CN115718702A - Automatic driving test scene library construction method and system - Google Patents
Automatic driving test scene library construction method and system Download PDFInfo
- Publication number
- CN115718702A CN115718702A CN202211478268.7A CN202211478268A CN115718702A CN 115718702 A CN115718702 A CN 115718702A CN 202211478268 A CN202211478268 A CN 202211478268A CN 115718702 A CN115718702 A CN 115718702A
- Authority
- CN
- China
- Prior art keywords
- scene
- information
- test
- road
- automatic driving
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention provides a method and a system for constructing an automatic driving test scene library, wherein the method comprises the following steps: determining a scene element information classification rule, dividing a scene element hierarchy according to the classification rule, and respectively identifying scene element information of each hierarchy in the drive test data according to the scene element hierarchy; fusing scene element information of each level and performing comprehensive description to generate an automatic driving test scene; and segmenting and storing the test scene to obtain a plurality of test scene segments with corresponding scene description. The invention describes the same scene through multi-level information, so that the scene information is more comprehensive and accurate, and the automatic driving can be more accurately tested in different scenes.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a method and a system for constructing an automatic driving test scene library, electronic equipment and a storage medium.
Background
The autopilot system test must be performed based on certain test scenarios. The test scenes comprise actual scenes and simulation scenes, most automatic driving test scenes are based on the simulation scenes, the classification levels of scene elements are set artificially, such as scene elements of levels of environment, ground features, dynamic information and the like, a large number of test scenes are obtained after the element combinations of all levels are crossed, the scenes are substantial repeated scenes, the special scenes (namely the long tail problem in automatic driving) needing to be responded in the actual driving scenes are difficult to screen, and the automatic driving algorithm or high-precision map data cannot be effectively verified according to the results of the test scenes.
Disclosure of Invention
The invention provides a method and a system for constructing an automatic driving test scene library, aiming at the technical problems in the prior art, wherein the same scene is described through multi-level information, so that the scene information is more comprehensive and accurate, and the automatic driving test scene library is beneficial to more accurately testing the automatic driving in different scenes.
According to a first aspect of the present invention, there is provided a method for constructing an automatic driving test scene library, including:
determining a scene element information classification rule, dividing a scene element hierarchy according to the classification rule, and respectively identifying scene element information of each hierarchy in the drive test data according to the scene element hierarchy;
fusing scene element information of each level and performing comprehensive description to generate an automatic driving test scene;
and segmenting and storing the test scene to obtain a plurality of test scene segments with corresponding scene description.
On the basis of the technical scheme, the invention can be improved as follows.
Optionally, the scene element hierarchy includes: road level information, traffic dynamic information and natural environment information; the road surface information is obtained through high-precision map elements, the traffic dynamic information is obtained through the current running condition and road passing condition of the vehicle in the driving process, and the natural environment information is obtained based on the natural environment in the current real vehicle driving scene marked by the drive test data.
Optionally, the process of acquiring the road surface information includes:
extracting road level information of each category according to data elements of a high-precision map, a road network topological structure and a division rule of the road level information in the scene element information classification rule, and extracting a corresponding high-precision map base map; the road surface information at least includes: road grade, road model, road network topology.
Optionally, the process of acquiring the traffic dynamic information includes:
under the condition that the drive test data exists, extracting corresponding traffic dynamic information and corresponding running track and timestamp information of the drive test data according to the driving behaviors and the running conditions of the self-vehicle which are actually recognized in the driving process of the self-vehicle and by combining the division rule of the traffic dynamic information in the scene element information classification rule; under the condition of the drive test data, the driving behavior is the actual driving behavior of the current vehicle;
under the condition of no drive test data, extracting corresponding traffic dynamic information and corresponding driving track information of the self vehicle according to possible driving routes and driving behaviors under the current road level information and by combining a division rule of the traffic dynamic information in the scene element information classification rule; and under the condition of no drive test data, the driving behaviors are all possible driving behaviors of the self-vehicle at the current position, which are automatically extracted according to the current high-precision map lane vector and road topological relation.
Optionally, the acquiring process of the natural environment information includes:
under the condition that the drive test data exist, extracting corresponding natural environment information and corresponding driving track and timestamp information of the drive test data according to the drive test data and by combining a dividing rule of the natural environment information in the scene element information classification rule;
under the condition of no drive test data, generating corresponding natural environment information according to all possible natural environment information under the current road level information and by combining the division rule of the scene element information classification rule on the natural environment information;
wherein the natural environment information includes at least: travel time, weather conditions, ambient temperature, lighting conditions during driving.
Optionally, the determining a classification rule of the scene element information, and dividing the scene element hierarchy according to the classification rule, further include:
defining a scene division rule, defining a parameter space described by scene elements at the same time, and generating a scene element parameter space, wherein the scene element parameter space comprises basic scene elements and special scene elements, and the basic scene elements at least comprise the width, the curvature radius and the gradient of a lane line.
Optionally, the scene element information of each level is fused and comprehensively described, and an automatic driving test scene is generated; the method comprises the following steps:
and fusing scene elements of the road surface information, the traffic dynamic information and the natural environment information, and comprehensively describing the fused scene according to a scene element information classification rule to obtain a multi-level driving scene and generate a final automatic driving test scene.
Optionally, the segmenting and storing the test scene to obtain a plurality of test scene segments with corresponding scene descriptions includes:
segmenting the identified original drive test data and the high-precision map according to the automatic drive test scene;
and classifying and storing the drive test data, the high-precision map and the scene description information which correspond to the same test scene obtained after segmentation so as to construct a scene library containing a plurality of test scene segments.
Optionally, the method further includes:
and extracting the parameter space and the parameter value of each specific test scene according to the scene description of different levels, generating a scene label and an index by adopting the parameter space and the parameter value of each specific test scene, and storing the scene label and the index in association with the corresponding specific test scene.
According to a second aspect of the present invention, there is provided an automatic driving test scenario library construction system, including:
the hierarchical identification module is used for determining a scene element information classification rule, dividing scene element hierarchies according to the classification rule and respectively identifying scene element information of each hierarchy in the drive test data according to the scene element hierarchies;
the information fusion module is used for fusing scene element information of each level and performing comprehensive description to generate an automatic driving test scene;
and the segmentation storage module is used for segmenting and storing the test scenes to obtain a plurality of test scene segments with corresponding scene descriptions.
According to a third aspect of the present invention, there is provided an electronic device, comprising a memory and a processor, wherein the processor is configured to implement the steps of the automatic driving test scenario library construction method when executing a computer management class program stored in the memory.
According to a fourth aspect of the present invention, there is provided a computer readable storage medium, on which a computer management-like program is stored, the computer management-like program, when being executed by a processor, implements the steps of the automatic driving test scenario library construction method.
According to the method, the system, the electronic equipment and the storage medium for constructing the automatic driving test scene library, provided by the invention, multi-level scenes are classified according to the information classification rule, different classification information is classified into different levels to describe the same scene, and various scene element information is identified according to different levels, so that the acquired scene information is more comprehensive and accurate. And secondly, scene recognition and extraction are carried out on the basis of a high-precision map and actually acquired drive test data, and compared with the traditional simulation scene, the simulation method is more representative and truthful, can reflect key factors influencing the functions of an automatic driving system in actual driving behaviors, and provides better reference for subsequent automatic driving function test evaluation. And thirdly, a large number of specific scenes can be generated based on high-precision map data, and quick and intelligent test iteration of the automatic driving function can be realized by combining an automatic test platform.
Drawings
FIG. 1 is a flow chart of a method for constructing an automatic driving test scene library according to the present invention;
FIG. 2 is a schematic diagram illustrating classification of inter-urban highways in road-level information according to an embodiment;
FIG. 3 is a schematic diagram illustrating classification of a city inner seal closed circuit in road level information according to an embodiment;
fig. 4 is a schematic diagram illustrating classification of parking scenes in road level information according to an embodiment;
fig. 5 is a schematic structural diagram of a component of an automatic driving test scene library construction system according to an embodiment;
fig. 6 is a schematic diagram of a hardware structure of a possible electronic device provided in the present invention;
fig. 7 is a schematic diagram of a hardware structure of a possible computer-readable storage medium according to the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
Fig. 1 is a flowchart of a method for constructing an automatic driving test scenario library, as shown in fig. 1, the method includes:
s1, determining a scene element information classification rule, dividing a scene element hierarchy according to the classification rule, and respectively identifying scene element information of each hierarchy in the real world according to the scene element hierarchy;
s2, fusing and comprehensively describing the scene element information of each level to generate an automatic driving test scene;
and S3, segmenting and storing the test scene to obtain a plurality of test scene segments with corresponding scene description.
It can be understood that, based on the defects in the background art, the embodiment of the invention provides a method for constructing an automatic driving test scene library.
The invention classifies multi-level scenes according to the information classification rule, classifies different classification information into different levels to describe the same scene, and identifies various scene element information according to different levels, so that the acquired scene information is more comprehensive and accurate. And secondly, scene recognition and extraction are carried out on the basis of the high-precision map and the actually acquired drive test data, so that the simulation system has higher representativeness and authenticity compared with the traditional simulation scene, can reflect key factors influencing the functions of the automatic driving system in actual driving behaviors, and provides better reference for subsequent automatic driving function test evaluation. And thirdly, a large number of specific scenes can be generated based on high-precision map data, and quick and intelligent test iteration of an automatic driving function can be realized by combining an automatic test platform.
In a possible embodiment, in step S1, the scene element hierarchy includes: road level information, traffic dynamic information and natural environment information; the road surface information is obtained through high-precision map elements, the traffic dynamic information is obtained through the running condition and road passing condition of the current vehicle in the driving process, and the natural environment information is obtained based on the natural environment in the current real vehicle driving scene marked by the road test data.
In a possible embodiment, the acquiring process of the road level information includes:
setting a certain rule based on SQL query language according to the data elements of the high-precision map, the road network topological structure and the division rule of the road level information in the scene element information classification rule, automatically extracting the road level information of each category, and extracting the corresponding base map of the high-precision map; the road level information may be understood as road structure information, which at least comprises: road grade, road model, road network topology.
Specifically, the road surface information is mainly divided into the following types according to the road type of the high-precision map: highway (urban high speed), urban expressway (urban closed circuit) and parking scene. Fig. 2 is a schematic diagram showing classification of a highway (inter-city highway) based on road level information, fig. 3 is a schematic diagram showing classification of an urban expressway (closed circuit inside a city) based on road level information, and fig. 4 is a schematic diagram showing classification of parking scenes based on road level information.
As shown in fig. 2, the road level information partition includes a primary structure and a secondary structure under the primary structure, where the primary structure has at least one type, and the secondary structure under the same primary structure has at least one type.
The main structure of the highway (urban highway) is divided into a common straight road, a curve road, a ramp/overpass, an intersection, an SA/PA (service area/parking area), an overpass, a toll station/inspection station, a bridge, a tunnel, an emergency scene and other special scenes according to the road type.
The main structure of the highway (urban highway) can be continuously divided into primary secondary structures according to specific road types, for example, the main structure of a ramp/overpass can be divided into an IC highway entrance and exit (a common road and a highway connecting road), a JTC highway connecting ramp, a JTC highway and an urban highway connecting ramp and the like.
The above-mentioned primary secondary structure of the highway (inter-urban highway) can be divided into secondary structures according to specific locations and road types. The secondary structure is the final road level information level. The primary secondary structure and the secondary structure can be added in a self-defining way according to the requirement.
As shown in fig. 3, the main structure of the urban expressway (urban closed circuit) is divided into a common straight road, a curve road, a ramp/overpass, an intersection, an SA/PA (service area/parking area), an overpass, a bridge, a tunnel and other special scenes according to the road type.
The main structure of the urban expressway (urban enclosed closed circuit) can be continuously divided into primary secondary structures according to specific road types, for example, the main structure of a ramp/overpass can be divided into an entrance and an exit of an IC urban expressway (a common road and an urban expressway connecting road), a JTC urban expressway connecting ramp and the like.
The primary secondary structure of the urban expressway (urban enclosed circuit) can be divided into secondary structures according to specific positions and road types. The secondary structure is the final road level information level. The primary secondary structure and the secondary structure can be added in a self-defining way according to the requirement.
As shown in FIG. 4, the main structure of the parking scene is divided into roadside parking spaces, outdoor parking spaces, underground parking spaces and other special scenes according to the types of roads.
The main structure of the parking scene can be continuously divided into primary secondary structures according to specific road types, for example, an outdoor parking scene can be divided into parking spaces, internal roads and the like.
The parking scene primary secondary structure may be divided into secondary structures according to specific locations and road types. The secondary structure is the final road level information level. The primary secondary structure and the secondary structure can be added in a self-defining way according to the requirement.
In a possible embodiment, the acquiring process of the traffic dynamics information includes:
under the condition that the drive test data exists, extracting corresponding traffic dynamic information and corresponding running track and timestamp information of the drive test data according to the driving behaviors and the running conditions of the self-vehicle which are actually recognized in the driving process of the self-vehicle and by combining the division rule of the traffic dynamic information in the scene element information classification rule; in the presence of drive test data, the driving behavior is the actual driving behavior of the current vehicle (whether or not traffic regulations are violated);
under the condition of no drive test data, extracting corresponding traffic dynamic information and corresponding driving track information of the self vehicle according to possible driving routes and driving behaviors under the current road level information and by combining a division rule of the traffic dynamic information in the scene element information classification rule; and under the condition of no drive test data, the driving behaviors are all possible driving behaviors of the self-vehicle at the current position, which are automatically extracted according to the current high-precision map lane vector and road topological relation.
The road traffic condition in the traffic dynamic information is defined according to the current road traffic state marked by the road test data when the road test data exists, and includes road congestion, road smoothness and other conditions.
In a possible embodiment, the acquiring process of the natural environment information includes:
under the condition that the drive test data exist, extracting corresponding natural environment information and corresponding driving track and timestamp information of the drive test data according to the drive test data and by combining a dividing rule of the natural environment information in the scene element information classification rule;
under the condition of no drive test data, generating corresponding natural environment information according to all possible natural environment information under the current road level information and by combining the division rule of the scene element information classification rule on the natural environment information;
wherein the natural environment information includes, but is not limited to: travel time during driving, weather conditions, ambient temperature, lighting conditions, etc.
The travel time in the natural environment information includes, but is not limited to: day, night, etc.
The weather conditions in the natural environment information include, but are not limited to: sunny days, rainy and snowy days, frozen ground, foggy days, strong wind days and the like.
The lighting conditions in the natural environment information include, but are not limited to, lighting angle, backlight, exit from a tunnel, and under an overhead channel.
In a possible embodiment, the determining a classification rule of scene element information, and dividing the scene element hierarchy according to the classification rule further includes:
defining a scene division rule, defining a parameter space described by scene elements at the same time, and generating a scene element parameter space, wherein the scene element parameter space comprises basic scene elements and special scene elements, and the basic scene elements at least comprise the width, curvature radius and gradient of a lane line.
It is understood that the basic scene element is a scene element provided for all scenes, such as the above-mentioned information of the width of the lane line, the curvature radius, and the gradient. The special scene element is defined when the special scene is described, for example, when the special scene is described in a tunnel scene, the width and height of the tunnel are the special scene elements. The parameter space of the scene element is composed of both the basic scene element and the special scene element.
In a possible embodiment, the scene element information of each level is fused and comprehensively described to generate an automatic driving test scene; the method comprises the following steps:
and fusing scene elements of the hierarchically identified road surface information, the traffic dynamic information and the natural environment information, and comprehensively describing the fused scene according to the classification rule of the scene element information to obtain a multi-level driving scene and generate a final automatic driving test scene.
In the scene fusion process, the scene element parameter space assigns specific parameter values when generating a specific driving scene, and the parameter values may be element parameters extracted from a high-precision map or parameters assigned manually. Specific scenes can be directly converted into test cases for subsequent function tests of the automatic driving system.
In this embodiment, in the scene generation processing process, the parameter space and the parameter value of each specific scene are extracted according to the scene descriptions of different levels.
In a possible embodiment, the segmenting and storing the test scenario to obtain a plurality of test scenario segments with corresponding scenario descriptions includes:
segmenting the identified original drive test data and the high-precision map according to the automatic drive test scene;
and classifying and storing the drive test data, the high-precision map and the scene description information which are obtained after segmentation and correspond to the same test scene so as to construct a scene library containing a plurality of test scene fragments.
It can be understood that, in the process of cutting and storing the scene segments, besides storing the scene raw drive test data and the high-precision map data segments, the description information (e.g. scene data metadata) of the scene classification design and the parameter space and parameter values of each specific scene need to be stored.
Generally, in the process of cutting and storing the scene segments, each specific scene may store, but is not limited to store, in addition to the above information, other information related to each specific scene, such as scene entry personnel and the like.
Generally, in the process of cutting and storing scene segments, a specific scene database is constructed, and each specific scene generated in the process of scene fusion is stored and managed.
In a possible embodiment, after the building a scenario library of each test scenario, the method further includes:
extracting the parameter space and the parameter value of each specific test scene according to the scene description of different levels, generating a scene label and an index by adopting the parameter space and the parameter value of each specific test scene, wherein the scene label is label information corresponding to each type of scene, and then associating the scene label and the index with the corresponding specific test scene for storage.
It can be understood that the generated scene labels and indexes can be used for searching various test required scenes obtained in the cutting and storing process in the subsequent automatic driving test.
Generally, in the scene retrieval and query process, data in a scene database can be retrieved and queried through a road level information primary structure and a road level information secondary structure in classification design, and query results are displayed in a list form.
Generally, in the scene retrieval and query process, data in a scene database can be retrieved and queried through a scene tag, and query results are displayed in a list form.
Generally, in the scene retrieval and query process, the scene query result may be subjected to secondary query through specific scene parameters, and the secondary query result is displayed in a list form.
Fig. 5 is a structural diagram of a system for constructing an automatic driving test scenario library according to an embodiment of the present invention, and as shown in fig. 5, the system for constructing an automatic driving test scenario library includes a hierarchical identification module, an information fusion module, and a segmentation storage module, where:
the hierarchical identification module is used for determining a scene element information classification rule, dividing scene element hierarchies according to the classification rule and respectively identifying scene element information of each hierarchy in the real world according to the scene element hierarchies;
the information fusion module is used for fusing scene element information of each level and performing comprehensive description to generate an automatic driving test scene;
and the segmentation storage module is used for segmenting and storing the test scenes to obtain a plurality of test scene segments with corresponding scene descriptions.
The system also comprises a retrieval module which is used for retrieving and inquiring each test scene segment.
It can be understood that the automatic driving test scenario base construction system provided by the present invention corresponds to the automatic driving test scenario base construction methods provided in the foregoing embodiments, and the relevant technical features of the automatic driving test scenario base construction system may refer to the relevant technical features of the automatic driving test scenario base construction method, and are not described herein again.
Referring to fig. 6, fig. 6 is a schematic view of an embodiment of an electronic device according to an embodiment of the invention. As shown in fig. 6, an embodiment of the present invention provides an electronic device 600, which includes a memory 610, a processor 620, and a computer program 611 stored in the memory 610 and operable on the processor 620, wherein the processor 620 implements the following steps when executing the computer program 611:
determining a scene element information classification rule, dividing a scene element hierarchy according to the classification rule, and respectively identifying scene element information of each hierarchy in the real world according to the scene element hierarchy;
fusing scene element information of each level and performing comprehensive description to generate an automatic driving test scene;
and segmenting and storing the test scene to obtain a plurality of test scene segments with corresponding scene description.
Referring to fig. 7, fig. 7 is a schematic diagram of an embodiment of a computer-readable storage medium according to the present invention. As shown in fig. 7, the present embodiment provides a computer-readable storage medium 700 having a computer program 711 stored thereon, the computer program 711, when executed by a processor, implementing the steps of:
determining a classification rule of the scene element information, dividing the scene element level according to the classification rule, and respectively identifying the scene element information of each level in the real world according to the scene element level;
fusing scene element information of each level and performing comprehensive description to generate an automatic driving test scene;
and segmenting and storing the test scene to obtain a plurality of test scene segments with corresponding scene description.
According to the method, the system and the storage medium for constructing the automatic driving test scene library, provided by the embodiment of the invention, multi-level scenes are classified according to the information classification rule, different classification information is classified into different levels to describe the same scene, and various scene element information is identified according to different levels, so that the acquired scene information is more comprehensive and accurate. And secondly, scene recognition and extraction are carried out on the basis of the high-precision map and the actually acquired drive test data, so that the simulation system has higher representativeness and authenticity compared with the traditional simulation scene, can reflect key factors influencing the functions of the automatic driving system in actual driving behaviors, and provides better reference for subsequent automatic driving function test evaluation. And thirdly, a large number of specific scenes can be generated based on high-precision map data, and quick and intelligent test iteration of the automatic driving function can be realized by combining an automatic test platform.
It should be noted that, in the foregoing embodiments, the description of each embodiment has an emphasis, and reference may be made to the related description of other embodiments for a part that is not described in detail in a certain embodiment.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A method for constructing an automatic driving test scene library is characterized by comprising the following steps:
determining a classification rule of the scene element information, dividing the scene element level according to the classification rule, and respectively identifying the scene element information of each level in the drive test data according to the scene element level;
fusing scene element information of each level and performing comprehensive description to generate an automatic driving test scene;
and segmenting and storing the test scene to obtain a plurality of test scene segments with corresponding scene description.
2. The method as claimed in claim 1, wherein the scene element hierarchy comprises: road surface information, traffic dynamic information and natural environment information; the road surface information is obtained through high-precision map elements, the traffic dynamic information is obtained through the running condition and road passing condition of the current vehicle in the driving process, and the natural environment information is obtained based on the natural environment in the current real vehicle driving scene marked by the road test data.
3. The method for constructing the automatic driving test scene library according to claim 2, wherein the process of acquiring the road level information comprises:
extracting road level information of each category according to the data elements of the high-precision map, the road network topological structure and the dividing rule of the road level information in the scene element information classification rule, and extracting a corresponding high-precision map base map; the road deck information at least includes: road grade, road model, road network topology.
4. The method for constructing the automatic driving test scene library according to the claim or the claim 3, wherein the process for acquiring the traffic dynamic information comprises the following steps:
under the condition of the presence of the drive test data, extracting corresponding traffic dynamic information and corresponding driving track and timestamp information of the drive test data according to driving behaviors and running conditions of the vehicle actually identified in the driving process of the vehicle and the division rule of the traffic dynamic information in the scene element information classification rule; under the condition of the drive test data, the driving behavior is the actual driving behavior of the current vehicle;
under the condition of no drive test data, extracting corresponding traffic dynamic information and corresponding driving track information of the self vehicle according to possible driving routes and driving behaviors under the current road level information and by combining a division rule of the traffic dynamic information in the scene element information classification rule; and under the condition of no drive test data, the driving behaviors are all possible driving behaviors of the self-vehicle at the current position, which are automatically extracted according to the current high-precision map lane vector and road topological relation.
5. The automatic driving test scene library construction method according to any one of claims 2 to 4, wherein the natural environment information acquisition process comprises:
under the condition that the drive test data exists, extracting corresponding natural environment information and corresponding driving tracks and timestamp information of the drive test data according to the drive test data and by combining a division rule of the natural environment information in the scene element information classification rule;
under the condition of no drive test data, generating corresponding natural environment information according to all possible natural environment information under the current road level information and by combining the division rule of the scene element information classification rule on the natural environment information;
wherein the natural environment information includes at least: travel time, weather conditions, ambient temperature, lighting conditions during driving.
6. The method as claimed in claim 2, wherein the determining a classification rule of scene element information and dividing a scene element hierarchy according to the classification rule further comprises:
defining a scene division rule, defining a parameter space described by scene elements at the same time, and generating a scene element parameter space, wherein the scene element parameter space comprises basic scene elements and special scene elements, and the basic scene elements at least comprise the width, the curvature radius and the gradient of a lane line.
7. The method for constructing the automatic driving test scene library according to claim 6, wherein scene element information of each level is fused and comprehensively described to generate an automatic driving test scene; the method comprises the following steps:
scene elements of the road surface information, the traffic dynamic information and the natural environment information are fused, and the fused scene is comprehensively described according to a scene element information classification rule, so that a multi-level driving scene is obtained, and a final automatic driving test scene is generated.
8. The method for constructing the automatic driving test scene library according to claim 1 or 7, wherein the step of segmenting and storing the test scene to obtain a plurality of test scene segments with corresponding scene descriptions comprises the steps of:
segmenting the identified original drive test data and the high-precision map according to the automatic drive test scene;
and classifying and storing the drive test data, the high-precision map and the scene description information which are obtained after segmentation and correspond to the same test scene so as to construct a scene library containing a plurality of test scene fragments.
9. The method for constructing the automatic driving test scene library according to claim 8, further comprising:
and extracting the parameter space and the parameter value of each specific test scene according to scene description of different levels, generating a scene label and an index by adopting the parameter space and the parameter value of each specific test scene, and associating the scene label and the index with the corresponding specific test scene for storage.
10. An automatic driving test scene library construction system is characterized by comprising:
the hierarchical identification module is used for determining a classification rule of the scene element information, dividing the scene element hierarchy according to the classification rule and respectively identifying the scene element information of each hierarchy in the drive test data according to the scene element hierarchy;
the information fusion module is used for fusing scene element information of each level and performing comprehensive description to generate an automatic driving test scene;
and the segmentation storage module is used for segmenting and storing the test scenes to obtain a plurality of test scene segments with corresponding scene descriptions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211478268.7A CN115718702A (en) | 2022-11-23 | 2022-11-23 | Automatic driving test scene library construction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211478268.7A CN115718702A (en) | 2022-11-23 | 2022-11-23 | Automatic driving test scene library construction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115718702A true CN115718702A (en) | 2023-02-28 |
Family
ID=85256132
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211478268.7A Pending CN115718702A (en) | 2022-11-23 | 2022-11-23 | Automatic driving test scene library construction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115718702A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116204791A (en) * | 2023-04-25 | 2023-06-02 | 山东港口渤海湾港集团有限公司 | Construction and management method and system for vehicle behavior prediction scene data set |
-
2022
- 2022-11-23 CN CN202211478268.7A patent/CN115718702A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116204791A (en) * | 2023-04-25 | 2023-06-02 | 山东港口渤海湾港集团有限公司 | Construction and management method and system for vehicle behavior prediction scene data set |
| CN116204791B (en) * | 2023-04-25 | 2023-08-11 | 山东港口渤海湾港集团有限公司 | Construction and management method and system for vehicle behavior prediction scene data set |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109754594B (en) | Road condition information acquisition method and equipment, storage medium and terminal thereof | |
| CN111930791B (en) | Similarity calculation method and system for vehicle track and storage medium | |
| CN111599183B (en) | Automatic driving scene classification and identification system and method | |
| CN110798805B (en) | Data processing method and device based on GPS track and storage medium | |
| Alzraiee et al. | Detecting of pavement marking defects using faster R-CNN | |
| CN109859505B (en) | Early warning processing method, device, server and medium for high-speed station | |
| CN115718702A (en) | Automatic driving test scene library construction method and system | |
| CN114707035A (en) | Visual traffic initial point analysis system | |
| CN113327248A (en) | Tunnel traffic flow statistical method based on video | |
| CN112446343B (en) | Vehicle-mounted point cloud road shaft-shaped object machine learning automatic extraction method integrating multi-scale features | |
| CN115440071B (en) | Automatic driving illegal parking detection method | |
| Khan | Vehicle and pedestrian detection using YOLOv3 and YOLOv4 for self-driving cars | |
| CN111368868A (en) | Method, device and equipment for determining vehicle fake plate | |
| Scholtes et al. | OMEGAFormat: A comprehensive format of traffic recordings for scenario extraction | |
| CN114116854A (en) | Track data processing method, device, equipment and storage medium | |
| CN112765392B (en) | High-speed rail train control positioning method and system based on image matching | |
| CN114527675A (en) | Automatic driving simulation test method, device, equipment and storage medium | |
| TWI617998B (en) | System and method for car number identification data filtering | |
| Chmiel et al. | Rule system for speed limit determination on national roads in Poland | |
| CN113592915B (en) | End-to-end rotating frame target searching method, system and computer readable storage medium | |
| LU503001B1 (en) | Method for predicting pedestrian crossing without signal lamp control | |
| Carnot et al. | Enhancing Roadway Safety: LiDAR-based Tree Clearance Analysis | |
| CN116739531B (en) | Digital management method and platform for expressway assets | |
| CN115577130A (en) | Scene library construction method and device for vehicle simulation test and vehicle | |
| CN116910986A (en) | Simulation drive test system of intelligent network-connected automobile |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |