Automatic driving data processing method and device

Abstract

The invention provides an automatic driving data processing method and device. The method comprises the following steps: acquiring a test data set of an automatic driving test, wherein the test data set comprises test data and a plurality of attribute features describing a test scene corresponding to the test data; receiving a data acquisition request for the test data set; wherein the data acquisition request indicates scene description information of a target test scene; the scene description information comprises attribute features used for describing the target test scene; and based on a plurality of attribute features of a test scene corresponding to the test data, obtaining target test data corresponding to the attribute features contained in the scene description information from the test data contained in the test data set, and generating a test data set corresponding to the target test scene based on the target test data.

Description

Technical Field

[0001] The implementation methods of this specification relate to the field of big data. More specifically, the implementation methods of this specification relate to an autonomous driving data processing method and device. Background Art

[0002] In the process of vehicle autonomous driving R&D and testing, driving tests need to be carried out in a variety of complex scenarios, which will generate a large amount of test data. Usually, a large amount of original test data needs to be classified according to the test scenarios, and further analysis and evaluation are carried out on the test data in different test scenarios. Summary of the invention

[0003] To overcome the problems existing in the related art, this specification provides the following methods and devices.

[0004] In a first aspect of the embodiments of this specification, a method for processing autonomous driving data is provided; the method comprises:

[0005] Acquire a test data set for an autonomous driving test, the test data set comprising test data and a number of attribute features describing a test scenario corresponding to the test data;

[0006] Receiving a data acquisition request for the test data set; wherein the data acquisition request indicates scene description information of a target test scene; the scene description information includes attribute features for describing the target test scene;

[0007] Test data corresponding to the attribute features included in the scene description information is acquired from the test data, and a test data set corresponding to the target test scene is generated based on the acquired test data.

[0008] In a second aspect of the embodiments of this specification, an autonomous driving data processing device is provided, the device comprising:

[0009] An acquisition unit, configured to acquire a test data set for an autonomous driving test, wherein the test data set includes test data and a plurality of attribute features describing a test scenario corresponding to the test data;

[0010] A receiving unit, configured to receive a data acquisition request for the test data set; wherein the data acquisition request indicates scene description information of a target test scene; the scene description information includes attribute features for describing the target test scene;

[0011] A generating unit obtains target test data corresponding to the attribute features contained in the scene description information from the test data contained in the test data set based on several attribute features of the test scene corresponding to the test data, and generates a test data set corresponding to the target test scene based on the target test data.

[0012] In a third aspect of the embodiments of the present specification, a storage medium is provided; a computer program is stored on the storage medium, and the computer program implements the steps of the above method when executed.

[0013] In a fourth aspect of the embodiments of the present specification, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the above method is implemented when the processor executes the program.

[0014] The above embodiments of this specification have at least the following beneficial effects:

[0015] In the above technical solution, by extracting attribute features from the raw data of the autonomous driving test and mining scenarios based on the attribute feature combinations, direct labeling and classification of the raw data based on the scenarios is avoided, providing a flexible and convenient method for obtaining data of the scenarios of interest from the raw test data. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification, and for ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0017] Figure 1 The following schematically shows an architecture diagram of an autonomous driving data processing method according to an embodiment of the present specification;

[0018] Figure 2 A flowchart of an autonomous driving data processing method according to an embodiment of this specification is schematically shown;

[0019] Figure 3 A schematic diagram schematically shows an attribute feature system of a scene according to an embodiment of this specification;

[0020] Figure 4 A schematic diagram schematically shows a state attribute feature of a target vehicle according to an embodiment of the present specification;

[0021] Figure 5A schematic diagram schematically shows a state attribute feature of another traffic participant according to an embodiment of the present specification;

[0022] Figure 6 A schematic diagram schematically shows a behavior attribute feature of other traffic participants according to an embodiment of the present specification;

[0023] Figure 7 A schematic diagram schematically shows an environment and road feature attribute characteristics according to an embodiment of the present specification;

[0024] Figure 8 A schematic diagram schematically shows a method for processing autonomous driving data according to an embodiment of this specification;

[0025] Fig. 9 A schematic diagram schematically shows a scene tag according to an embodiment of the present specification;

[0026] Fig.10 A flowchart of an autonomous driving data processing method according to an embodiment of this specification is schematically shown;

[0027] Fig.11 A block diagram schematically shows an automatic driving data processing device according to an embodiment of the present specification;

[0028] Fig.12 A block diagram schematically shows another autonomous driving data processing device according to an embodiment of the present specification;

[0029] Fig.13 A hardware structure diagram of a computer device in which an autonomous driving data processing method according to an embodiment of the present specification is located is schematically shown.

[0030] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0031] The principles and spirit of this specification will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement this specification, and are not intended to limit the scope of this specification in any way. On the contrary, these embodiments are provided to make this specification more thorough and complete, and to fully convey the scope of this specification to those skilled in the art.

[0032] Those skilled in the art know that the embodiments of this specification can be implemented as a system, device, equipment, method or computer program product. Therefore, this specification can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0033] In the process of vehicle autonomous driving R&D and testing, driving tests need to be carried out in a variety of complex scenarios, which will generate a large amount of test data. Usually, a large amount of original test data needs to be classified according to the test scenarios, and further analysis and evaluation are carried out on the test data in different test scenarios.

[0034] Among them, the test data can be generated based on the data collected and recorded by the sensors of the autonomous driving test vehicle and the cameras, radars and other perception devices on the road.

[0035] During the testing process, the amount of data generated is massive, and the scenarios are unpredictable, inexhaustible, complex to handle, and flexible and multi-category. It is difficult to define all scenario classifications from the beginning, and many scenarios cannot be summarized by a single category.

[0036] Moreover, different systems or test functions of autonomous vehicles often focus on different scene features. Therefore, it is usually unrealistic to mine high-value scenes simply by manual labeling.

[0037] In a related technology, a method for organizing multi-sensor data annotation and simulation test scene labeling through a common data structure is provided. Specifically, the original test data of the autonomous driving test is annotated with scene labels by running the dimensions of design domain label, behavior label and management label. Each original test data can be annotated with its corresponding test scene in a fixed format of scene label.

[0038] However, the above-mentioned related technology is still a method of directly annotating scenes of original test data. The number of labels is limited, and the way to describe the labels is qualitative semantic expression. It cannot quantitatively describe the characteristics of the scenes represented by the labels, and cannot flexibly define scenes and mine them according to needs.

[0039] To overcome the problems existing in the related art, this specification provides the following embodiments.

[0040] The scheme of this specification is described in detail below with reference to the accompanying drawings.

[0041] See also Figure 1 , Figure 1 FIG. 1 is a schematic diagram of an architecture of an autonomous driving data processing system provided by an exemplary embodiment. Figure 1As shown, the system may include a network 10, a server 11, and several electronic devices, such as a computer 12, a computer 13, and a computer 14.

[0042] The server 11 may be a physical server including an independent host, or the server 11 may be a virtual server, a cloud server, etc. hosted by a host cluster. Computers 12-14 are just one type of electronic device that a user may use. In fact, a user may also use electronic devices such as tablet devices, mobile phones, PDAs (Personal Digital Assistants), wearable devices (such as smart glasses, smart watches, etc.), etc., and one or at least one embodiment of this specification does not limit this. The network 10 may include various types of wired or wireless networks.

[0043] In one embodiment, the server 11 can cooperate with the computers 12-14; wherein the computers 12-14 can accept user operations and upload the received commands and files to the server 11 through the network 10, and then the server 11 processes the files based on the scheme of this specification. In another embodiment, the computers 12-14 can independently implement the scheme of this specification; wherein the computers 12-14 accept user operations and process the received commands and files based on the scheme of this specification to realize autonomous driving data processing.

[0044] See also Figure 2 , Figure 2 is a flowchart of an automatic driving data processing method provided by an exemplary embodiment, the method is applied to a processing device, and the processing device can be Figure 1 The server 11 or computers 12-14 etc. are shown.

[0045] The method may include the following steps:

[0046] Step 202: Obtain a test data set for autonomous driving testing, where the test data in the test data set includes several attribute features extracted from the test data and used to describe a test scene corresponding to the test data.

[0047] In the process of vehicle autonomous driving R&D and testing, driving tests need to be carried out in a variety of complex scenarios, which will generate a large amount of test data. Due to the complexity of the test scenarios, the amount of test data is often very large, and the test data set for autonomous driving testing contains a large amount of raw test data.

[0048] These raw data can be generated based on data collected and recorded by sensors of autonomous driving test vehicles and perception devices such as cameras and radars on the road.

[0049] Generally, the original test data can be divided into two parts. The first part is the vehicle-side test data collected by the vehicle-side sensors of the target vehicle for autonomous driving testing. The second part is the roadside test data based on the vehicle trajectory data collected and recorded by perception devices such as cameras and radars on the road.

[0050] The vehicle-side test data may specifically include various types of data, such as information such as the position, speed, size and shape of surrounding objects recorded by the vehicle's laser radar (LiDAR), camera, radar and ultrasonic sensors; information such as the vehicle's position, speed and direction provided by the GPS (Global Positioning System); the vehicle's motion state and posture information such as the vehicle's acceleration and angular velocity measured by the inertial navigation system; vehicle state data such as the vehicle speed, steering angle, braking status, etc. provided by the vehicle operating system; control command data such as steering angle, braking force and acceleration generated by the automatic driving system, etc. This specification does not limit the specific types of test data.

[0051] Similarly, roadside test data may also include multiple types of data, such as the location and speed information of the test vehicle recorded by fixed cameras, radars, etc. on the road, the type and location information of other traffic participants related to the test vehicle, etc.

[0052] In an exemplary embodiment shown in this specification, a standardized data format may be used to represent the original test data, or when processing the original test data, the original test data may be converted into a standardized data format. For example, the OpenDrive / OpenScenario format may be used to represent the test data.

[0053] OpenDrive is a standardized simulation static scene format, and OpenScenario is a standardized description language and variable information format for simulation dynamic driving scenes. Standardized test data formats are conducive to extracting attribute features corresponding to test scenarios from test data.

[0054] In practical applications, all original test data may be converted into a standardized data format; since converting standardized format data usually takes up more computing resources and storage resources, part of the original test data may also be converted into a standardized data format.

[0055] During the driving test, for some special scenarios, such as scenarios where certain specific raw data (such as the main vehicle speed, acceleration, etc.) change dramatically, it can be determined that they are important data and converted into standardized format data for storage.

[0056] In order to determine the test scenario corresponding to the test data, attribute features corresponding to the test data may be extracted from the test data.

[0057] The attribute features corresponding to the test data can characterize the features of the test scenarios corresponding to the test data. By constructing a set of attribute feature systems, the test data can be characterized by attribute features without directly defining the test data as a fixed scenario.

[0058] Specifically, the attribute characteristics of the test data may include the attribute characteristics of the target vehicle undergoing the autonomous driving test, as well as the attribute characteristics of the environment and roads related to the target vehicle. When there are other traffic participants related to the target vehicle, it may also include relevant attribute characteristics of the other traffic participants related to the target vehicle.

[0059] like Figure 3 As shown, Figure 3 A schematic diagram of an attribute feature system of a scene according to an implementation manner of this specification is schematically shown.

[0060] In a typical test scenario, the test scenario usually includes the target vehicle (also called the main vehicle, Ego) performing autonomous driving, other traffic participants Actors related to the target vehicle, and the environment and roads in the test scenario.

[0061] Therefore, a scene attribute feature system can be constructed through the above three aspects.

[0062] The test data can be uniquely determined by the test time and the ID information of the target vehicle: the status of the target vehicle can be determined through the test time and the ID information of the target vehicle; the status of other traffic participants related to the target vehicle can also be determined through the test time and the ID information of the target vehicle; the location of the target vehicle can be determined through the test time and the ID information of the target vehicle, thereby determining the environment and road information in the test scene.

[0063] ODD (Operational Design Domain) refers to the operational design domain in which the autonomous driving system can operate safely. It can describe the environment, road type, weather conditions, speed range and other operational restrictions applicable to the autonomous driving system. In this manual, ODD is used to refer to the environment and road conditions in the test scenario.

[0064] For the target vehicle, its attribute characteristics may mainly include the behavior characteristics of the target vehicle.

[0065] like Figure 4 As shown, Figure 4 A schematic diagram schematically shows a state attribute feature of a target vehicle according to an embodiment of the present specification.

[0066] The behavior characteristics of the target vehicle can be specifically divided into lateral behavior characteristics and longitudinal behavior characteristics.

[0067] The lateral behavior characteristics refer to the behavior characteristics in the direction perpendicular to the lane in which the vehicle is traveling, such as turning behavior and lane change behavior. Turning behavior can be further divided into left turn, right turn and U-turn, and lane change can be further divided into left lane change and right lane change.

[0068] The longitudinal behavior characteristics refer to the behavior characteristics of the vehicle in the lane direction. For example, it can include forward behavior, braking behavior and parking behavior. The forward behavior can be further divided into normal cruising behavior, acceleration behavior and deceleration behavior, and can include specific forms of speed and acceleration.

[0069] The attribute characteristics of other traffic participants related to the target vehicle may mainly include their state attribute characteristics and behavior attribute characteristics.

[0070] Other traffic participants related to the target vehicle refer to other traffic participants that may have an impact on the driving behavior of the target vehicle. Generally, other traffic participants whose distance from the target vehicle is less than a preset threshold, including other vehicles, non-motor vehicles, pedestrians, animals or other objects, can be regarded as other traffic participants related to the target vehicle. Specifically, different thresholds can be preset for different types of traffic participants or traffic participants in different directions. For example, a smaller threshold can be preset for stationary obstacles that are not in the same lane as the target vehicle, and a larger threshold can be preset for other vehicles that are in the same lane as the target vehicle and located in front of the target vehicle, etc. This specification does not specifically limit the method and value of the preset threshold.

[0071] like Figure 5 As shown, Figure 5 A schematic diagram schematically shows a state attribute feature of another traffic participant according to an embodiment of the present specification.

[0072] Since the test scenario focuses on the changes in the driving environment of the target vehicle, the state attribute characteristics of other traffic participants in the scenario are relative to the target vehicle, which may include: driving direction (same, opposite, intersecting, etc. as the target vehicle); position (lateral position, longitudinal position, distance, etc. from the target vehicle).

[0073] like Figure 6 As shown, Figure 6 A schematic diagram schematically shows the behavioral attribute characteristics of other traffic participants according to an embodiment of the present specification.

[0074] For other traffic participants in the scene, their behavior attribute characteristics that may affect the target vehicle are calculated, and they can be specifically classified according to the types of other traffic participants.

[0075] Other traffic participants in the scene may include other motor vehicles, non-motor vehicles, pedestrians or animals related to the target vehicle, roadblocks and other obstacles, etc.

[0076] For example, for other motor vehicles in the scene, their behavior attribute characteristics can be represented by attribute characteristics similar to those of the target vehicle, such as various lateral behavior characteristics and longitudinal behavior characteristics.

[0077] As for other obstacles in the scene, such as pedestrians or animals, their behaviors may not be restricted by lanes and can be described using behavioral attribute features such as walking, sudden stop, and standing still.

[0078] like Figure 7 As shown, Figure 7 A schematic diagram of an environment and road feature attribute characteristics according to an embodiment of the present specification is schematically shown.

[0079] The ODD of the test scene, or the attribute characteristics of the environment and road characteristics, can be described using multiple dimensions such as road scene attribute characteristics, environmental condition attribute characteristics, dynamic element attribute characteristics, etc. For example, the road scene attribute characteristics can be divided into road type attribute characteristics, road shape attribute characteristics, road edge attribute characteristics, road surface attribute characteristics, sign attribute characteristics, etc.

[0080] In summary, the attribute features of the test scenario can describe the test scenario in detail. The more attribute features used, the more detailed the description.

[0081] Step 204: Receive a data acquisition request for the test data set; wherein the data acquisition request indicates scene description information of a target test scene; and the scene description information includes attribute features for describing the target test scene.

[0082] When a user needs to obtain test data of a specific test scenario from a test data set for further analysis and processing, he can send a data acquisition request. The above data acquisition request should be able to indicate the attribute characteristics of the test scenario that needs to be acquired.

[0083] Specifically, the scene description information may be used to describe the attribute characteristics of the target test scene.

[0084] The scenario description information needs to clearly describe one or more attribute characteristics corresponding to the target test scenario so as to filter the test data according to the attribute characteristics.

[0085] The target test scenario may be a user-defined test scenario defined by the user according to the needs. The user-defined test scenario is defined according to the scenario description information.

[0086] The format of the scenario description information should correspond to the format of the attribute characteristics to clearly describe the attribute characteristics of the target test scenario.

[0087] For example, a user-defined test scenario: when the main vehicle is driving, another motor vehicle cuts in from the lane to the side of the lane where the main vehicle is located and moves in front of the main vehicle.

[0088] The customized test scenario may correspond to the following attribute characteristics: the type of other participants related to the main vehicle is a motor vehicle, the motor vehicle is in front of the main vehicle, the distance between the motor vehicle and the main vehicle is less than 30 meters, and the motor vehicle has the behavior of entering the lane where the main vehicle is located.

[0089] In an illustrative embodiment, the above four statements may be used as the scene description information of the target test scene.

[0090] In another exemplary embodiment, the following standardized operator logic statements may be used as the scenario description information of the target test scenario:

[0091] actor.type = 'car';

[0092] actor.in_front_of_ego=true;

[0093] actor.distance_to_ego<30;

[0094] actor.entering_ego_lane=true.

[0095] This specification does not specifically limit the representation form of the scene description information. However, the scene description information should be able to clearly describe the attribute characteristics of the target test scene.

[0096] The scene description information may describe the distribution and value of the attribute characteristics of the target test scene, wherein the value of the attribute characteristics may be a fixed value or a range interval. If the value of a certain attribute characteristic of the target test scene described by the scene description information is a range interval, the value of the attribute characteristic may be deemed to satisfy the description information if it falls within the range interval.

[0097] Through the above description information, a user-defined target scenario can be constructed based on the attribute characteristics of the target scenario. By using attribute characteristics or a combination of attribute characteristics, it is easy to mine interesting scenarios from the test database. Customizing the scenario mining logic based on attribute characteristics can mine complex test scenarios. In addition, different test scenarios can have overlapping parts, which is more flexible.

[0098] Step 206: Based on several attribute features of the test scene corresponding to the test data, obtain target test data corresponding to the attribute features contained in the scene description information from the test data contained in the test data set, and generate a test data set corresponding to the target test scene based on the target test data.

[0099] The target test scene is determined through the scene description information and the distribution and value of the attribute characteristics corresponding to the scene description information. Based on the distribution and value of the attribute characteristics of the target test scene, the test data that meets the target test scene can be obtained from the test data set of the autonomous driving test. The test data set composed of the data that meets the target test scene is the test data set under the target test scene.

[0100] In addition to directly building user-defined test scenarios, you can also determine a batch of fixed, well-defined, and more commonly used general test scenarios in advance based on the attribute characteristics of the scenarios.

[0101] For some typical test scenarios, which are usually used frequently, there may be a need to repeatedly obtain test data for these scenarios.

[0102] For these commonly used test scenarios, it is troublesome to enter complex description information every time they are used. Therefore, a group of commonly used typical test scenarios can be preset in advance, which are called general test scenarios.

[0103] Specifically, the user may input a general test scenario saving instruction in advance. The general test scenario saving instruction includes the scenario description information corresponding to the common scenario to be saved, and the scenario tag corresponding to the general test scenario.

[0104] After receiving the general test scenario saving instruction, the general test scenario tag and the scenario description information corresponding to the general test scenario may be associated and saved.

[0105] When a data acquisition request for a test data set for an autonomous driving test is received as a general scenario data acquisition request including a general test scenario tag, the scene description information corresponding to the general test scenario tag can be automatically determined based on the general test scenario tag.

[0106] Through the above method, the corresponding scenario description information function can be automatically implemented for the saved general test scenarios using only simple test scenario tags.

[0107] In an exemplary embodiment shown in this specification, the test data set of the autonomous driving test includes a dynamically updated autonomous driving test data set. As the autonomous driving test is gradually improved, the data of the autonomous driving test set can be dynamically updated.

[0108] For dynamically updated data sets, even for the same target scene, the target test set obtained each time is different. Therefore, it is more convenient and faster to use a general scene data acquisition method when targeting the same target scene in a dynamically updated test data set.

[0109] See also Figure 8 , Figure 8 A schematic diagram of an autonomous driving data processing method according to an embodiment of the present specification is schematically shown.

[0110] A variety of general test scenarios can be set to meet various general test scenario data acquisition requirements. Each general scenario is stored in the form of a corresponding relationship between a general scenario tag and the scenario description information corresponding to the general scenario tag.

[0111] In addition, during the data acquisition process of the user-defined test scenario, it is also possible to support saving the user-defined test scenario as a common test scenario, and the user only needs to specify a common test scenario tag for it to save it.

[0112] For example, the user-defined test scenario in which another motor vehicle cuts into the front of the main vehicle from a lane to the side of the lane where the main vehicle is traveling, can be saved as a general test scenario and marked with the general test scenario label "cutting into lane".

[0113] When the user needs to obtain test data for this test scenario next time, he only needs to include the general test scenario tag "entering the lane" in the data acquisition request. When the general test scenario tag "entering the lane" is obtained, the saved correspondence can be automatically queried to determine the scene description information corresponding to the general test scenario tag "entering the lane" "the type of other participants related to the main car is a motor vehicle, the motor vehicle is in front of the main car, the distance between the motor vehicle and the main car is less than 30 meters, and the motor vehicle has the behavior of entering the lane where the main car is located" or the scene description information "actor.type = 'car'; actor.in_front_of_ego = true; actor.distance_to_ego<30; actor.entering_ego_lane = true.".

[0114] See also Fig. 9 , Fig. 9 A schematic diagram of an autonomous driving data processing method according to an embodiment of the present specification is schematically shown.

[0115] The following combination Fig. 9 The overall process of the autonomous driving data processing method is explained.

[0116] A large amount of raw data during the autonomous driving test, including vehicle-side test data and roadside test data, is stored in a standard format;

[0117] By establishing an attribute feature system, the corresponding attribute features are extracted from the original data, including the main vehicle behavior attribute features, other participant types, state attribute features and behavior attribute features, as well as ODD attribute features (road and environment attribute features).

[0118] When it is necessary to obtain test data of a specific target test scenario from the original data, the target test scenario can be constructed based on the attribute feature system. Specifically, the distribution and value of the attribute features can be described using the scenario description information.

[0119] For custom scenarios, you can directly use the scenario description information, such as custom query operators, to build target test scenarios to filter and select test data for the target scenarios from the original data set.

[0120] For custom scenarios with reusable value, their scenario description information or custom query operators can be saved, precipitated into common test scenarios, and represented using common test scenario labels.

[0121] In addition, a batch of common test scenarios can be saved separately, and the scenario description information can be represented by common test scenario tags.

[0122] For common test scenarios, you only need to use relatively simple common test scenario tags, which can be automatically converted into corresponding scenario description information.

[0123] After obtaining the test data set of the target test scenario, the performance of autonomous driving in the target test scenario can be analyzed based on the test data set corresponding to the target test scenario, and an autonomous driving performance analysis report for the target vehicle can be generated to evaluate the autonomous driving performance of the target vehicle in the target test scenario.

[0124] In the exemplary embodiment of this specification, another method for processing autonomous driving data is also provided. Fig.10 , Fig.10 is a flowchart of an automatic driving data processing method provided by an exemplary embodiment, the method is applied to a processing device, and the processing device can be Figure 1 The server 11 or computers 12-14 etc. are shown.

[0125] The method may include the following steps:

[0126] Step 1002: Obtain a test data set for autonomous driving testing.

[0127] Step 1004: extracting a number of attribute features corresponding to the test data in the test data set; wherein the attribute features are used to describe the test scenario corresponding to the test data.

[0128] Step 1006: Add the several attribute features as labels of the test data to the test data to form a labeled test data set; the labeled test data set is used to respond to a data acquisition request received for the test data set, wherein the data acquisition request indicates scene description information of a target test scene, and the scene description information includes attribute features for describing the target test scene, obtain test data corresponding to the attribute features included in the scene description information from the test data set, and generate a test data set corresponding to the target test scene based on the acquired test data.

[0129] The autonomous driving data processing method shown in this embodiment is a method for extracting several attribute features in the test data that can describe the test scenario from the original test data obtained from the autonomous driving test.

[0130] Through this method, the original test data in the test data set can be added with attribute feature labels related to the test scenario.

[0131] When it is necessary to obtain test data for a specific target scenario from a test data set, the target test scenario can be described using scene description information containing attribute feature labels, and test data having attribute features corresponding to the target test scenario can be obtained from the test data set to generate a test data set corresponding to the target test scenario.

[0132] For how to extract attribute features from the test data set, please refer to the detailed description in the previous embodiment of this specification, which will not be repeated here.

[0133] In an exemplary embodiment shown in this specification, the extracting of several attribute features corresponding to the test data in the test data set includes:

[0134] Based on the test data in the test data set, extracting a number of attribute features for describing a test scenario corresponding to the test data;

[0135] The several attribute characteristics include one or more of the following:

[0136] Behavioral attribute characteristics of the target vehicle for autonomous driving testing;

[0137] Environmental attribute characteristics related to the target vehicle;

[0138] Road attribute characteristics related to the target vehicle.

[0139] In an exemplary embodiment of this specification, an automatic driving data processing device is also provided. Fig.11 , Fig.11 This is a block diagram of an autonomous driving data processing device according to an embodiment of this specification.

[0140] The device may include the following components:

[0141] A first acquisition unit 1110 is configured to acquire a test data set for an autonomous driving test, wherein the test data set includes test data and a plurality of attribute features describing a test scenario corresponding to the test data;

[0142] The receiving unit 1120 is configured to receive a data acquisition request for the test data set; wherein the data acquisition request indicates scene description information of a target test scene; and the scene description information includes attribute features for describing the target test scene.

[0143] The generating unit 1130 is used to obtain target test data corresponding to the attribute features contained in the scene description information from the test data contained in the test data set based on several attribute features of the test scene corresponding to the test data, and generate a test data set corresponding to the target test scene based on the target test data.

[0144] Optionally, the data acquisition request includes a custom scenario data acquisition request; the target test scenario includes a user-defined test scenario;

[0145] The data acquisition request indicates the scene description information of the target test scene, including:

[0146] The custom scene data acquisition request includes scene description information of the user-customized test scene.

[0147] Optionally, the device further comprises:

[0148] The settling unit 1140 is configured to obtain a general test scenario saving indication, wherein the general test scenario saving indication includes a scenario tag corresponding to the general test scenario and scenario description information corresponding to the general test scenario;

[0149] The correspondence between the scenario label of the general test scenario and the scenario description information of the general test scenario is saved.

[0150] Optionally, the data acquisition request includes a general scenario data acquisition request; the general scenario data acquisition request includes a scenario tag of the general test scenario;

[0151] The receiving unit 1120 is specifically used to query the saved corresponding relationship and determine the scene description information corresponding to the scene label of the general test scene.

[0152] Optionally, the test data set for the autonomous driving test includes a dynamically updated test data set for the autonomous driving test.

[0153] Optionally, the test data includes vehicle-side test data collected by vehicle-side sensors of a target vehicle undergoing autonomous driving testing, and roadside test data collected by roadside sensors corresponding to the target vehicle deployed in a roadside environment.

[0154] Optionally, the attribute features corresponding to the test data include:

[0155] Behavioral attribute characteristics of the target vehicle for autonomous driving testing;

[0156] Environmental attribute characteristics related to the target vehicle;

[0157] Road attribute features related to the target vehicle.

[0158] Optionally, the attribute features corresponding to the test data further include:

[0159] Behavioral attribute characteristics of other traffic participants related to the target vehicle;

[0160] State attribute characteristics of other traffic participants related to the target vehicle.

[0161] Optionally, the other traffic participants related to the target vehicle include at least one of the following:

[0162] Motor vehicles related to the target vehicle; non-motor vehicles related to the target vehicle; pedestrians related to the target vehicle; other obstacles related to the target vehicle.

[0163] Optionally, the first acquisition unit 1110 is specifically configured to extract the distribution and value of the attribute feature corresponding to the test data from the test data contained in the test data set;

[0164] The scenario description information includes attribute features used to describe the target test scenario, including:

[0165] The scene description information includes the distribution and values ​​of attribute features used to describe the target test scene.

[0166] Optionally, the device further comprises:

[0167] The analyzing unit 1150 is used to analyze the performance of the autonomous driving in the target test scenario based on the test data set corresponding to the target test scenario.

[0168] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0169] In an exemplary embodiment of this specification, another autonomous driving data processing device is also provided. Fig.12 , Fig.12 This is a block diagram of an autonomous driving data processing device according to an embodiment of this specification.

[0170] The device may include the following components:

[0171] A second acquisition unit 1210 is configured to acquire a test data set for an autonomous driving test;

[0172] An extraction unit 1220, configured to extract a plurality of attribute features corresponding to the test data in the test data set; wherein the attribute features are used to describe the test scenario corresponding to the test data;

[0173] An adding unit 1230 is used to add the several attribute features as labels of the test data to the test data to form a labeled test data set; the labeled test data set is used to respond to a data acquisition request received for the test data set, wherein the data acquisition request indicates scene description information of a target test scene, and the scene description information includes attribute features for describing the target test scene, obtain test data corresponding to the attribute features included in the scene description information from the test data set, and generate a test data set corresponding to the target test scene based on the acquired test data.

[0174] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can refer to the partial description of the method embodiments. The device embodiments described above are only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on at least one network unit. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this specification. Ordinary technicians in this field can understand and implement it without paying creative work.

[0175] In an exemplary embodiment of the present specification, an embodiment of a device and a terminal to which the device is applied is also provided.

[0176] The embodiments of the device of this specification can be applied to computer devices, such as servers or terminal devices. The device embodiments can be implemented by software, hardware, or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor in which it is located reading the corresponding computer program instructions in the non-volatile memory into the memory and running them. From the hardware level, if Fig.13 As shown, Fig.13 A hardware structure diagram of a computer device 130 in which a device in an embodiment of this specification is located, except Fig.13 In addition to the processor 1310, memory 1330, network interface 1320, and non-volatile memory 1340 shown, the server or electronic device where the device is located in the embodiment may also include other hardware according to the actual function of the computer device, which will not be described in detail.

[0177] In the exemplary embodiments of this specification, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of this specification is stored. In some possible embodiments, various aspects of this specification can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of this specification described in the above "Exemplary Method" section of this specification.

[0178] According to the embodiment of the present specification, the program product for implementing the above method can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present specification is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0179] The program product may adopt any combination of one or at least one readable medium. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or at least one wire, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0180] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0181] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0182] Program code for performing the operations of the present specification may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0183] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0184] Although this specification contains many specific implementation details, these should not be interpreted as limiting the scope of any invention or the scope of the claimed protection, but are mainly used to describe the features of the specific embodiments of specific inventions. Certain features described in at least one embodiment in this specification may also be implemented in combination in a single embodiment. On the other hand, the various features described in a single embodiment may also be implemented separately in at least one embodiment or in any suitable sub-combination. In addition, although the features may work in certain combinations as described above and even initially claim protection, one or at least one feature from the claimed combination may be removed from the combination in some cases, and the claimed combination may point to a sub-combination or a variation of a sub-combination.

[0185] Similarly, although operations are depicted in a particular order in the accompanying drawings, this should not be understood as requiring these operations to be performed in the particular order shown or to be performed sequentially, or requiring all illustrated operations to be performed to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. In addition, the separation of various system units and components in the above-described embodiments should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product, or packaged into at least one software product.

[0186] Thus, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In some implementations, multitasking and parallel processing may be advantageous.

[0187] The above description is only a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.

Claims

1. A method for processing autonomous driving data, the method comprising: Acquire a test data set for an autonomous driving test, the test data set comprising test data and a number of attribute features describing the test data; Receiving a data acquisition request for the test data set; wherein the data acquisition request indicates scene description information of a target test scene; the scene description information includes attribute features for describing the target test scene; Based on several attribute features of the test scene corresponding to the test data, target test data corresponding to the attribute features contained in the scene description information is obtained from the test data contained in the test data set, and a test data set corresponding to the target test scene is generated based on the target test data.

2. The method according to claim 1, further comprising: Obtaining a general test scenario saving instruction, wherein the general test scenario saving instruction includes a scenario tag corresponding to the general test scenario and scenario description information corresponding to the general test scenario; The correspondence between the scenario label of the general test scenario and the scenario description information of the general test scenario is saved.

3. The method according to claim 2, The data acquisition request includes a general scenario data acquisition request; the general scenario data acquisition request includes a scenario tag of the general test scenario; By querying the saved corresponding relationship, the scene description information corresponding to the scene label of the general test scene is determined, and the scene description information indicated by the general scene data acquisition request is obtained.

4. The method according to claim 1, The test data set for the autonomous driving test includes a dynamically updated test data set for the autonomous driving test.

5. The method according to claim 1, The obtaining of the autonomous driving data set includes: For a target vehicle undergoing an autonomous driving test, vehicle-side test data of the target vehicle is collected by a vehicle-side sensor on the target vehicle, and roadside test data of the target vehicle is collected by a roadside sensor corresponding to the target vehicle and deployed in a roadside environment.

6. The method according to claim 1, The attribute characteristics corresponding to the test data include at least one of the following: Behavioral attribute characteristics of the target vehicle for autonomous driving testing; Environmental attribute characteristics related to the target vehicle; Road attribute features related to the target vehicle.

7. The method according to claim 6, The attribute features corresponding to the test data also include: Behavioral attribute characteristics of other traffic participants related to the target vehicle; Status attribute characteristics of other traffic participants related to the target vehicle; If the distance between other traffic participants and the target vehicle is less than a preset threshold, it is determined that the other traffic participants are related to the target vehicle.

8. The method according to claim 1, The scenario description information includes attribute features used to describe the target test scenario, including: The scene description information includes the distribution and values ​​of attribute features used to describe the target test scene.

9. The method according to claim 1, further comprising: Based on the test data set corresponding to the target test scenario, an autonomous driving performance analysis report for the target test scenario is generated.

10. A method for processing autonomous driving data, the method comprising: Get the test data set for autonomous driving testing; Extracting a number of attribute features corresponding to the test data in the test data set; wherein the attribute features are used to describe the test scenario corresponding to the test data; adding the number of attribute features as labels of the test data to the test data to form a labeled test data set; the labeled test data set is used to respond to receiving a data acquisition request for the test data set, wherein the data acquisition request indicates scene description information of a target test scenario, and the scene description information contains attribute features for describing the target test scenario, acquire test data corresponding to the attribute features contained in the scene description information from the test data set, and generate a test data set corresponding to the target test scenario based on the acquired test data.

11. The method according to claim 10, The extracting of a plurality of attribute features corresponding to the test data in the test data set includes: Based on the test data in the test data set, extracting a number of attribute features for describing a test scenario corresponding to the test data; The several attribute characteristics include one or more of the following: Behavioral attribute characteristics of the target vehicle for autonomous driving testing; Environmental attribute characteristics related to the target vehicle; Road attribute characteristics related to the target vehicle.

12. An automatic driving data processing device, the device comprising: An acquisition unit, configured to acquire a test data set for an autonomous driving test, wherein the test data set includes test data and a plurality of attribute features describing a test scenario corresponding to the test data; A receiving unit, configured to receive a data acquisition request for the test data set; wherein the data acquisition request indicates scene description information of a target test scene; the scene description information includes attribute features for describing the target test scene; A generating unit is used to obtain target test data corresponding to the attribute features contained in the scene description information from the test data contained in the test data set based on several attribute features of the test scene corresponding to the test data, and generate a test data set corresponding to the target test scene based on the target test data.

13. An automatic driving data processing device, the device comprising: A second acquisition unit is used for acquiring a test data set for an autonomous driving test; An extraction unit, used to extract a plurality of attribute features corresponding to the test data in the test data set; wherein the attribute features are used to describe the test scenario corresponding to the test data; An adding unit is used to add the several attribute features as labels of the test data to the test data to form a labeled test data set; the labeled test data set is used to respond to a data acquisition request received for the test data set, wherein the data acquisition request indicates scene description information of a target test scene, and the scene description information includes attribute features for describing the target test scene, obtain test data corresponding to the attribute features included in the scene description information from the test data set, and generate a test data set corresponding to the target test scene based on the acquired test data.

14. A storage medium having a computer program stored thereon, which, when executed, implements the steps of the method according to any one of claims 1 to 11.

15. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the following method is implemented: wherein, when the processor executes the program, the method as claimed in any one of claims 1 to 11 is implemented.

Images