.

According to the verification method, the verification result of the at least two non-point cloud data is determined according to the difference between the first point cloud data and the second point cloud data acquired by the radar unit for the preset area, and the verification result indicates the data quality of the at least two non-point cloud data when the at least two non-point cloud data are used for constructing the electronic map. According to the technical scheme, the data quality of related data for making the map is determined by converting various data for making the map in the preset area acquired on the acquisition vehicle into point cloud data and comparing the point cloud data with high data quality acquired by the radar unit aiming at the preset area, the method does not need to be manually participated, so that the waste of manpower resources is saved, the verification efficiency of the related data for electronic map construction is improved, and on the other hand, the method converts a plurality of non-point cloud data acquired aiming at the preset area into the point cloud data, verifies the difference of the point cloud data with high accuracy acquired by the radar unit with the data quality of the plurality of non-point cloud data for constructing the electronic map, and further improves the accuracy of data verification for electronic map construction.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.

FIG. 1 is a schematic diagram of an architecture for electronic map production;

Fig. 2 is a flowchart illustrating a method for verifying electronic map data according to an embodiment of the present application;

fig. 3 is a second flowchart of a verification method of electronic map data according to an embodiment of the present application;

fig. 4 is a flowchart illustrating a third method for verifying electronic map data according to an embodiment of the present application;

Fig. 5 is a flowchart illustrating a method for verifying electronic map data according to an embodiment of the present application;

FIG. 6 is a schematic diagram of verification provided by an embodiment of the present application;

fig. 7 is a schematic flow chart of an electronic map construction method according to an embodiment of the present application;

Fig. 8 is a schematic structural diagram of a verification device for electronic map data according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of an electronic map building device according to an embodiment of the present application;

Fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Specific embodiments of the present disclosure have been shown by way of the above drawings and will be described in more detail below. These drawings and the written description are not intended to limit the scope of the disclosed concepts in any way, but rather to illustrate the disclosed concepts to those skilled in the art by reference to specific embodiments.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

Before describing the embodiments of the present application, an application background of the embodiments of the present application will be explained first:

In the development process of the automatic driving technology, the electronic map is a very core infrastructure, can provide road information with the highest effectiveness and accuracy, and has decisive influence on the safety and the high efficiency of the automatic driving from the aspects of environment perception, positioning, path planning and the like.

Fig. 1 is a schematic diagram of an architecture for electronic map production, as shown in fig. 1, including field collection, data processing, in-industry production, quality check, and data output.

In one possible implementation, the field collection stage is generally implemented by collecting high-precision equipment such as a Camera (English: camera), a radar (English: lidar), a high-precision global positioning system (Global Positioning System, GPS), an inertial measurement unit (Inertial Measurement Unit, IMU), a central control unit and the like on a vehicle, collecting road tracks, images, space positions and three-dimensional point clouds in one station to obtain basic data for generating subsequent electronic map production, preprocessing the collected basic data to remove stain data and the like, performing an in-industry production stage to generate an initial electronic map on the data after the data processing, performing verification on the initial electronic map by using a mode of real vehicle testing or field testing, entering a data output stage after verification, namely, performing online on the electronic map after verification, and the like.

In the implementation process, due to the fact that the Lidar acquisition equipment is high in cost and large in data processing and storage magnitude, in the mass production scheme of the electronic map, 80% -90% of the electronic map acquisition vehicles do not start the Lidar equipment in the field acquisition process, and laser point cloud acquisition is carried out by starting the Lidar equipment only on a 10% -20% road section, so that the electronic map acquisition vehicles are used for precision positioning matching and production.

Based on the above situation, according to the basic data collected by the units such as the camera, the GPS, the IMU and the like, the accuracy of the basic data has a certain inaccuracy, and thus the generation of the electronic map is affected, and the safety and the high efficiency of the automatic driving are affected, so how to verify the data quality of the basic data becomes a technical problem to be solved urgently.

In the implementation of the above related art, the following verification method for generating basic data related to an electronic map is provided:

firstly, carrying an electronic map generated based on basic data on a vehicle end for real-time drive test, and checking the accuracy and the precision of the map in the drive test process, namely determining the accuracy of the basic data;

Secondly, in-situ measurement, coordinate dotting is carried out on the actual road position by using a GPS measuring instrument, and the spatial precision of the measured coordinate point is compared with that of an electronic map generated based on basic data, so that the accuracy of the basic data is determined.

However, in the prior art, an electronic map needs to be generated based on the acquired data, which is complicated for data verification, and then the electronic map is used for field measurement, such as a coordinate dotting mode, so that the method not only needs participation of manpower resources, but also involves the defects of verification of a large amount of data, long time consumption, limited sample magnitude and incapability of meeting large-magnitude sample test.

Aiming at the technical problems in the prior art, the inventor of the application has the following conception that before the electronic map is generated, the obtained basic data can be verified, further, the generation of the electronic map can be directly carried out based on the basic data with qualified data quality of a verification result after the verification is passed, the complex realization in the prior art is avoided, under most scenes, the basic data are collected based on high-precision equipment such as a camera, a GPS (global positioning system), an IMU (inertial measurement unit) and the like, and then, the manufactured map has certain precision, at the moment, the data measured by the measuring units such as the camera, the GPS, the IMU and the like are reversely pushed to obtain point cloud data, and the point cloud data acquired by the radar in the area are compared, so that the precision of the data for generating the map can be verified, the participation of manpower is not needed, the high cost is not needed, and the verification accuracy is improved.

Further, in the subsequent implementation, if the verification result of the data quality is higher, for example, the verification result meets the map data quality requirement, an accurate and more practical electronic map can be generated based on the basic data.

In one embodiment, the above manner may be further utilized to extract basic data corresponding to some areas in the high-precision map for verification after the electronic map is generated, so as to determine the accuracy of the electronic map.

It should be understood that the electronic map in the embodiment of the present application may be any one of a high-precision map, a standard-precision map, and other precision maps, and in the embodiment of the present application, the high-precision map is used as an example of the electronic map.

Therefore, the embodiment of the application provides a verification method for electronic map data, and one possible application scenario of the method can be that after high-precision equipment such as a camera, a GPS (global positioning system) and an IMU (inertial measurement unit) based on a vehicle collects data of a whole section of road, the data of a part of the road can be intercepted for precision verification, and after a verification result of the data of the part of the road is accurate, an electronic map can be generated based on the data of the whole section of road.

The technical scheme of the application is described in detail through specific embodiments. It should be noted that the following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

The execution subject of the application is an electronic device, and may specifically be a vehicle, a server, a computing device, or the like.

Fig. 2 is a flowchart of a method for verifying electronic map data according to an embodiment of the present application, as shown in fig. 2, the method may include the following steps:

And 21, acquiring at least two non-point cloud data acquired by the acquisition vehicle aiming at a preset area.

In this step, in order to generate a more accurate electronic map that is more fit to the actual electronic map, verification needs to be performed on data for generating the electronic map, where the data of the electronic map may be at least two non-point cloud data acquired based on the acquisition vehicle.

Alternatively, the non-point cloud data may be based on data collected by a non-radar unit in the collection vehicle for subsequent drawing of an electronic map, such as the basic data obtained by the collection vehicle collecting a preset area as mentioned above.

In order to improve verification efficiency, at least two non-point cloud data in a preset area can be acquired, and verification is performed on the at least two non-point cloud data in the area.

Optionally, the preset area may be any designated area in the actual scene corresponding to the electronic map, may be an area within a certain range along the preset road, may also be an area measured by using radar on some special road sections, and may also be an area covered by the whole map in some cases.

Optionally, the non-point cloud data includes image information collected by a camera unit in the collected vehicle, positioning information collected by a position measurement unit in the collected vehicle, and vehicle speed information collected by an inertia measurement unit in the collected vehicle.

In one possible implementation, the camera unit for acquiring the image information may be a camera or a camera installed on the acquisition vehicle, the position measuring unit for acquiring the positioning information may be a GPS installed on the vehicle, and the inertial measuring unit for acquiring the vehicle speed information may be at least one or more of an accelerometer, a gyroscope and a magnetometer.

In some possible implementations, the image information is an image captured during actual collection, the positioning information is longitude and latitude and the like of a collected vehicle during running, and the vehicle speed information is corresponding speed, direction and the like of the collected vehicle during running.

In addition, the positioning information may also include the altitude at which the acquisition vehicle is located, which may be an altitude or a ground level, and in some implementations may be used to create a three-dimensional model or scene of a geographic entity, such as a building, mountain, river, etc. By combining longitude and latitude with altitude, the position and shape of these geographic elements in the vertical direction can be accurately located and presented.

In a subsequent possible implementation, the "spatial position" of the point cloud mentioned below may be latitude and longitude information of the point cloud, and in some implementations, may also include the altitude of the point cloud on the latitude and longitude, that is, may be three-dimensional coordinates.

And 22, performing point cloud conversion processing on at least two non-point cloud data to obtain first point cloud data corresponding to the at least two non-point cloud data.

In this scheme, in order to verify the electronic map data, more accurate point cloud data obtained by a radar unit is adopted for comparison, so in this step, at least two non-point cloud data need to be subjected to point cloud conversion processing to obtain point cloud data corresponding to the at least two non-point cloud data, and the point cloud data is recorded as first point cloud data, where the radar unit may be radar equipment (such as a laser radar or a radar range finder) installed on an acquisition vehicle.

Step 23, determining verification results of the at least two non-point cloud data according to differences between the first point cloud data and second point cloud data acquired by the radar unit for a preset area.

The verification result indicates the data quality when at least two non-point cloud data are used for constructing the electronic map.

In this step, since the second point cloud data corresponding to the preset area acquired by the radar unit is relatively accurate, but the cost is relatively high, the point cloud data determined by the imaging unit, the position measurement unit and the inertia measurement unit can be compared in actual implementation, so as to determine the accuracy and reliability of the non-point cloud data acquired by the imaging unit, the position measurement unit and the inertia measurement unit in actual operation.

Optionally, the verification result may be two types of data quality qualification and data quality disqualification, or may be divided according to the degree, attribute, etc. of the data difference based on the actual situation.

In one possible implementation, taking the degree of the data difference as an example, in the same position, the distance between the point in the converted point cloud and the point cloud midpoint acquired by the radar unit can be divided by a preset difference threshold value to obtain a corresponding difference range, taking the attribute as the intensity information of the point cloud midpoint as an example, in the same position, the intensity information of the point in the converted point cloud and the intensity information of the point in the position in the point cloud acquired by the radar unit are compared, and different intensity ranges and corresponding materials can be set to detect the material classification corresponding to different points.

Further, after step 23, parameters of the three of the image capturing unit, the position measuring unit, and the inertial measuring unit may be adjusted according to the verification results of the at least two non-point cloud data.

Under the realization, the verification results corresponding to the at least two non-point cloud data can reflect the precision of each acquisition unit in the acquisition vehicle in the actual map acquisition process to a certain extent, so that the parameters of the shooting unit, the position measurement unit and the inertia measurement unit can be adjusted based on the verification results so as to improve the precision of the subsequent acquisition operation.

For example, compared with the difference in spatial positions (such as three-dimensional coordinates), directions, angles and the like of two points in the same area, the data of the units are determined to be more accurate or consistent under specific conditions through statistical analysis or visual inspection, and when parameters are adjusted, the change of the parameters and the influence of the change on the point cloud data are carefully recorded, so that the adjustment of the parameters is realized.

For the imaging unit, internal parameters (such as focal length and distortion coefficient) and external parameters (such as imaging unit position and posture) of the imaging unit can be adjusted to improve the spatial accuracy of point cloud data, for the position measurement unit, precision parameters of the imaging unit or the precision of position measurement can be improved by using differential GPS, and for the inertial measurement unit, zero offset and scale factors of a gyroscope and an accelerometer are adjusted to reduce drift and errors in inertial sensor data.

In one possible implementation, in the same area, the spatial position deviation between the point in the converted point cloud and the point in the point cloud acquired by the radar unit is larger, for the imaging unit, for example, if the deviation indicates that the point is farther away from the imaging unit than the radar unit, the value of the translation vector is increased to push the scene toward the camera, for the GPS, for example, if the deviation indicates that the positioning information is larger in phase difference, the updating frequency of the GPS in the precision parameters of the GPS can be increased, and for the gyroscope and the accelerometer, for example, if the deviation indicates that the speed information of the vehicle is actually lower than the speed information of the vehicle, the scaling factor of the gyroscope can be increased.

According to the verification method for the electronic map data, at least two non-point cloud data acquired by the acquisition vehicle aiming at the preset area are acquired, then point cloud conversion processing is carried out on the at least two non-point cloud data to obtain first point cloud data corresponding to the at least two non-point cloud data, and finally verification results of the at least two non-point cloud data are determined according to differences between the first point cloud data and second point cloud data acquired by the radar unit aiming at the preset area, wherein the verification results indicate data quality when the at least two non-point cloud data are used for constructing the electronic map. According to the technical scheme, the data quality of related data for making the map is determined by converting various data for making the map in the preset area acquired on the acquisition vehicle into point cloud data and comparing the point cloud data with high data quality acquired by the radar unit aiming at the preset area, the method does not need to be manually participated, so that the waste of manpower resources is saved, the verification efficiency of the related data for electronic map construction is improved, and on the other hand, the method converts a plurality of non-point cloud data acquired aiming at the preset area into the point cloud data, verifies the difference of the point cloud data with high accuracy acquired by the radar unit with the data quality of the plurality of non-point cloud data for constructing the electronic map, and further improves the accuracy of data verification for electronic map construction.

On the basis of the foregoing embodiment, fig. 3 is a second flowchart of a verification method of electronic map data according to the embodiment of the present application, as shown in fig. 3, the foregoing step 22 may be implemented as follows:

step 31, determining and collecting the running track information of the vehicle according to the positioning information and the vehicle speed information;

in this step, the positioning information is combined with the vehicle speed information, and the running track of the collected vehicle, including the position, direction and speed, can be more accurately determined.

One possible implementation is to perform time-stamp synchronization from the positioning information and the vehicle speed information, ensure that the positioning information and the vehicle speed information are available at the same time point, and then perform gesture calculation on the positioning information and the vehicle speed information after time-stamp synchronization by using a fusion algorithm (such as an Extended kalman filter (KALMAN FILTER, EKF)), so as to calculate the position, the speed and the direction of the acquired vehicle at each time point during the whole running period, thereby constructing a running track during the running period.

The running track can be a path for collecting the position change of the vehicle in a coordinate system along with time, and can reflect the position, the speed and the direction of the vehicle.

Step 32, acquiring three-dimensional point cloud information around the acquired vehicle according to the image information;

in this step, the image information includes depth information of each object in the actual acquisition scene, and three-dimensional point cloud information around the acquisition vehicle can be determined based on the depth information.

In one possible implementation, taking a stereo camera as an example, depth information of objects in a three-dimensional scene can be calculated using parallax information between cameras. Such as by a stereo matching algorithm, to obtain a depth estimate for each pixel, i.e., three-dimensional point cloud information.

In another possible implementation, a three-dimensional map of the surrounding environment of the acquisition vehicle may be created in real-time in combination with simultaneous localization and mapping (Simultaneous Localization AND MAPPING, SLAM) techniques. SLAM can estimate not only the position and attitude of a vehicle, but also create and update three-dimensional models of the environment and represent them as three-dimensional point cloud information.

In yet another possible implementation, depth information of a scene may be inferred from a single or multiple images using a depth learning model, such as a semantic segmentation model or a depth estimation model based on convolutional neural networks, which may be trained to directly predict the depth or point cloud data of each pixel from images acquired from a camera, i.e., to obtain three-dimensional point cloud information.

And step 33, converting the three-dimensional point cloud information into a global coordinate system according to the moving track information and the visual angle information of the camera unit to obtain first point cloud data.

The visual angle information provided by the image capturing unit refers to the position and the direction (gesture) of the camera, and can be obtained by calibrating the image capturing unit.

In this step, three-dimensional point cloud information is typically acquired under the coordinate system of the camera unit (e.g., the camera coordinate system or the local acquisition vehicle coordinate system), and the position of each point is relative to the camera or the acquisition vehicle coordinate system.

Therefore, the three-dimensional point cloud information is converted into the global coordinate system by combining the running track information, so that the first point cloud data which can be used for accuracy verification can be obtained.

In one possible implementation, first, the position and posture of each frame of camera are interpolated based on the motion trajectory information, then the pose (rotation matrix and translation vector) of each frame of camera coordinate system with respect to the global coordinate system is calculated according to the internal parameters and external parameters of the camera, and each three-dimensional point is transformed from the camera coordinate system (local coordinate system) into the global coordinate system.

According to the verification method for the electronic map data, the running track information of the collected vehicle is determined according to the positioning information and the vehicle speed information, then the three-dimensional point cloud information around the collected vehicle is obtained according to the image information, and finally the three-dimensional point cloud information is converted into a global coordinate system according to the running track information and the visual angle information of the camera unit, so that first point cloud data are obtained. According to the technical scheme, the point cloud information corresponding to the image information is laid out under the same coordinate system through the positioning information and the vehicle speed information, so that the point cloud information can be more conveniently and accurately compared with the point cloud data acquired by the follow-up radar.

On the basis of the foregoing embodiment, fig. 4 is a flowchart third of a verification method of electronic map data according to the embodiment of the present application, as shown in fig. 4, the foregoing step 23 may be implemented as follows:

step 41, determining third point cloud data corresponding to the environment to be verified from the first point cloud data;

in this step, in order to reduce the amount of processing data required for precision verification to improve verification efficiency, or to verify only part of data under some demands, some verification scenes may be set to limit, only comparing point cloud data in an environment to be verified.

The method comprises the steps of carrying out region cutting on first point cloud data based on features related to an environment to be verified, and obtaining third point cloud data corresponding to the environment to be verified.

Optionally, the environment to be verified comprises at least one of a region to be verified in a preset region, a scene to be verified, an element to be verified and a sample magnitude to be verified.

For example, the to-be-verified area in the preset area may be an element within a preset distance range of the point a in the preset area, the preset distance may be set according to actual requirements, for example, the preset distance may be, but not limited to, 1 meter, 1.5 meters or 2 meters, the to-be-verified scene may be one or more elements corresponding to the scenes of a bridge, a ramp, a tunnel, a main road, an auxiliary road, etc., the to-be-verified element may be an element in the to-be-verified scene, or may be one or more elements in the preset area of a tree, a marking, etc., the number of to-be-verified samples may be at least a certain preset value, that is, the element scale, etc.

Step 42, determining verification results of at least two non-point cloud data according to differences of fourth point cloud data corresponding to the environment to be verified in the third point cloud data and the second point cloud data.

In this step, in order to ensure the rationality and comparability of verification, the point cloud data corresponding to the environment to be verified in the second point cloud data acquired by the radar needs to be acquired and recorded as fourth point cloud data, and then the verification results corresponding to at least two non-point cloud data are determined based on the comparison between the third point cloud data and the fourth point cloud data.

According to the verification method of the electronic map data, the third point cloud data corresponding to the environment to be verified is determined from the first point cloud data, and the verification results of the at least two non-point cloud data are determined according to the difference of the fourth point cloud data corresponding to the environment to be verified in the third point cloud data and the second point cloud data. In the technical scheme, the specific environment to be verified is selected from the preset area, so that verification accuracy is improved, verification data size is reduced, verification efficiency is improved, and diversity of verification results can be improved in different verification environments.

On the basis of the foregoing embodiment, fig. 5 is a flow chart of a method for verifying electronic map data according to an embodiment of the present application, and as shown in fig. 5, the foregoing step 42 may be implemented as follows:

step 51, determining first truth value data corresponding to third point cloud data and second truth value data corresponding to fourth point cloud data;

in this step, the truth data may be generated for the corresponding point cloud data based on the production specification requirement, so as to obtain the first truth data corresponding to the third point cloud data and the second truth data corresponding to the fourth point cloud data.

Optionally, the first truth data is used for describing the attribute and/or the geometric vector of the element in the third point cloud data, and the second truth data is used for describing the attribute and/or the geometric vector of the element in the fourth point cloud data.

In one possible implementation, for any one of the third point cloud data and the fourth point cloud data, the attribute or the geometric vector of the element of interest may be extracted from the point cloud data using a computer vision and geometric processing manner, which may involve the following:

1, attribute of element:

Color attributes, intensity attributes, which are common attributes in laser radar point clouds and represent the reflection intensity of each point, can reflect the change of the surface reflection property, classification attributes, which are attributes for marking each point to belong to different categories or objects, such as ground points, building points, tree points and the like, timestamp attributes, which can identify the acquisition time of each point if the point cloud data is acquired from a dynamic scene, and other custom attributes, which can also define and extract various custom attributes, such as smoothness, surface normal direction, density and the like, according to specific application scenes.

2, Geometric vector:

The position vector describes the position of each point in three-dimensional space, usually represented by (x, y, z) coordinates, the normal vector describes the normal direction of each point on the surface, which is used for representing the direction and curvature of the surface, is important geometrical information in many point cloud analysis tasks, the curvature reflects the bending degree or curvature of the surface at a certain point, is usually calculated through the change rate of the normal, the bounding box or envelope vector describes the minimum bounding box or minimum bounding sphere of the point cloud or a subset thereof, is used for rapidly calculating the appearance range of the point cloud, other local or global descriptors are used for representing the local or global structure in the point cloud, if a plurality of elements are contained in the point cloud data, different elements can be identified and distinguished by using segmentation and classification algorithms, and then the element attribute or geometrical vector is extracted for each element.

Step 52, determining verification results of at least two non-point cloud data according to the difference between the first truth data and the second truth data.

In this step, the difference between the first truth value data corresponding to the third point cloud data and the second truth value data corresponding to the fourth point cloud data is compared, so that information related to the difference is obtained, and the information is used as verification results corresponding to at least two non-point cloud data, wherein the verification results can also represent verification results for generating the electronic map corresponding to the preset area.

Optionally, the step may be implemented as follows (the verification results corresponding to the at least two non-point cloud data include a first verification result and/or a second verification result):

Firstly, determining a first verification result according to a space position difference value between a geometric vector in the first truth value data and a geometric vector in the second truth value data and a first preset threshold value;

Wherein, the spatial position difference value can be the difference value of the geometric vector in the three-dimensional coordinates;

in this implementation, the geometry vector is given by way of example (only illustrative) of the following:

The difference of the position vectors is that the position pose of the midpoint of the first truth value data and the position pose of the midpoint of the second truth value data in the same geographic position can be respectively represented by the position vectors in one three-dimensional coordinate, and the difference of the two position vectors, namely the distance difference, can be calculated;

And 2, normal vector difference, wherein the normal vector describes the direction of the surfaces of two adjacent points, namely the normal vector of the adjacent points corresponding to the same geographic position in the first truth value data and the second truth value data, and the difference value of the two normal vectors is used as the difference value of the normal vectors.

For example, the spatial position difference is compared with a threshold (i.e. a first preset threshold) which can be accommodated, and at any position, first and second truth data corresponding to the spatial position difference being greater than the first preset threshold, and first and second truth data corresponding to the spatial position difference not being greater than the first preset threshold are determined as a first verification result, i.e. the first verification result includes first and second truth data based on whether the spatial position is good or bad.

In one implementation, element geometries, coordinates, element representations, element magnitudes that are greater than a first preset threshold in the first verification result may be output.

For example, taking the example that the spatial position difference value includes a distance difference value, for the same geographic position, the coordinates of a point a in the first truth value data are (2, 2), the coordinates of a point B in the second truth value data are (2, 3), a first preset threshold value is set to be 0.4, the distance difference value between a and B is calculated to be 1, and the point a and B are considered as points in a first verification result with unqualified data quality.

For another example, taking the difference of the spatial positions as an example, the difference of the spatial positions includes a difference of the distances, for the same geographic position, the coordinates of a point C in the first truth value data are 5,2,2, the coordinates of a point D in the second truth value data are 5.2,2,2, a first preset threshold value is set to be 0.5, the difference of the distances between C and D is calculated to be 0.2, and then the point A and the point B in the first verification result with qualified data quality are considered.

It should be appreciated that the magnitude of the corresponding first preset threshold may be different for different geometric vectors, and may be adjusted based on actual conditions.

And secondly, determining a second verification result according to the attribute difference value between the elements in the first truth value data and the elements in the second truth value data and a second preset threshold value.

In this implementation, the attribute is exemplified (by way of example only) by differences that may include differences in brightness, chromaticity, contrast, saturation, etc. between two elements.

That is, the two elements are respectively the attribute/expression element of the expression point corresponding to the first point cloud data and the second point cloud data at the same geographic position in the physical world.

Taking the brightness difference value as an example, comparing the brightness difference value between two elements at the same geographic position with a threshold value (namely a second preset threshold value) which can be accommodated, so as to determine first truth value data and second truth value data corresponding to the brightness difference value which are larger than the second preset threshold value, and first truth value data and second truth value data corresponding to the brightness difference value which are not larger than the second preset threshold value, wherein the first truth value data and the second truth value data are used as a second verification result, namely the second verification result comprises first truth value data and second truth value data which are distinguished based on the attribute of the elements.

In one implementation, element geometries, coordinates, element IDs, element magnitudes that are inconsistent or deviate from a threshold in the second validation result may be output.

For example, taking the example that the attribute difference value includes a brightness difference, for an element H in the first truth data at the same geographic location, the brightness is 15, the brightness is 16, a second preset threshold value is set at 0.6, the brightness difference value between H and I is calculated to be 1, and H and I are considered as elements with unqualified data quality in the second verification result.

For another example, taking the attribute difference value as an example, regarding the element M in the first truth value data at the same geographic location, the brightness is 15, the element N in the second truth value data, the brightness is 15.2, the second preset threshold value is set at 0.6, the brightness difference value between M and N is calculated to be 0.2, and then M and N are considered as the elements with qualified data quality in the second verification result.

It should be understood that the size of the corresponding second preset threshold may be different for different attributes of the element, and may be adjusted based on the actual situation.

In one possible implementation, fig. 6 is a schematic verification diagram provided by an embodiment of the present application, and as shown in fig. 6, the schematic verification diagram includes a lane line 1, a lane line 2, a lane line 3, and a lane line 4.

If lane 1 is information corresponding to an element in the first truth value data and lane 2 is information corresponding to an element in the second truth value data, it can be judged that the difference value of the spatial positions between the two lane lines is 6 and is larger than a first preset threshold value 5, and the element in the map, namely "lane 1", is considered to be inaccurate;

For another example, the lane line 3 is information corresponding to an element in the first truth value data, the lane line 4 is information corresponding to an element in the second truth value data, and it can be determined that the attribute between the two lane lines is inconsistent, that is, the lane line 3 is a dotted line lane line, the lane line 4 is a solid line lane line, the degree of the attribute inconsistency can correspond to a certain value, for example, 30%, if the degree of the attribute inconsistency is greater than a second preset threshold value of 10%, the element in the map, that is, "lane line 3", is considered inaccurate.

Further, after step 52, the verification result may be further divided into sub-verification results of different deviation dimensions according to the spatial position difference and the attribute difference.

In the implementation, the accuracy verification level can be analyzed according to the verification result, for example, the sub-verification result can be a multi-dimensional evaluation result such as a qualified level, a unqualified position, a unqualified scene, a unqualified accuracy deviation threshold value and the like.

That is, different threshold ranges, different scenes, etc. may be set.

It should be understood that the number of features such as the environment to be verified, etc. involved in the embodiments of the present application is not limited.

According to the verification method of the electronic map data, the verification results of at least two non-point cloud data are determined according to the difference between the first truth value data and the second truth value data. According to the technical scheme, based on comparison between the attributes and/or geometric vectors of the elements in the two point cloud data, verification results of different dimensions can be determined more remarkably.

On the basis of the above embodiment, fig. 7 is a schematic flow chart of an electronic map construction method according to an embodiment of the present application, and as shown in fig. 7, the method may include the following steps:

step 71, determining a verification result of collecting at least two non-point cloud data aiming at a preset area based on a verification method of electronic map data.

In this step, the verification result of verifying the at least two non-point cloud data is described in detail in the above embodiment, and indicates whether the data quality of the at least two non-point cloud data used for constructing the electronic map is acceptable.

And step 72, when the data quality indicated by the verification result meets the map data quality requirement, constructing an electronic map corresponding to the preset area through at least two non-point cloud data.

In this step, if the data quality indicated by the verification result meets the map data quality requirement, it is considered that at least two non-point cloud data can be used to construct an electronic map corresponding to the preset area.

In contrast, if the data quality indicated by the verification result does not meet the map data quality requirement, at least two non-point cloud data are considered to be unavailable for constructing the electronic map corresponding to the preset area, parameters of the image capturing unit, the position measuring unit and the inertia measuring unit can be adjusted to acquire the map data again, and other more accurate modes can be used for acquiring the map data so as to avoid inaccurate generated electronic maps.

Wherein the implementation of adjusting the parameters in the respective units may refer to the relevant content in step 23.

According to the electronic map construction method provided by the embodiment of the application, based on the verification method of the electronic map data in the embodiment, the verification result of collecting at least two non-point cloud data for the preset area is determined, and when the data quality indicated by the verification result meets the map data quality requirement, the electronic map corresponding to the preset area is constructed through the at least two non-point cloud data. According to the technical scheme, the verification result indicates whether the data quality meets the map data quality requirement, and the electronic map corresponding to the preset area can be accurately and reliably generated based on the non-point cloud data after the data quality meets the map data quality requirement, so that the safety and the high efficiency of automatic driving are improved.

The following are examples of the apparatus of the present application that may be used to perform the method embodiments of the present application. For details not disclosed in the embodiments of the apparatus of the present application, please refer to the embodiments of the method of the present application.

Fig. 8 is a schematic structural diagram of an electronic map data verification device according to an embodiment of the present application. As shown in fig. 8, the apparatus includes:

the acquisition module 81 is configured to acquire at least two non-point cloud data acquired by the acquisition vehicle for a preset area;

the processing module 82 is configured to perform a point cloud conversion process on at least two non-point cloud data, so as to obtain first point cloud data corresponding to the at least two non-point cloud data;

The determining module 83 is configured to determine a verification result of the at least two non-point cloud data according to a difference between the first point cloud data and the second point cloud data acquired from the preset area through the radar unit, where the verification result indicates data quality when the at least two non-point cloud data are used for constructing the electronic map.

In one or more embodiments, the processing module 82 is specifically configured to:

determining and collecting the running track information of the vehicle according to the positioning information and the vehicle speed information;

acquiring three-dimensional point cloud information around the collected vehicle according to the image information;

And converting the three-dimensional point cloud information into a global coordinate system according to the moving track information and the visual angle information of the camera unit to obtain first point cloud data.

In one or more embodiments, the determining module 83 determines a verification result of at least two non-point cloud data according to a difference between the first point cloud data and the second point cloud data acquired by the radar unit for the preset area, specifically for:

And determining verification results of at least two non-point cloud data according to differences of fourth point cloud data corresponding to the environment to be verified in the third point cloud data and the second point cloud data.

In one or more embodiments, the determining module 83 determines the verification result of at least two non-point cloud data according to the difference between the third point cloud data and the fourth point cloud data corresponding to the environment to be verified in the second point cloud data, which is specifically configured to:

Determining first truth value data corresponding to third point cloud data and second truth value data corresponding to fourth point cloud data;

And determining verification results of at least two non-point cloud data according to the difference between the first truth data and the second truth data.

In one or more embodiments, the determining module 83 determines a verification result of at least two non-point cloud data according to a difference between the first truth data and the second truth data, specifically for:

and/or the number of the groups of groups,

Determining a second verification result according to the attribute difference value between the elements in the first truth value data and the elements in the second truth value data and a second preset threshold value;

the verification results comprise a first verification result and/or a second verification result.

In one or more embodiments, the determining module 83 is further configured to:

And dividing the verification result into sub-verification results with different deviation dimensions according to the spatial position difference value and the attribute difference value.

In one or more embodiments, the determining module 83 is further configured to:

And adjusting parameters of the shooting unit, the position measuring unit and the inertia measuring unit according to verification results of at least two non-point cloud data.

In one or more embodiments, the environment to be verified includes at least one of a region to be verified in a preset region, a scene to be verified, an element to be verified, and a sample magnitude to be verified.

The verification device for electronic map data provided by the embodiment of the application can be used for executing the verification method for electronic map data in any embodiment, and the implementation principle and the technical effect are similar and are not repeated here.

Fig. 9 is a schematic structural diagram of an electronic map building device according to an embodiment of the present application. As shown in fig. 9, the apparatus includes:

the determining module 91 is configured to determine a verification result of collecting at least two non-point cloud data for a preset area based on the above-mentioned electronic map construction method;

and a construction module 92, configured to construct an electronic map corresponding to the preset area through at least two non-point cloud data when the data quality indicated by the verification result meets the map data quality requirement.

The verification device for electronic map data provided by the embodiment of the application can be used for executing the electronic map construction method in any embodiment, and the implementation principle and the technical effect are similar and are not repeated here.

It should be noted that, it should be understood that the division of the modules of the above apparatus is merely a division of a logic function, and may be fully or partially integrated into a physical entity or may be physically separated. The modules can be realized in the form of software which is called by the processing element, in the form of hardware, in the form of software which is called by the processing element, and in the form of hardware. In addition, all or part of the modules may be integrated together or may be implemented independently. The processing element described herein may be an integrated circuit having signal processing capabilities. In implementation, each step of the above method or each module above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in a software form.

Fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application, as shown in fig. 10, the electronic device may include a processor 101, a memory 102, and computer program instructions stored in the memory 102 and capable of running on the processor 101, where the verification method or the electronic map construction method of electronic map data provided in any of the foregoing embodiments is implemented when the processor 101 executes the computer program instructions.

Alternatively, the above devices of the electronic apparatus may be connected by a system bus.

The memory 102 may be a separate memory unit or may be a memory unit integrated into the processor 101. The number of processors 101 is one or more.

It should be appreciated that the Processor 101 may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors 101, digital signal Processor 101 (DIGITAL SIGNAL Processor, DSP), application SPECIFIC INTEGRATED Circuit (ASIC), etc. The general purpose processor 101 may be a microprocessor 101 or the processor 101 may be any conventional processor 101 or the like. The steps of a method disclosed in connection with the present application may be embodied directly in hardware performed by the processor 101, or in a combination of hardware and software modules in the processor 101.

The system bus may be a peripheral component interconnect (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The system bus may be classified into an address bus, a data bus, a control bus, and the like. For ease of illustration, the figures are shown with only one bold line, but not with only one bus or one type of bus. The memory 102 may include a random access memory 102 (random access memory, RAM) and may also include a non-volatile memory 102 (NVM), such as at least one disk memory 102.

All or part of the steps for implementing the method embodiments described above may be performed by hardware associated with program instructions. The foregoing program may be stored in a readable memory 102. The program, when executed, performs the steps comprising the method embodiments described above, and the aforementioned memory 102 (storage medium) comprises a read-only memory 102 (ROM), RAM, flash memory 102, a hard disk, a solid state disk, a magnetic tape (english: MAGNETIC TAPE), a floppy disk (english: floppy disk), an optical disk (english: optical disk), and any combination thereof.

The electronic device provided by the embodiment of the application can be used for executing the verification method or the electronic map construction method of the electronic map data provided by any of the method embodiments, and the implementation principle and the technical effect are similar, and are not repeated here.

The embodiment of the application provides a computer readable storage medium, wherein computer instructions are stored in the computer readable storage medium, and when the computer instructions run on a computer, the computer is caused to execute the verification method or the electronic map construction method of electronic map data.

The computer readable storage medium described above may be implemented by any type of volatile or non-volatile memory device or combination thereof, such as static random access memory, electrically erasable programmable read-only memory, magnetic memory, flash memory, magnetic disk or optical disk. A readable storage medium can be any available medium that can be accessed by a general purpose or special purpose computer.

In the alternative, a readable storage medium is coupled to the processor such that the processor can read information from, and write information to, the readable storage medium. In the alternative, the readable storage medium may be integral to the processor. The processor and the readable storage medium may reside in an Application SPECIFIC INTEGRATED Circuits (ASIC). The processor and the readable storage medium may reside as discrete components in a device.

The embodiment of the application also provides a computer program product, which comprises a computer program, wherein the computer program is stored in a computer readable storage medium, the computer program can be read from the computer readable storage medium by at least one processor, and the verification method or the electronic map construction method of the electronic map data can be realized when the computer program is executed by the at least one processor.

It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for verifying electronic map data, comprising:

Determining verification results of the at least two non-point cloud data according to differences between the first point cloud data and the second point cloud data acquired by the radar unit for the preset area, wherein the verification results indicate data quality of the at least two non-point cloud data when the electronic map is constructed, and if the data quality indicated by the verification results meets the map data quality requirement, constructing the electronic map corresponding to the preset area through the at least two non-point cloud data.

2. The method of claim 1, wherein the non-point cloud data includes image information acquired by a camera unit in the acquisition vehicle, positioning information acquired by a position measurement unit in the acquisition vehicle, and vehicle speed information acquired by an inertial measurement unit in the acquisition vehicle.

3. The method of claim 2, wherein performing the point cloud conversion process on the at least two non-point cloud data to obtain first point cloud data corresponding to the at least two non-point cloud data, comprises:

4. A method according to any one of claims 1-3, wherein determining the verification result of the at least two non-point cloud data according to the difference between the first point cloud data and the second point cloud data acquired by the radar unit for the preset area comprises:

5. The method of claim 4, wherein the determining the verification result of the at least two non-point cloud data according to the difference between the third point cloud data and the fourth point cloud data corresponding to the environment to be verified in the second point cloud data comprises:

6. The method according to claim 5, wherein said determining a validation result of the at least two non-point cloud data based on a difference between the first and second truth data comprises:

and/or the number of the groups of groups,

7. The method of claim 6, wherein the method further comprises:

8. The method according to claim 2, wherein the method further comprises:

9. The method of claim 4, wherein the environment to be verified comprises at least one of an area to be verified, a scene to be verified, an element to be verified, and a sample level to be verified in the predetermined area.

10. The electronic map construction method is characterized by comprising the following steps of:

Determining a verification result of collecting at least two non-point cloud data for a preset area based on the verification method of the electronic map data of any one of claims 1 to 9;

11. An electronic device comprising a processor and a memory communicatively coupled to the processor;

The memory stores computer-executable instructions;

The processor executes computer-executable instructions stored in the memory to implement the method of any one of the preceding claims 1 to 9 or 10.

12. A computer readable storage medium having stored therein computer executable instructions which when executed by a processor are adapted to carry out the method of any one of the preceding claims 1 to 9 or 10.

13. A computer program product, characterized in that it comprises a computer program stored in a computer-readable storage medium, from which at least one processor can read, said at least one processor implementing the method according to any of the preceding claims 1 to 9 or 10 when executing said computer program.