Method for constructing automatic driving test scene based on solid-state laser radar system

Abstract

The invention relates to the technical field of automatic driving, in particular to an automatic driving test scene construction method based on a solid-state laser radar system, and the method comprises the steps: firstly obtaining mixed traffic flow real vehicle data based on a pre-configured traffic data collection device; screening target test scenes and test cases corresponding to the target test scenes in a preset test scene library according to the mixed traffic flow real vehicle data; obtaining vehicle motion data based on a pre-established road data acquisition vehicle; and finally configuring detailed information of the target test scene and detailed information of the test case corresponding to each target test scene according to the vehicle motion data. The scene constructed by the method enables the test process to be more real and effective, and is beneficial for verifying the reliability and environmental adaptability of the radar system in facing complex and diverse real traffic environments.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a method for constructing an autonomous driving test scenario based on a solid-state lidar system. Background Art

[0002] The solid-state LiDAR system is an important part of the vehicle's autonomous driving system. Therefore, in order to ensure the reliability and adaptability of the solid-state LiDAR system during the autonomous driving process, the national standard "Performance Requirements and Test Methods for Vehicle-mounted LiDAR" stipulates the performance requirements and test methods of the LiDAR itself. However, the national standard does not specify the relevant content of the autonomous driving test scenario.

[0003] In the related technologies, when faced with complex and diverse real traffic environments, there is a lack of methods for constructing autonomous driving test scenarios based on solid-state lidar systems, and the reliability and adaptability of solid-state lidar systems during autonomous driving cannot be guaranteed. Summary of the invention

[0004] In order to solve or partially solve the problems existing in the related technologies, the present application provides a method for constructing an autonomous driving test scenario based on a solid-state lidar system, which can make the testing process more realistic and effective, and is conducive to verifying the reliability and environmental adaptability of the radar system when facing complex and diverse real traffic environments.

[0005] The first aspect of the present application provides a method for constructing an autonomous driving test scenario based on a solid-state laser radar system, comprising:

[0006] Acquire real vehicle data of mixed traffic flow based on pre-configured traffic data collection devices;

[0007] Filtering target test scenarios and test cases corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data;

[0008] Obtain vehicle motion data based on a pre-built road data collection vehicle;

[0009] The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured according to the vehicle motion data.

[0010] Optionally, the obtaining of mixed traffic flow real vehicle data based on a pre-configured traffic data collection device includes:

[0011] Collecting traffic data of all target objects based on pre-configured traffic data collection devices;

[0012] Invoking an image analysis and machine learning algorithm to filter and mark the autonomous driving car from the traffic data to obtain first traffic data; and invoking a preset programming language to distinguish the target object as one of a plurality of traffic participants to obtain second traffic data;

[0013] The first traffic data and the second traffic data are processed based on the time period data combing rule to obtain mixed traffic flow real vehicle data.

[0014] Optionally, the preconfigured traffic data collection device includes a mechanical laser radar and an industrial camera.

[0015] Optionally, the preset test scenario library is obtained by finitely mapping, logically abstracting, and organizing and combining infinite driving scenarios in the real world.

[0016] Optionally, the step of screening a target test scenario and a test case corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data includes:

[0017] Obtaining traffic flow operation data obtained in the mixed traffic flow real vehicle data collection and processing;

[0018] Based on the traffic flow data, a preset test scenario library is screened to obtain target test scenarios and test cases corresponding to each target test scenario.

[0019] Optionally, the obtaining of vehicle motion data based on a pre-built road data collection vehicle includes:

[0020] Receive vehicle motion data in different weather conditions and / or different scenarios collected by a public road data collection vehicle; wherein the vehicle motion data includes vehicle bus data, Ethernet data and / or reference video.

[0021] Optionally, configuring the detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario according to the vehicle motion data includes:

[0022] associating metadata corresponding to the vehicle motion data;

[0023] Perform multi-source data fusion playback based on different acquisition methods in the acquired data;

[0024] Receiving a marking operation during the multi-source data fusion playback process to obtain processed collected data;

[0025] The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured based on the scenario-based refined conditions; wherein the refined conditions at least characterize the frequency of occurrence of the scenario, the safety consequences caused and / or the importance.

[0026] Optionally, the pre-built road data collection vehicle includes at least a plurality of laser radars installed outside the vehicle body, at least one intelligent visual sensor inside the vehicle, at least one GPS and inertial navigation combination test sensor, at least one industrial computer and / or at least one data acquisition and storage industrial computer.

[0027] A second aspect of the present application provides a device for constructing an autonomous driving test scenario based on a solid-state laser radar system, comprising:

[0028] An acquisition module, used for acquiring real vehicle data of mixed traffic flow based on a pre-configured traffic data acquisition device;

[0029] A construction module is used to select a target test scenario and a test case corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data;

[0030] Obtain vehicle motion data based on a pre-built road data collection vehicle;

[0031] The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured according to the vehicle motion data.

[0032] Optionally, the acquisition module is further used to:

[0033] Collecting traffic data of all target objects based on pre-configured traffic data collection devices;

[0034] Invoking an image analysis and machine learning algorithm to filter and mark the autonomous driving car from the traffic data to obtain first traffic data; and invoking a preset programming language to distinguish the target object as one of a plurality of traffic participants to obtain second traffic data;

[0035] The first traffic data and the second traffic data are processed based on the time period data combing rule to obtain mixed traffic flow real vehicle data.

[0036] Optionally, the preconfigured traffic data collection device includes a mechanical laser radar and an industrial camera.

[0037] Optionally, the preset test scenario library is obtained by finitely mapping, logically abstracting, and organizing and combining infinite driving scenarios in the real world.

[0038] Optionally, the building block is further used to:

[0039] Obtaining traffic flow operation data obtained in the mixed traffic flow real vehicle data collection and processing;

[0040] Based on the traffic flow data, a preset test scenario library is screened to obtain target test scenarios and test cases corresponding to each target test scenario.

[0041] Optionally, the building block is further used to:

[0042] Receive vehicle motion data in different weather conditions and / or different scenarios collected by a public road data collection vehicle; wherein the vehicle motion data includes vehicle bus data, Ethernet data and / or reference video.

[0043] Optionally, the building block is further used to:

[0044] associating metadata corresponding to the vehicle motion data;

[0045] Perform multi-source data fusion playback based on different acquisition methods in the acquired data;

[0046] Receiving a marking operation during the multi-source data fusion playback process to obtain processed collected data;

[0047] The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured based on the scenario-based refined conditions; wherein the refined conditions at least characterize the frequency of occurrence of the scenario, the safety consequences caused and / or the importance.

[0048] Optionally, the pre-built road data collection vehicle includes at least a plurality of laser radars installed outside the vehicle body, at least one intelligent visual sensor inside the vehicle, at least one GPS and inertial navigation combination test sensor, at least one industrial computer and / or at least one data acquisition and storage industrial computer.

[0049] The present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned construction method when executing the program.

[0050] The present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of any one of the above-mentioned methods for prototyping a BMC chip based on a PCIE interface are implemented.

[0051] The present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned BMC chip prototype verification methods based on a PCIE interface.

[0052] The technical solution provided by this application may have the following beneficial effects:

[0053] The embodiment of the present application first obtains mixed traffic flow real vehicle data based on a pre-configured traffic data collection device; selects target test scenarios and test cases corresponding to each target test scenario in a preset test scenario library based on the mixed traffic flow real vehicle data; then obtains vehicle motion data based on a pre-built road data collection vehicle; and finally configures detailed information of the target test scenario and detailed information of the test cases corresponding to each target test scenario based on the vehicle motion data. The scenario constructed by this method can make the test process more realistic and effective, which is conducive to verifying the reliability and environmental adaptability of the radar system when facing complex and diverse real traffic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The above and other objects, features and advantages of the present application will become more apparent by describing in more detail exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0055] Figure 1 is a flow chart of a method for constructing an autonomous driving test scenario based on a solid-state lidar system shown in an embodiment of the present application;

[0056] Figure 2 It is a mapping diagram of a test scenario library provided in an embodiment of the present application;

[0057] Figure 3 This is one of the test scenario examples provided in the embodiment of the present application;

[0058] Figure 4 This is a second test scenario example provided in an embodiment of the present application;

[0059] Figure 5 This is a third test scenario example provided in an embodiment of the present application;

[0060] Figure 6 This is a fourth example of a test scenario provided in an embodiment of the present application;

[0061] Figure 7 This is a fifth test scenario illustration provided in an embodiment of the present application;

[0062] Figure 8 This is a sixth example of a test scenario provided in an embodiment of the present application;

[0063] Fig. 9 This is a seventh example of a test scenario provided in an embodiment of the present application;

[0064] Fig.10This is an eighth example of a test scenario provided in an embodiment of the present application;

[0065] Fig.11 This is a ninth example of a test scenario provided in an embodiment of the present application;

[0066] Fig.12 This is a tenth example of a test scenario provided in an embodiment of the present application;

[0067] Fig.13 This is an eleventh test scenario illustration provided by an embodiment of the present application;

[0068] Fig.14 is a structural schematic diagram of a device for constructing an autonomous driving test scenario based on a solid-state laser radar system provided in an embodiment of the present application;

[0069] Fig.15 It is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0070] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0071] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0072] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0073] The embodiment of the present application aims to construct a test scenario for a solid-state LiDAR system during autonomous driving, through which the autonomous driving function performance of an autonomous vehicle can be tested, verified or evaluated. Since developers need to obtain expected feedback conclusions through test evaluation, the application test scenario must have a clear test purpose, which may include expected behavior, performance requirements, etc. Testers can verify and evaluate the vehicle through the performance of the tested vehicle in the test scenario, thereby feeding back the expected indicators or conclusions to the developer.

[0074] See also Figure 1 , Figure 1 It is a flow chart of a method for constructing an autonomous driving test scenario based on a solid-state lidar system disclosed in an embodiment of the present application.

[0075] The method for constructing an autonomous driving test scenario based on a solid-state laser radar system disclosed in an embodiment of the present application includes:

[0076] S100, obtaining mixed traffic flow real vehicle data based on a pre-configured traffic data collection device;

[0077] S200, selecting a target test scenario and a test case corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data;

[0078] S300, obtaining vehicle motion data based on a pre-built road data collection vehicle;

[0079] S400: configuring detailed information of the target test scenario and detailed information of the test case corresponding to each target test scenario according to the vehicle motion data.

[0080] The technical solution adopted in the embodiment of the present application is a method for constructing an autonomous driving test scenario based on a solid-state lidar system, including real vehicle data collection and processing of mixed traffic flow, test scenario selection, and public road data collection and processing.

[0081] Among them, mixed traffic flow real-vehicle data collection and processing: used to collect traffic flow information on urban public roads, and pre-process the collected data, classify the target object types, record the operating data of each target object, and divide the collected data according to different time periods.

[0082] Among them, test scenario selection: based on traffic flow operation data, preliminarily screen test scenarios and test cases from the test scenario library for intelligent connected vehicles.

[0083] Among them, public road data collection and processing: used to build public road data collection equipment, collect data and process it, and refine the test scenarios based on the processed data.

[0084] In the embodiment of the present application, a traffic data collection device is pre-configured.

[0085] In some embodiments, the preconfigured traffic data collection device includes a mechanical laser radar and an industrial camera.

[0086] In the application, a HESAI 64-line mechanical laser radar and a Hikvision industrial camera can form a traffic data collection device to collect the speed, position, headway, size, etc. of all targets. It is mainly used to collect data of self-driving cars and manually driven cars in mixed traffic flows on urban public roads as real vehicle data of mixed traffic flows.

[0087] In the application, taking into account the operation scale and brand distribution of autonomous driving taxis Robotaxi in different autonomous driving demonstration operation areas, as well as the driving style of the drivers, Changsha and Shanghai are selected as the collection locations. The environmental information, collection time, etc. can be shown in Table 1 below.

[0088] Table 1

[0089]

[0090] In the embodiment of the present application, the image analysis and machine learning methods are combined to fuse the point cloud data of the lidar with the 3-channel color image information of the camera. Taking into account the biggest difference between self-driving cars and traditional manually driven cars - the lidar and bracket installed on the top of the self-driving car, the ROI information is extracted to calculate the HOG features and train the SVM classifier, so as to screen out the self-driving cars and mark them.

[0091] The acquired data is processed using the C++ programming language, and the targets are divided into self-driving cars, manually driven cars, pedestrians, two-wheeled vehicles, etc. The size of each target and its position information at different times are recorded; the driving direction is defined as the X-axis and the direction perpendicular to the lane is defined as the Y-axis, and the speed, acceleration, front / rear vehicle type and head-to-head distance of each target in the X and Y directions are calculated.

[0092] Data sorting by time period:

[0093] The biggest difference between high-level self-driving cars and manually driven cars lies in the driving subject. Manually driven cars require the driver to receive road information and make judgments, and input actions to the vehicle through the steering wheel, brake / accelerator pedal, etc.; while self-driving cars rely on the perception-control system on board, and do not require the driver's action input to complete the driving task. When the car issues a takeover request or the driver predicts that there will be danger, the driving right is handed over to the driver. During the on-site collection, it was found that when the Robotaxi in Changsha was running, the driver rarely took over the vehicle, and most of the driving tasks were completed by the automatic driving function; when the Robotaxi in Shanghai was running, the driver's hands were mostly placed on the steering wheel, making it difficult to distinguish the driving subject. In order to study the actual operating characteristics of self-driving cars, the collected data of the Meixi Lake section in Changsha was used as a sample.

[0094] Taking into account that the operation time of the autonomous Robotaxi vehicle avoids the morning / evening peak hours, we selected the data collected during the morning and evening peak hours on weekdays and the midday peak hours on non-working days in Changsha for analysis, and divided the collected targets into manually driven cars, medium-sized cars and Robotaxi vehicles.

[0095] According to the provisions of JTGB01-2014 Highway Engineering Technical Standards, as shown in Table 2:

[0096] Table 2

[0097] Representative car models Vehicle conversion factor illustrate Passenger car 1 Passenger cars with ≤19 seats and trucks with a load capacity of ≤2t Mid-size car 1.5 Passenger cars with more than 19 seats and trucks with a load capacity of 2t and less than 7t Large car 2.5 Trucks with a load capacity of 7t or less and a load capacity of ≤20t Car train 4 Trucks with a load capacity of >20t

[0098] The cars and mid-sized cars in the traffic flow data are converted into standard cars. The specific collection time and data are shown in Table 3:

[0099] Table 3

[0100]

[0101] The infinite driving scenarios in the real world are mapped, logically abstracted, and organized to form a test scenario library for intelligent connected vehicles, such as Figure 2 As shown in Table 4:

[0102] Table 4

[0103] Serial number Test Project Highways and expressways Urban roads Other roads 1 Speed ​​limit signs √ √ √ 2 Lane lines √ √ √ 3 Stop and yield sign marking √ √ 4 Traffic lights at intersections √ √ 5 Direction indicator lights √ √ 6 Expressway lane signal lights √ 7 tunnel √ √ √ 8 Roundabout √ √ 9 Ramp √ √ 10 Toll booths √ √ 11 Conflict between vehicles going straight at an intersection without traffic lights √ √ 12 Conflict between vehicles turning right at an intersection without traffic lights √ 13 Conflict between vehicles turning left at an intersection without traffic lights √ 14 Conventional obstacles √ √ √ 15 Stationary vehicles occupying part of the lane √ √ √ 16 Pedestrians crossing the crosswalk √ √ 17 Pedestrians walking along the road √ √ 18 Bicycle riding along the road √ √ 19 Motorcycle riding along the road √ √ 20 Pedestrians crossing the road √ √ √ 21 Cycling across the road √ √ 22 The vehicle ahead cuts in √ √ √ 23 The vehicle ahead cuts out √ √ √ 24 Oncoming vehicles use the same lane √ √ 25 Vehicle ahead stop-go √ √ √ 26 There is a stationary vehicle ahead of the following vehicle √ √ √ 27 The vehicle ahead brakes suddenly √ √ √ 28 Fixed-point parking √ √ 29 <![CDATA[Dynamic driving task takeover * > √ √ √

[0104] For example, based on the traffic flow operation data obtained in the mixed traffic flow real vehicle data collection and processing, 9 test scenarios and 15 test cases were preliminarily screened out as shown in the three columns of "Scene Description", "Lighting" and "Legend" in Table 5 below:

[0105] Table 5

[0106]

[0107] Public road data collection and processing: Build vehicles with public road data collection equipment, collect and process data, and refine test scenarios based on the processed data.

[0108] Vehicles for building open road data collection equipment: Open road data collection equipment is portable and can be installed quickly and easily, including the installation of sensors and data recording equipment inside and outside the vehicle. To ensure the installation effect, the test equipment generally has an adaptive mounting bracket. Under the premise of meeting the sensor performance and location requirements, implement the sensor installation and vehicle modification plan, and rely on the original vehicle space for concealed installation as much as possible to avoid major changes and damage to the original vehicle body structure. At the same time, the equipment installation needs to consider the routine debugging and maintenance of the equipment. Specifically, this application uses the Cadillac CT6 as the survey car, mainly installing 6 Ibeo LUX 4-line laser radars outside the car body, 1 intelligent visual sensor inside the car, 1 RT3000 GPS and inertial navigation combination test sensor, and 1 digital acquisition display screen; 1 IBEO industrial computer in the trunk of the car, 1 digital acquisition storage industrial computer (Carbox), cooling fan, power switch and several connecting harnesses.

[0109] IBEO LiDAR Installation:

[0110] Six IBEO LUX 4-line laser radars, each with a horizontal field of view of 110°, need to be installed exposed on the front, rear and sides of the vehicle body. The installation height is generally 30-70cm, and the installation height of the side laser radars should match it. Under the premise of meeting the technical parameter requirements of the sensor, the original vehicle body parts should not be changed or damaged too much during the installation process.

[0111] Left front and right front laser radar:

[0112] After comprehensive consideration of the external shape and structure of the left and right front headlights and the interior space of the vehicle, the installation method of the left and right front laser radars is designed as follows: remove the guard plates on both sides of the bumper as connecting holes, make a profile bracket of appropriate size to extend to the outside of the bumper, fix one end of the profile to the bar of the front bumper with screws, and fix the laser radar mounting backplate at the other end to install the laser radar.

[0113] Front LiDAR:

[0114] After evaluating the installation position, use screws to fix the installation backplate to the center of the front grille of the vehicle, and fix the lidar on the backplate for exposed installation.

[0115] Laser radar directly behind:

[0116] Without damaging the car body, remove the original rear bumper bar and keep it separately, replace it with a new one, fix the laser radar installation back plate to the rear bumper with screws, open a hole at the target position in the middle of the rear bumper, and install the laser radar externally and fix it on the back plate.

[0117] Left rear and right rear laser radar:

[0118] Remove the rear bumper, use the metal plates at the upper and lower edges of the left and right rear vents of the vehicle body as fixed inner layers, add two metal plates of appropriate size inside and outside the interlayer and fix them with screws, install the LiDAR backplane on the added metal plates, open the target positions at the left and right rear of the rear bumper, install the LiDAR exposed, and fix it on the backplane. The figure shows the installation diagram of the left rear radar, and the installation of the right rear radar follows the same method.

[0119] Smart Camera Installation:

[0120] The intelligent vision camera is installed in the center of the vehicle's front windshield to collect information such as the position and speed of traffic participants in front of the vehicle, including pedestrians, vehicles and other traffic participants. The camera comes with a fixed bracket, which is fixed to the front windshield with transparent double-sided tape to ensure that the camera bracket and the glass plane are adhered without affecting the driver's line of sight.

[0121] HD audio and video installation:

[0122] The high-definition video uses Logitech cameras, which are mainly used to monitor and record scenes inside and outside the car, as well as driver operations. The camera has a mounting bracket, which is fixed with transparent double-sided tape through the bonding surface between the mounting bracket and the vehicle glass.

[0123] Inertial GPS combined test system installation:

[0124] The inertial GPS combination test system can be installed between the sunroof and the rear floor of the vehicle through the built-in telescopic rod. The bracket is retractable and can be flexibly supported. The RT3000 is fixed to the telescopic rod through a fixed bracket and screws, and the line of sight X, Y and Z axes are fixed to make the inertial GPS combination test system equipment and the vehicle rigidly connected.

[0125] Data acquisition industrial computer and its electrical device:

[0126] The data acquisition industrial computer is a terminal device for data acquisition and storage. The base of the device can be fixed with Velcro and pasted on the trunk. The wiring harness is wired according to the equipment harness connection diagram. The cables of the Reference system sensors and other equipment are arranged and fixed. The beginning and end of the wiring harness are clearly marked. Relying on the original body structure, the original vehicle pipeline path is used as much as possible for concealed installation. To avoid damage to the body, the roof innovusion radar harness enters the car through the gap of the left rear door window glass and is arranged on the trunk control unit through the interior space. The gap of the left rear door window glass is sealed with tape, and opening and closing are avoided as much as possible.

[0127] Vehicle motion data is collected in different weather conditions (including temperature, humidity, rainfall, light and other environmental conditions) and scenarios (pedestrians crossing, longitudinally following two-wheeled vehicles) on autonomous driving demonstration sections in Tianjin, Shanghai and other places, including vehicle CAN / CANFD bus data, Ethernet data, reference video data, etc.

[0128] The data collected by the public road data collection vehicles are uploaded to the cloud via storage media;

[0129] The cloud associates the corresponding data through project name, location information, etc., and automatically filters and cleans the data during the process;

[0130] Joint playback of data collected by camera / truth sensor / bus, and fusion playback of multi-source data;

[0131] During the vehicle-side test, the tags can be reviewed, positioned, and adjusted; during playback, manual / automatic annotations can be added in the cloud;

[0132] During playback, the vehicle position when the tag event occurs is displayed on the interface, and click jump is supported, so that manufacturers can repeat the test and iterate the optimization at this position;

[0133] The processed data can be sliced ​​by sensor / channel and exported into csv format for subsequent MATLAB and Python analysis.

[0134] Manually sort out the data to filter the data based on frequency of occurrence, importance, etc.

[0135] Refine the test scenario based on the processed data:

[0136] According to the collected data, the selected scenarios are refined by considering the frequency of occurrence, safety consequences, importance, etc., and detailed information such as the test vehicle speed in the actual vehicle test is determined, as shown in the "Own vehicle behavior" and "Target vehicle behavior" columns in Table 5.

[0137] The embodiment of the present application constructs an autonomous driving test scenario for a solid-state laser radar system, collects data from open road traffic flows, and uses this data to preliminarily screen test scenarios and use cases. Subsequently, a set of open road data collection equipment is built to collect vehicle motion data in different weather conditions and scenarios, in order to obtain richer test scenario elements and refine the screened scenarios. The scenario constructed by this method can make the test process more realistic and effective, and is conducive to verifying the reliability and environmental adaptability of the radar system in the face of complex and diverse real traffic environments.

[0138] Reference Fig.14 A schematic diagram of the structure of a device 20 for constructing an autonomous driving test scenario based on a solid-state laser radar system provided in an embodiment of the present application is shown, wherein the device comprises:

[0139] An acquisition module 201 is used to obtain mixed traffic flow real vehicle data based on a pre-configured traffic data acquisition device;

[0140] A construction module 202 is used to select a target test scenario and a test case corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data;

[0141] Obtain vehicle motion data based on a pre-built road data collection vehicle;

[0142] The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured according to the vehicle motion data.

[0143] Optionally, the acquisition module 201 is further used to:

[0144] Collecting traffic data of all target objects based on pre-configured traffic data collection devices;

[0145] Invoking an image analysis and machine learning algorithm to filter and mark the autonomous driving car from the traffic data to obtain first traffic data; and invoking a preset programming language to distinguish the target object as one of a plurality of traffic participants to obtain second traffic data;

[0146] The first traffic data and the second traffic data are processed based on the time period data combing rule to obtain mixed traffic flow real vehicle data.

[0147] Optionally, the preconfigured traffic data collection device includes a mechanical laser radar and an industrial camera.

[0148] Optionally, the preset test scenario library is obtained by finitely mapping, logically abstracting, and organizing and combining infinite driving scenarios in the real world.

[0149] Optionally, the construction module 202 is further used to:

[0150] Obtaining traffic flow operation data obtained in the mixed traffic flow real vehicle data collection and processing;

[0151] Based on the traffic flow data, a preset test scenario library is screened to obtain target test scenarios and test cases corresponding to each target test scenario.

[0152] Optionally, the construction module 202 is further used to:

[0153] Receive vehicle motion data in different weather conditions and / or different scenarios collected by a public road data collection vehicle; wherein the vehicle motion data includes vehicle bus data, Ethernet data and / or reference video.

[0154] Optionally, the construction module 202 is further used to:

[0155] associating metadata corresponding to the vehicle motion data;

[0156] Perform multi-source data fusion playback based on different acquisition methods in the acquired data;

[0157] Receiving a marking operation during the multi-source data fusion playback process to obtain processed collected data;

[0158] The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured based on the scenario-based refined conditions; wherein the refined conditions at least characterize the frequency of occurrence of the scenario, the safety consequences caused and / or the importance.

[0159] Optionally, the pre-built road data collection vehicle includes at least a plurality of laser radars installed outside the vehicle body, at least one intelligent visual sensor inside the vehicle, at least one GPS and inertial navigation combination test sensor, at least one industrial computer and / or at least one data acquisition and storage industrial computer.

[0160] Fig.15 An example of a physical structure diagram of an electronic device is shown in FIG. Fig.15 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute a method for constructing an autonomous driving test scenario based on a solid-state lidar system, the method comprising:

[0161] Acquire real vehicle data of mixed traffic flow based on pre-configured traffic data collection devices;

[0162] Filtering target test scenarios and test cases corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data;

[0163] Obtain vehicle motion data based on a pre-built road data collection vehicle;

[0164] The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured according to the vehicle motion data.

[0165] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0166] On the other hand, the present application further provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the above-mentioned method for constructing an autonomous driving test scenario based on a solid-state lidar system, which includes:

[0167] Acquire real vehicle data of mixed traffic flow based on pre-configured traffic data collection devices;

[0168] Filtering target test scenarios and test cases corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data;

[0169] Obtain vehicle motion data based on a pre-built road data collection vehicle;

[0170] The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured according to the vehicle motion data.

[0171] In another aspect, the present application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the above-mentioned method for constructing an autonomous driving test scenario based on a solid-state lidar system, the method comprising:

[0172] Acquire real vehicle data of mixed traffic flow based on pre-configured traffic data collection devices;

[0173] Filtering target test scenarios and test cases corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data;

[0174] Obtain vehicle motion data based on a pre-built road data collection vehicle;

[0175] The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured according to the vehicle motion data.

[0176] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0177] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for constructing an autonomous driving test scenario based on a solid-state laser radar system, characterized in that: include: Acquire real vehicle data of mixed traffic flow based on pre-configured traffic data collection devices; Filtering target test scenarios and test cases corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data; Obtain vehicle motion data based on a pre-built road data collection vehicle; The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured according to the vehicle motion data.

2. The method according to claim 1, characterized in that: The method of obtaining mixed traffic flow real vehicle data based on a pre-configured traffic data collection device includes: Collecting traffic data of all target objects based on pre-configured traffic data collection devices; Invoking an image analysis and machine learning algorithm to filter and mark the autonomous driving car from the traffic data to obtain first traffic data; and invoking a preset programming language to distinguish the target object as one of a plurality of traffic participants to obtain second traffic data; The first traffic data and the second traffic data are processed based on the time period data combing rule to obtain mixed traffic flow real vehicle data.

3. The method according to claim 2, characterized in that The pre-configured traffic data collection device includes a mechanical laser radar and an industrial camera.

4. The method according to claim 1, characterized in that The preset test scenario library is obtained by finitely mapping, logically abstracting, and organizing and combining infinite driving scenarios in the real world.

5. The method according to claim 4, characterized in that The step of screening target test scenarios and test cases corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data includes: Obtaining traffic flow operation data obtained in the mixed traffic flow real vehicle data collection and processing; Based on the traffic flow data, a preset test scenario library is screened to obtain target test scenarios and test cases corresponding to each target test scenario.

6. The method according to claim 1, characterized in that The method of obtaining vehicle motion data based on a pre-built road data collection vehicle includes: Receive vehicle motion data in different weather conditions and / or different scenarios collected by a public road data collection vehicle; wherein the vehicle motion data includes vehicle bus data, Ethernet data and / or reference video.

7. The method according to claim 6, characterized in that The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario configured according to the vehicle motion data include: associating metadata corresponding to the vehicle motion data; Perform multi-source data fusion playback based on different acquisition methods in the acquired data; Receiving a marking operation during the multi-source data fusion playback process to obtain processed collected data; The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured based on the scenario-based refined conditions; wherein the refined conditions at least characterize the frequency of occurrence of the scenario, the safety consequences caused and / or the importance.

8. The method according to claim 6, characterized in that The pre-built road data collection vehicle includes at least a plurality of laser radars installed outside the vehicle body, at least one intelligent visual sensor inside the vehicle, at least one GPS and inertial navigation combination test sensor, at least one industrial computer and / or at least one data acquisition and storage industrial computer.

9. A device for constructing an autonomous driving test scenario based on a solid-state laser radar system, characterized in that: include: An acquisition module, used for acquiring real vehicle data of mixed traffic flow based on a pre-configured traffic data acquisition device; A construction module is used to select a target test scenario and a test case corresponding to each target test scenario in a preset test scenario library according to the mixed traffic flow real vehicle data; Obtain vehicle motion data based on a pre-built road data collection vehicle; The detailed information of the target test scenario and the detailed information of the test case corresponding to each target test scenario are configured according to the vehicle motion data.

10. An electronic 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 construction method as described in claims 1-8 is implemented.

Images