Simulation test system and method for autonomous driving

Abstract

The application relates to the technical field of automatic driving, and provides an automatic driving simulation test system and method.The system comprises a scene simulation module, a map forwarding module and an automatic driving controller module, wherein the scene simulation module is used for analyzing a scene file to obtain initial map data and storing the initial map data, and the stored initial map data is shared with the map forwarding module; the map forwarding module is used for screening target map data from the initial map data based on vehicle information of a test vehicle, and assigning values to the target map data, and the data is sent to the automatic driving controller module in the form of middleware; and the automatic driving controller module is used for generating control parameters of the test vehicle based on the target map data and a configuration file, and sending the control parameters to the scene simulation module, and the scene simulation module controls the test vehicle based on the control parameters. The simulation test system provided by the application can improve the effectiveness of the test.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to an autonomous driving simulation testing system and method. Background Technology

[0002] The simulation testing environment for autonomous vehicles can simulate the geographical environment and road conditions of the real world, enabling simulated testing of autonomous vehicles under various road conditions. This helps developers test the navigation, positioning, path planning and other functions of autonomous driving systems, thereby improving the safety and reliability of autonomous driving.

[0003] Regarding the issue of map data acquisition during the research and development and testing phase of autonomous driving systems, OEMs currently rely on map providers for the necessary map data when developing intelligent driving functions. However, map providers sometimes do not provide Open Drive files for specific vehicle maps, but only provide map data interfaces. This leads to a mismatch between map simulation scenarios and map data for autonomous driving controllers, increasing the development and testing costs of autonomous driving technology. Summary of the Invention

[0004] In view of this, embodiments of this application provide a simulation testing system and method for autonomous driving to solve the problem in the prior art where the test effectiveness is not high due to the mismatch between the map simulation scene and the actual map data of the autonomous driving controller.

[0005] A first aspect of this application provides a simulation testing system for autonomous driving, comprising a scene simulation module, a map forwarding module, and an autonomous driving controller module; wherein...

[0006] The scene simulation module is used to parse the pre-stored scene file to obtain the initial map data, store the initial map data based on the data interface protocol of the autonomous driving controller module, and share the stored initial map data with the map forwarding module through shared memory; the scene file includes road environment information and test vehicle information, and the initial map data includes map feature information and test vehicle information.

[0007] The map forwarding module is used to filter target map data from the initial map data based on the vehicle information of the test vehicle, assign values ​​to the target map data based on the data interface protocol of the autonomous driving controller module, and send the configuration file corresponding to the map forwarding module and the assigned target map data to the autonomous driving controller module through middleware; the configuration file is used to make the simulation environment configuration of the scene simulation module consistent with that of the autonomous driving controller module.

[0008] The autonomous driving controller module generates control parameters for the test vehicle based on target map data and configuration files, and sends the control parameters to the scenario simulation module, which then controls the test vehicle based on the control parameters.

[0009] A second aspect of this application provides a simulation testing method for autonomous driving, applied to a map forwarding module in a simulation testing system. The simulation testing system further includes a scene simulation module and an autonomous driving controller module. The simulation testing method includes:

[0010] The initial map data stored by the scene simulation module is obtained. The initial map data is stored by the scene simulation module based on the data interface protocol of the autonomous driving controller module, and the initial map data is obtained by parsing the pre-stored scene file.

[0011] Target map data is obtained by filtering the initial map data based on the vehicle information of the test vehicle;

[0012] The target map data is assigned based on the data interface protocol of the autonomous driving controller module. The configuration file corresponding to the map forwarding module and the assigned target map data are sent to the autonomous driving controller module through middleware, so that the autonomous driving controller module can generate control parameters for the test vehicle based on the target map data and configuration file.

[0013] A third aspect of this application 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 executes the computer program to implement the steps of the above-described method.

[0014] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0015] The beneficial effects of this application embodiment compared with the prior art are as follows: The simulation test system includes a scene simulation module, a map forwarding module, and an autonomous driving control module. The scene simulation module is used to parse pre-stored scene files to obtain initial map data. The scene files include road environment information and vehicle information of the test vehicle. The initial map data includes map feature information and vehicle information of the test vehicle. The initial map data is stored according to the data interface protocol of the autonomous driving controller module, so that the stored initial map data can be called through the data interface protocol of the autonomous driving controller module, saving computing power. The stored initial map data is shared with the map forwarding module through shared memory. The map forwarding module is used to filter target map data from the initial map data according to the vehicle information of the test vehicle, assign values ​​to the target map data according to the data interface protocol of the autonomous driving controller module, and forward the map. The configuration file corresponding to the module and the assigned target map data are sent to the autonomous driving controller module via middleware. The configuration file ensures consistency between the scenario simulation module and the autonomous driving controller module, enabling the target map data to be sent to the autonomous driving controller module via the map forwarding module. This allows the autonomous driving controller module to test the autonomous driving algorithm based on the target map data. The autonomous driving module generates control parameters for the test vehicle based on the target map data and the configuration file, and sends these parameters to the scenario simulation module. The scenario simulation module then controls the test vehicle according to these parameters, thus enabling the visibility of autonomous driving applications on the test vehicle. Furthermore, the autonomous driving controller module obtains map data from an external data source (i.e., scenario files) to execute the autonomous driving algorithm, ensuring consistency of map information between the scenario simulation module and the autonomous driving controller module, and improving the accuracy and effectiveness of simulation testing. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of an autonomous driving simulation test system provided in an embodiment of this application;

[0018] Figure 2 This is a schematic diagram of another autonomous driving simulation test system provided in an embodiment of this application;

[0019] Figure 3 This is a schematic diagram of the autonomous driving controller module receiving target map data according to an embodiment of this application;

[0020] Figure 4 This is a schematic diagram of a decision control submodule provided in an embodiment of this application;

[0021] Figure 5 This is a schematic diagram illustrating data transmission between the time submodule and the visualization submodule provided in an embodiment of this application;

[0022] Figure 6 This is a flowchart illustrating a simulation testing method for autonomous driving provided in an embodiment of this application;

[0023] Figure 7 This is a flowchart illustrating another simulation testing method provided in the embodiments of this application;

[0024] Figure 8 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0025] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0026] The following will describe in detail, with reference to the accompanying drawings, an autonomous driving simulation test system and method according to an embodiment of this application.

[0027] Figure 1 This is a schematic diagram of an autonomous driving simulation test system provided in an embodiment of this application. Figure 1 As shown, the simulation test system 100 includes a scene simulation module 101, a map forwarding module 102, and an autonomous driving controller module 103, wherein:

[0028] The scene simulation module 101 is used to parse the pre-stored scene file to obtain the initial map data, store the initial map data based on the data interface protocol of the autonomous driving controller module, and share the stored initial map data with the map forwarding module through shared memory.

[0029] The scene file includes road environment information and test vehicle information, while the initial map data includes map feature information and test vehicle information.

[0030] The map forwarding module 102 is used to filter target map data from the initial map data based on the vehicle information of the test vehicle, assign values ​​to the target map data based on the data interface protocol of the autonomous driving controller module, and send the configuration file corresponding to the map forwarding module and the assigned target map data to the autonomous driving controller module through middleware; wherein the configuration file is used to make the configuration of the scene simulation module consistent with that of the autonomous driving controller module.

[0031] The autonomous driving controller module 103 is used to generate control parameters for the test vehicle based on the target map data and configuration file, and send the control parameters to the scene simulation module, which then controls the test vehicle based on the control parameters.

[0032] Specifically, the scenario simulation module can process simulation environment map data and provide the processed simulation environment map data to the autonomous driving controller module. The autonomous driving controller module stores autonomous driving algorithms to ensure that the simulation environment of the scenario simulation module is consistent with the environment in which the autonomous driving algorithm of the test vehicle is applied, thereby improving the effectiveness of the simulation test system.

[0033] The scene simulation module parses pre-stored scene files to obtain initial map data. The scene files include road environment information and test vehicle information. Road environment information may include road geometry, road hierarchy, lane line geometry, lane line color and type, traffic lights, various signs, obstacles, etc. Vehicle information may include the test vehicle's motion status, location information, vehicle model, etc. Thus, the scene simulation module can construct initial map data based on the scene files. Furthermore, map element information may include road environment information, such as road geometry, road hierarchy, lane line geometry, lane line color and type, traffic lights, various signs, obstacles, etc.

[0034] The method for parsing the pre-stored scene file can be through searching and segmentation. Different data in the scene file are parsed using corresponding parsing functions. The different data include lane information, road boundaries, lane line information, traffic light information, etc. The parsing functions can include parsing functions for line segments, parsing functions for boundaries, parsing functions for signals, parsing functions for lanes, etc.

[0035] In addition, the scene simulation module stores the parsed initial map data according to the data interface protocol defined by the autonomous driving controller module. This ensures that the initial map data stored by the scene simulation module is consistent with the data interface protocol of the autonomous driving controller module. When calling the initial map data, the module can directly read the initial map data that is consistent with the data interface protocol of the autonomous driving controller module, thereby improving the utilization efficiency of the initial map data.

[0036] Furthermore, by sharing memory, the stored initial map data is shared with the map forwarding module. This shared memory approach allows the map forwarding module to obtain the stored initial map data in real time without delay. It also reduces the latency and overhead of data exchange, thereby improving the real-time performance and accuracy of the simulation test system.

[0037] In addition, after reading the initial map data through shared memory, the map forwarding module filters the initial map data based on the vehicle information of the test vehicle to obtain the target map data. This ensures that the target map data filtered by the map forwarding module can accurately and timely match the autonomous driving task. The map forwarding module can be developed and compiled in a Linux environment.

[0038] In addition, by assigning values ​​to the target map data according to the data interface protocol of the autonomous driving controller module, the target map data can be filled into a specific data structure or object and the corresponding data fields can be assigned values. This ensures that the transfer of target map data between different components conforms to the data interface protocol of the autonomous driving controller module, thus guaranteeing the consistency and stability of the simulation test system.

[0039] Furthermore, before assigning values ​​to the target map data according to the data interface protocol of the autonomous driving controller module, the target map data can be processed twice. This second processing can be to convert the format of the target map data into a data format that matches the map forwarding module, or to aggregate the same map feature information to reduce data redundancy and improve processing efficiency. It can also convert the coordinates of each map feature information in the target map data into the reference coordinate system corresponding to the autonomous driving controller module, so that the autonomous driving controller module can generate test parameters for controlling the test vehicle based on the target map data.

[0040] Furthermore, the configuration file corresponding to the map forwarding module and the assigned target map data are sent to the autonomous driving controller module via middleware. The middleware, acting as the communication hub within the simulation testing system, is responsible for data encapsulation, transformation, and routing. It allows modules to call functions or services located on different servers, enabling remote calls and integration, as well as asynchronous message transmission and communication in a distributed system. The configuration file ensures consistency between the scene simulation module and the autonomous driving controller module, allowing data transferred between them via the map forwarding module to be directly applied without reconfiguration, thus improving the efficiency of the simulation testing system.

[0041] Specifically, after receiving the target map data and configuration file sent by the map forwarding module, the autonomous driving controller module generates control parameters for the test vehicle based on the target map data and configuration file, combined with built-in algorithms and logic. The built-in algorithms can be the autonomous driving algorithm to be tested, and the control parameters can include the test vehicle's steering angle, throttle opening, brake opening, gear shifting commands, etc. The control parameters can determine the behavior of the test vehicle in the simulation environment. The autonomous driving control module can be developed and compiled in a Linux environment.

[0042] In addition, the autonomous driving controller module sends control parameters to the scene simulation module in real time. The scene simulation module then controls the test vehicle based on the control parameters to achieve autonomous driving simulation testing. Specifically, the control parameters can be sent to the scene simulation module through the map forwarding module. That is, after generating the control parameters, the autonomous driving controller module sends the control parameters to the map forwarding module. After receiving the control parameters, the map forwarding module sends them to the scene simulation module, enabling the scene simulation module to visualize the autonomous driving mode based on the control parameters.

[0043] According to the technical solution provided in this application, map data injection is achieved by autonomously generating map data through the scene simulation module, without the need for third-party map vendors to provide Open Drive files. Furthermore, the selected target map data and configuration files are sent to the autonomous driving controller module via middleware through map forwarding data, maintaining the matching between the simulation scene in the simulation test system and the map data in the autonomous driving controller module. This improves the development efficiency and maturity of the autonomous driving controller module and enables the simulation of various complex and extreme map conditions in the virtual environment. This enhances the coverage, diversity, and effectiveness of simulation testing for the autonomous driving controller module, avoiding the problem of low test effectiveness caused by the mismatch between the simulation scene and the map information of the autonomous driving controller module when the map vendor fails to provide Open Drive files.

[0044] In some embodiments, the scene simulation module includes a parsing submodule, which is used for:

[0045] Convert scene files into lightweight text files;

[0046] Vehicle information of the test vehicle is extracted from the text file, including the test vehicle's motion status and position information;

[0047] The text file is parsed using preset parsing functions corresponding to different categories of map elements to obtain map element information, and the map element information is then fused with vehicle information to obtain initial map data.

[0048] like Figure 2 As shown, the scene simulation module 101 includes a parsing submodule 1011. The parsing submodule can receive scene files pre-stored externally or internally to the scene simulation module. These scene files may include various map feature information, and the data volume of these files is typically large and not easily used directly for simulation scene testing and construction. The parsing submodule can convert the scene files into lightweight text files. For example, if the initial format of the scene file is plain text (Extensible Markup Language, XML), the parsing submodule can convert its format into a lightweight data exchange (JavaScript Object Notation, JavaScript, JSON) format. This makes the text file structure clear, the format uniform, and allows for the storage of map feature information of the simulation scene in a concise form, thereby reducing the complexity of processing scene files and the consumption of parsing resources.

[0049] Map element information may include lane information, lane line information, road boundaries, traffic lights, and other information.

[0050] Furthermore, the initial map data can be packaged into a database for storage, so that it can be directly accessed when performing autonomous driving test tasks later.

[0051] Furthermore, the initial map data includes target map data and other data required to execute the current autonomous driving test task. The other data represents data that is not needed to execute the current autonomous driving test task and also needs to be stored again according to the data interface protocol corresponding to the autonomous driving controller module.

[0052] In addition, after obtaining the lightweight text file, the parsing submodule can further parse the contents of the text file to determine the vehicle information corresponding to the test vehicle. The vehicle information includes motion state and position information. The motion state can include the speed, acceleration, and direction of travel of the test vehicle.

[0053] In addition, the parsing submodule can use search and segmentation to perform pre-defined parsing functions corresponding to different categories of map feature information in the scene file, thereby parsing and extracting map feature information from the text file, and thus obtaining map feature information. Among them, map features are the infrastructure and elements that constitute the traffic environment, such as bridges, roads, tunnels, traffic signs and traffic lights.

[0054] The preset parsing functions may include Line() for parsing line segments, Boundary() for parsing boundaries, Signal() for parsing signals, Lane() for parsing lanes, and may also include type parsing functions for parsing road types, signal parsing functions for parsing traffic lights, etc.

[0055] Furthermore, the parsing submodule fuses the extracted map feature information with the previously acquired test vehicle information to construct initial map data that includes both dynamic vehicle behavior and static environmental information. Through this fusion process, the simulation testing system can more accurately simulate the driving conditions of test vehicles in real-world road environments, including interactions between vehicles and between vehicles and the environment, thereby improving the realism of the simulation testing system.

[0056] According to the technical solution provided in the embodiments of this application, the parsing submodule can convert scene files into lightweight text files that are easy to process, reducing the computation time and amount of data processing. It can also extract map feature information and vehicle information from the text file and merge them into initial map data, providing more accurate and easier-to-filter data for the simulation test system and improving the testing efficiency of the simulation test system.

[0057] In some embodiments, the scene simulation module further includes a scene building submodule, used for:

[0058] Import map information from a preset path and build a simulation map scene based on the map information;

[0059] Based on the simulated map scene, generate road files;

[0060] In response to the operation of adding vehicle information and environmental information of the test vehicle to the road file, the vehicle information of the test vehicle is added to the road file, and based on the preset correspondence between different road types and environmental information, the corresponding environmental information is added to the roads of different road types in the road file to generate a scene file.

[0061] Send the scene file to the parsing submodule.

[0062] Specifically, such as Figure 2As shown, the scene simulation module 101 also includes a scene building submodule 1012. This submodule can construct a highly realistic simulation test scene by importing map information from a preset path. The map information includes the relative coordinates of each map element, terrain, landforms, and road network information. After obtaining the map information, it can be used to build a simulated map scene using graphics rendering technology and geographic information system principles. During the rendering process, the map information can be rendered based on factors such as terrain elevation, road width and material, building layout, and appearance.

[0063] In addition, based on the established simulation map scenario, the road network structure is further analyzed, and road files are generated based on the analyzed road network structure. The road files can include the geometry of the roads, road categories (such as highways, urban roads, rural roads, etc.), intersection information, etc. The road files can provide an accurate road basis for the autonomous driving algorithm of the test vehicle.

[0064] In addition, vehicle information for test vehicles can be specified in the road file. This means you can choose the type of test vehicle, such as a sedan or a truck, and you can also add information such as the size, maximum speed, maximum acceleration, initial position, and direction of travel of the test vehicle. The vehicle information is then integrated into the road file to facilitate simulation testing based on the vehicle information.

[0065] In addition, to simulate real-world driving environments under different road types, the scenario building submodule stores preset correspondences between different road types and environmental information. Road types can include urban roads, highways, rural roads, mountain roads, etc. The preset correspondences detail the matching relationships between different road types and environmental information. Different environmental information (e.g., sunny, rainy, foggy) affects the autonomous driving of the test vehicle. For example, the complex traffic flow on urban roads corresponds to different autonomous driving modes under different environmental conditions. For instance, in rainy, foggy, and snowy weather, a slower speed should be used compared to sunny weather to prevent skidding. Environmental information can also include light intensity, road surface slippage, etc. Therefore, the simulation testing system can add different environmental information based on the road type where the test vehicle is located and test the autonomous driving task under different environmental conditions, thereby enhancing the realism and complexity of the simulation testing system.

[0066] In addition, the scenario building submodule integrates all the above information to generate a scenario file, which includes a simulation map, road network, test vehicles and their driving paths, and environmental information, thus forming a complete simulation test environment. The integrated scenario file is then sent to the parsing submodule for further processing and analysis.

[0067] According to the technical solution provided in the embodiments of this application, sub-modules can be built according to the scenario to increase the realism and complexity of the simulation test system, and improve the accuracy and efficiency of simulation testing.

[0068] In some embodiments, the map forwarding module includes a configuration submodule and a data forwarding submodule;

[0069] The configuration submodule is used to configure the simulation environment, task management, communication interface and data interface to obtain configuration information, and to encapsulate the data interface and simulation environment to obtain encapsulated information; based on the configuration information and encapsulated information, a configuration file is generated and sent to the data forwarding submodule.

[0070] Specifically, such as Figure 2 As shown, the map forwarding module 102 includes a configuration submodule 1021 and a data forwarding submodule 1022. The configuration submodule and the data forwarding submodule can work together to ensure that the map forwarding module can efficiently and accurately transmit the corresponding initial map data, target map data or control parameters between the scene simulation module and the autonomous driving controller module.

[0071] In addition, the configuration submodule 1021 can set or adjust the environmental parameters of the simulation environment according to actual needs, such as map range, accuracy requirements, terrain features, environmental information, traffic flow, etc., so as to ensure that the simulation environment can closely match the actual application scenario.

[0072] The task management can also be configured to improve the efficiency of the map forwarding module in forwarding map data. The map data includes initial map data, target map data, and control parameters. The task management configuration can include the priority order of each task, the arrangement of the execution schedule, the definition of task execution conditions, and the retry strategy after task failure. Configuring the task management of the map forwarding module enables the simulation test system to transfer data between modules in an orderly and efficient manner.

[0073] The communication interface can also be configured. This means that the communication interface can be configured in detail according to the actual network environment and transmission requirements. The configuration content can include the selection of the interface type, the definition of the communication protocol, the allocation of port numbers, the agreement of data format, etc. Through the configuration of the communication interface, the smooth transmission of target map data or control parameters can be guaranteed.

[0074] The data interface can also be configured, including defining access permissions, implementing data encryption measures, and customizing data exchange protocols. By configuring the data interface, the efficiency of the map forwarding module in transmitting map data can be improved.

[0075] In addition, by encapsulating the data interface and simulation environment, the encapsulated information can be obtained, thereby ensuring the security and integrity of the map data. Through encapsulation, the target map data or control parameters can be prevented from being tampered with or leaked during transmission.

[0076] In addition, after configuring the simulation environment, task management, communication interface and data interface, they are integrated into configuration information. Based on the configuration information and the packaged file, a configuration file is generated. The configuration file records various configuration parameters and settings, and also includes execution scripts or instructions, which can guide the specific operation of the data forwarding submodule and make the configuration environment in the scenario simulation module consistent with the configuration environment in the autonomous driving controller module.

[0077] In addition, sending the configuration file to the data forwarding submodule 1022 can serve as the basis for executing the target map data or control parameter forwarding task.

[0078] Furthermore, the data forwarding submodule 1022 can forward target map data or control parameters. Specifically, it parses the received configuration file, extracts the necessary configuration parameters and execution commands, and forwards the target map data to the autonomous driving controller module and the control parameters to the scene simulation module according to the communication protocol and interface configured in the configuration file.

[0079] According to the technical solution provided in the embodiments of this application, the configuration environment in the scene simulation module and the configuration environment in the autonomous driving controller module can be made consistent through the collaborative work of the configuration submodule and the data forwarding submodule, and the configuration file generated by the configuration submodule, so that the process of forwarding target map data or control parameters remains smooth.

[0080] In some embodiments, the map forwarding module further includes a data reading submodule; the data reading submodule is used for:

[0081] Determine the storage path of the initial map data after storage, and read the initial map data after storage based on the storage path;

[0082] Target map data is obtained by filtering the initial map data based on the vehicle information of the test vehicle;

[0083] The target map data is assigned a value based on the data interface protocol of the autonomous driving controller module, and the assigned target map data is sent to the data forwarding submodule.

[0084] The data forwarding submodule is used to receive the assigned target map data and configuration file, and call the middleware to send the assigned target map data and configuration file to the autonomous driving controller module.

[0085] Specifically, such as Figure 2As shown, the map forwarding module 102 also includes a data reading submodule 1023. The data reading submodule 1023 can retrieve and read the initial map data from the storage medium, meeting the data transmission requirements of the simulation test system. After determining the storage path of the stored initial map data, the data reading submodule 1023 can open and read the file corresponding to the initial map data through a file operation interface, such as a file system application programming interface (API). The initial map data includes road networks, traffic signs, traffic light information, etc., which can support the autonomous driving algorithm in the autonomous driving controller module for environmental perception and decision-making of the test vehicle.

[0086] In addition, target map data can be obtained by filtering the initial map data based on the vehicle information of the test vehicle. The initial map data can be filtered based on the vehicle information of the test vehicle, such as location coordinates, driving direction, driving speed, vehicle type, etc. For example, when the simulation test scenario corresponding to the test vehicle is driving on a highway, relevant ramp information, traffic signs, service area information, etc. can be used as the key filtering objects of the target data.

[0087] In addition, the target map data is assigned values ​​according to the data interface protocol of the autonomous driving controller module, thereby ensuring the accuracy and readability of the target map data, improving the efficiency of the simulation test system, and sending the assigned target map data to the data forwarding submodule.

[0088] Furthermore, the target map data can be encrypted or compressed to improve the transmission efficiency and security of the target map data.

[0089] In addition, the data forwarding submodule 1022 is used to receive the target map data and related configuration files after being assigned by the data reading submodule 1023. The configuration files include configuration parameters for data transmission, configuration parameters for the simulation scene being built, etc. The data forwarding submodule 1022 can call one or more middleware to send the assigned target map data and configuration files to the autonomous driving controller module through the middleware.

[0090] The middleware acts as a bridge between the data forwarding submodule and the autonomous driving controller module. It is used for tasks such as routing target map data and handling errors. Through the middleware, the assigned target map data and configuration files can be sent to the autonomous driving controller module securely and efficiently, so that the autonomous driving controller module can make autonomous driving-related decisions and generate control parameters based on the target map data and configuration files.

[0091] According to the technical solution provided in the embodiments of this application, the map forwarding module, through the cooperation between the data reading submodule and the data forwarding submodule, realizes the data transmission between the initial map data and the target map data and sends the target map data to the autonomous driving controller module, providing detailed and accurate map data support for the simulation test tasks of the test vehicle in the autonomous driving controller module.

[0092] In some embodiments, the vehicle information of the test vehicle includes the location information of the test vehicle; obtaining target map data from the initial map data based on the vehicle information of the test vehicle includes:

[0093] The location information of the test vehicle is used as the navigation starting point, and the navigation endpoint corresponding to the autonomous driving test task is determined.

[0094] The navigation map path is determined based on the navigation start point and navigation end point. The navigation map path includes map element information existing in the path within a preset path range centered on the navigation path.

[0095] Based on the navigation map path, the initial map data is filtered to obtain the target map data.

[0096] Specifically, the vehicle information for the test vehicle includes its location information, which can be obtained through methods such as the Global Positioning System (GPS). The location information can include the test vehicle's latitude and longitude, altitude, and orientation, thereby enabling the accurate construction of a simulation scenario and allowing the autonomous driving controller module to generate accurate control parameters based on the actual situation, such as the location information.

[0097] In addition, the location information of the test vehicle is determined as the navigation starting point, and the navigation endpoint is determined according to the specific requirements of the autonomous driving test task. The navigation endpoint can be a preset simulation scene location, a simulated traffic accident scene, or a specific road intersection, etc. In this way, comprehensive control parameters generated by the autonomous driving controller module can be obtained to determine the performance of the autonomous driving algorithm in the autonomous driving controller module under different situations.

[0098] In addition, a navigation path can be determined based on the navigation start point and navigation end point according to the path planning algorithm, and a navigation map path can be determined based on the navigation path. The navigation map path includes map element information existing in the path within a preset path range centered on the navigation path. For example, map element information within 5 kilometers around the navigation path can be used as the navigation map path, and the map element information can be saved according to the actual road network, shape, size, etc. in the navigation map path.

[0099] Furthermore, multiple feasible paths can be determined based on the navigation start and end points, and multiple corresponding navigation map paths can be determined based on each feasible path. Different navigation map paths may include different road types, speed limit information, construction sections, and other factors, thereby increasing the comprehensiveness of autonomous driving test tasks.

[0100] In addition, target map data is filtered from the initial map data based on the navigation map route. The target map data may include road markings (such as lane lines, lane dividers, stop lines, etc.), traffic light status, pedestrian crossings, traffic signs (such as speed limits, no left turns, etc.), intersection layout, location and type of obstacles (such as construction fences, vehicles, pedestrians, etc.), and other environmental factors that may affect vehicle driving safety and efficiency.

[0101] According to the technical solution provided in the embodiments of this application, target map data is filtered from the initial map data by using navigation start point, navigation end point and navigation path information, making the target map data more accurate and comprehensive and reducing the amount of data, thereby improving the data transmission efficiency and processing efficiency in the simulation test system.

[0102] In some embodiments, the autonomous driving controller module includes a map component submodule, a data interface submodule, and a decision control submodule;

[0103] The map component submodule is used to receive Open Drive files of maps sent by third parties;

[0104] The data interface submodule is used to disconnect from the map component submodule when it is detected that the map component submodule has not received the Open drive file, and to receive the target map data and configuration file and send the target map data and configuration file to the decision control submodule.

[0105] The decision control submodule is used to generate control parameters for the test vehicle based on target map data and configuration files.

[0106] Specifically, such as Figure 2 As shown, the autonomous driving controller module 103 includes a map component submodule 1031, a data interface submodule 1032, and a control decision submodule 1033. These submodules can work together to achieve autonomous driving of the test vehicle.

[0107] In addition, the map component submodule 1031 is used to receive and process map data. It can receive map data files in Open Drive format sent by third parties. Open Drive is a map format that includes key data such as road network structure, lane information, traffic signs, and intersection layout, providing an accurate simulation testing environment for the test vehicle. By parsing the file containing map data, a digital map required for the test vehicle's operation can be constructed, providing a foundation for the decision control process in the decision control submodule.

[0108] like Figure 3 As shown, the data interface submodule is used to disconnect from the map component submodule when it detects that the map component submodule has not received the Open drive file. That is, the data interface submodule can detect the working status of the map component submodule. When it detects that the map component submodule has failed to receive the map's Open drive file as expected, the data interface submodule disconnects the direct connection with the map component submodule, receives the target map data and the corresponding configuration file through the backup data channel or directly, and sends the target map data and configuration file to the decision control submodule so that the autonomous driving controller module can execute the autonomous driving test task.

[0109] The map component submodule may fail to receive the map's Open drive file due to network latency, file corruption, or third-party service terminals.

[0110] In addition, the decision control submodule is used to generate control parameters for controlling the test vehicle to perform autonomous driving test tasks based on the target map data and configuration file, combined with the vehicle information of the test vehicle. The control parameters include acceleration, braking, steering and other parameters.

[0111] Furthermore, the decision control submodule can optimize decision logic through algorithmic models, such as machine learning models and deep learning models, to improve the safety, comfort, and efficiency of autonomous driving.

[0112] Furthermore, such as Figure 4 As shown, the decision control submodule may include a positioning and planning unit, a sensor fusion unit, and a decision control unit, thereby providing intelligent decision-making for autonomous driving test tasks. The positioning and planning unit can plan the vehicle position of the test vehicle in the target map data, and can integrate the physical layout of the road and various factors, such as traffic rules and road speed limits, to update the planned position information of the test vehicle in the autonomous driving test task in real time, and at the same time generate the real-time position information of the test vehicle when actually performing the autonomous driving test task;

[0113] The sensor fusion unit fuses the real-time location information and planned location information of the test vehicle based on the real-time location information and planned location information transmitted by the positioning plan. At the same time, it can also simulate the fused information received by other sensors of the test vehicle, such as sensor information received by LiDAR, cameras, etc., and fuse them to obtain fused information. It can also perform data synchronization, noise reduction, calibration, etc. on the real-time location information and planned location information to obtain fused real-time location information and planned location information synchronized with the simulation scene, and send the fused information to the decision control unit.

[0114] The decision control unit generates control parameters based on the planned path and fused information. The control parameters correspond to vehicle information and may include acceleration parameters, braking parameters, steering parameters, lane change parameters, and obstacle avoidance parameters.

[0115] According to the technical solution provided in the embodiments of this application, the autonomous driving controller module can generate control parameters for the test vehicle based on the target map data and configuration file through the map component submodule, data interface submodule, and decision control submodule, so that the test vehicle can accurately execute autonomous driving test tasks.

[0116] In some embodiments, the map forwarding module further includes a time submodule, and the scene simulation module further includes a visualization submodule;

[0117] The time submodule is used to: determine the first transmission time when the map forwarding module sends the current target map data to the autonomous driving controller module; determine the first reception time when the map forwarding module receives the control parameters; and send the first transmission time and the first reception time to the visualization submodule.

[0118] The visualization submodule is used to control the test vehicle based on the control parameters corresponding to the first reception time, on the target map data corresponding to the first transmission time.

[0119] Specifically, such as Figure 5 As shown, the map forwarding module 102 also includes a time submodule 1024. The time submodule can improve the real-time performance and accuracy of the simulation test system. The time submodule can synchronize the transmission process of initial map data, target map data and control parameters in the simulation test system. The scene simulation module 101 also includes a visualization submodule 1013. The visualization submodule can intuitively visualize the target map data and control parameters.

[0120] In addition, the time submodule 1024 determines the specific time point at which the map forwarding module sends the target map data to the autonomous driving controller module, i.e., the first sending time, and determines the first receiving time corresponding to the control parameters sent by the autonomous driving controller module to the map forwarding module. The first receiving time is used to guide the driving behavior of the test vehicle. The time submodule 1024 sends the first sending time and the first receiving time to the visualization submodule 1013 of the scene simulation module. The target map data is updated in real time. The first sending time corresponds to the first receiving time. That is, the control parameters received at the first receiving time are the control parameters of the target map data corresponding to the first sending time. Similarly, the second sending time corresponds to the second receiving time, and so on, with the nth sending time corresponding to the nth receiving time, until the test vehicle completes the autonomous driving test task. The time submodule will record and store all sending times and receiving times corresponding to the autonomous driving test task.

[0121] In addition, after the visualization submodule receives the first transmission time and the first reception time sent by the time submodule, it can control the driving of the test vehicle based on the control parameters corresponding to the first reception time in the target map data corresponding to the first transmission time.

[0122] The visualization submodule may include a dynamics control submodule, which is used to control the driving of the test vehicle in the visualized simulation scene according to control parameters.

[0123] Furthermore, the visualization submodule needs to load the target map data, the corresponding configuration file for the target map data, and the configuration sensor information.

[0124] According to the technical solution provided in the embodiments of this application, the timeline of autonomous driving test tasks is managed through the time submodule, and the target map data and corresponding control parameters are displayed intuitively according to the timeline through the visualization submodule. This helps developers to deeply understand the performance of autonomous driving test tasks in different scenarios in the autonomous driving controller module. Furthermore, through playback and comparative analysis, the autonomous driving algorithm in the autonomous driving controller module can be optimized, thereby improving the safety and reliability of the autonomous driving system and building an efficient and accurate platform for autonomous driving test tasks and verification.

[0125] In some embodiments, the map forwarding module further includes an update submodule, which is used for:

[0126] Receive update instructions sent by the scene simulation module. The update instructions include map data for updating the simulation scene.

[0127] Based on the map data in the update instruction, the latest target map data received by the map forwarding module is updated, so that the autonomous driving controller module can update the control parameters of the test vehicle based on the updated target map data.

[0128] Specifically, the map forwarding module also includes an update submodule. The update submodule receives update instructions sent by the scene simulation module. The update instructions are given by the scene simulation module based on test requirements and actual scene changes, thereby enabling the map forwarding module to update the current test environment synchronously. Based on the update instructions, the flexibility and practicality of the control parameters generated by the autonomous driving controller module can be better tested, and it can also be used to update autonomous driving test tasks.

[0129] In addition, after receiving the update command, the update submodule will merge the map data used to update the simulation scene with the latest target map data received by the map forwarding module to obtain the updated target map data.

[0130] The update process can include adding, deleting, modifying, and querying map element information, and may also involve complex operations such as adjusting the map coordinate system and optimizing road connectivity. This enables the map forwarding module to ensure that the autonomous driving controller module makes autonomous driving decisions based on the latest and most accurate target map data and generates corresponding control parameters.

[0131] In addition, the autonomous driving controller module recalculates and updates the control parameters of the test vehicle based on the updated target map data. These control parameters directly guide the test vehicle's trajectory, speed, acceleration, and other parameters. Simultaneously, the scenario simulation module visualizes the test vehicle based on the updated control parameters.

[0132] According to the technical solution provided in the embodiments of this application, the simulation testing system can continuously adapt to testing requirements and obtain more accurate and reliable test results corresponding to autonomous driving test tasks.

[0133] Figure 6 This is a flowchart illustrating a simulation testing method for autonomous driving provided in an embodiment of this application. The simulation testing method is applied to the map forwarding module in the simulation testing system described above. The simulation testing system also includes a scene simulation module and an autonomous driving controller module. The simulation testing method includes the following steps:

[0134] S601, Obtain the initial map data stored in the scene simulation module;

[0135] The initial map data stored is obtained by the scene simulation module based on the data interface protocol of the autonomous driving controller module. The scene file includes road environment information and vehicle information of the test vehicle. The initial map data obtained by parsing includes map feature information and vehicle information of the test vehicle.

[0136] It should be noted that the specific details of this step can be found in the relevant content of the simulation test system mentioned above, and will not be repeated here.

[0137] S602, the target map data is obtained by filtering the initial map data based on the vehicle information of the test vehicle;

[0138] It should be noted that the specific details of this step can be found in the relevant content of the simulation test system mentioned above, and will not be repeated here.

[0139] S603 assigns values ​​to the target map data based on the data interface protocol of the autonomous driving controller module, and sends the configuration file corresponding to the map forwarding module and the assigned target map data to the autonomous driving controller module through middleware, so that the autonomous driving controller module can generate control parameters for the test vehicle based on the target map data and configuration file.

[0140] It should be noted that the specific details of this step can be found in the relevant content of the simulation test system mentioned above, and will not be repeated here.

[0141] According to the technical solution provided in the embodiments of this application, it is possible to simulate various complex and extreme map conditions in a simulation environment. The map forwarding module transmits data between the scene simulation module and the autonomous driving controller module, ensuring the real-time performance and consistency of map data in the scene simulation module and the autonomous driving controller module.

[0142] Figure 7 This is a flowchart illustrating another simulation testing method provided in an embodiment of this application. Figure 7 As shown, the simulation test process includes:

[0143] According to the requirements of autonomous driving test tasks, a simulation map scenario (i.e., simulation scenario) is built through the scenario simulation module. The simulation scenario includes information such as road geometry type, road hierarchy structure, lane line geometry information, lane line color and line type, traffic lights, and various signs. Road files and visualization files are generated based on the simulation scenario.

[0144] In the scene simulation module, use dynamic scene editing software to add road files and visualization files. Then, add dynamic information based on the added road files and visualization files, such as the behavior of test vehicles, the actions of other vehicles, and information about obstacles. After completion, save it as a scene file, which can be run directly in the scene software.

[0145] Scene files can be converted into lightweight text files using open-source libraries. The motion status and location information of the test vehicle can be obtained from the text file. Through the processing of the text file (i.e. the lightweight scene file), the initial map data can be extracted and processed. The initial map data includes lane information, lane line information, road boundaries, traffic lights, and other information. The processed initial map data is then passed to the map forwarding module via shared memory.

[0146] The map forwarding module receives data from the map plugin (which reads initial map data via shared memory) through shared memory. After receiving the initial map data, the map forwarding module performs secondary processing to obtain the target map data, implements an interface protocol consistent with the autonomous driving controller module, and assigns values ​​to the target map data. After the assignment is completed, the map data processed by the map forwarding module and the map configuration information are published to the data communication framework of the autonomous driving controller module through middleware by configuring the communication port, task enablement, and interface data management of the map forwarding module. The data connection between the map forwarding module and the map component sub-module inside the autonomous driving controller module is then cut off. The data sent by the map forwarding module through the middleware is directly passed to subsequent modules for use.

[0147] The decision control module processes the target map data to generate horizontal and vertical control parameters, which are then output to the dynamic control submodule of the scene simulation module to control the motion of the test vehicle.

[0148] In addition, after generating the scene file, the scene simulation module loads the configuration file and scene file, configures the sensor information and activation status, and configures the power system.

[0149] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0150] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0151] Figure 8 This is a schematic diagram of the electronic device 8 provided in an embodiment of this application. For example... Figure 8As shown, the electronic device 8 of this embodiment includes a processor 801, a memory 802, and a computer program 803 stored in the memory 802 and executable on the processor 801. When the processor 801 executes the computer program 803, it implements the steps in the various method embodiments described above. Alternatively, when the processor 801 executes the computer program 803, it implements the functions of each module / unit in the various device embodiments described above.

[0152] Electronic device 8 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 8 may include, but is not limited to, processor 801 and memory 802. Those skilled in the art will understand that... Figure 8 This is merely an example of electronic device 8 and does not constitute a limitation on electronic device 8. It may include more or fewer components than shown, or different components.

[0153] The processor 801 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0154] The memory 802 can be an internal storage unit of the electronic device 8, such as a hard disk or RAM of the electronic device 8. The memory 802 can also be an external storage device of the electronic device 8, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the electronic device 8. The memory 802 can also include both internal and external storage units of the electronic device 8. The memory 802 is used to store computer programs and other programs and data required by the electronic device.

[0155] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0156] If integrated modules / units are implemented as software functional units and sold or used as independent products, they can be stored in a readable storage medium (e.g., a computer-readable storage medium). Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable storage medium may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0157] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An automated driving simulation test system, characterized by, The simulation test system comprises a scene simulation module, a map forwarding module and an automatic driving controller module; wherein The scene simulation module is configured to parse a pre-stored scene file to obtain initial map data, store the initial map data based on a data interface protocol of the automatic driving controller module, and share the stored initial map data with the map forwarding module through a shared memory; wherein the scene file comprises road environment information and vehicle information of a test vehicle, and the initial map data comprises map element information and vehicle information of the test vehicle; The map forwarding module comprises a data reading submodule and a data forwarding submodule; the data reading submodule is configured to determine a storage path of the stored initial map data, read the stored initial map data based on the storage path, filter target map data from the initial map data based on the vehicle information of the test vehicle, assign values to the target map data based on the data interface protocol of the automatic driving controller module, and send the assigned target map data to the data forwarding submodule; the data forwarding submodule is configured to receive the assigned target map data and a configuration file corresponding to the map forwarding module, and send the configuration file corresponding to the map forwarding module and the assigned target map data to the automatic driving controller module by calling a middleware; wherein the configuration file is configured to make the configuration of the scene simulation module consistent with that of the automatic driving controller module; The automatic driving controller module is configured to generate control parameters of the test vehicle based on the target map data and the configuration file, and send the control parameters to the scene simulation module, so that the scene simulation module controls the test vehicle based on the control parameters.

2. The simulation test system for autonomous driving according to claim 1, wherein The scene simulation module comprises an analysis submodule, which is configured to: convert the scene file into a lightweight text file; extract vehicle information of the test vehicle from the text file, wherein the vehicle information comprises motion state and position information of the test vehicle; analyze the text file by using preset analysis functions corresponding to different types of map elements to obtain map element information, and fuse the map element information with the vehicle information to obtain the initial map data.

3. The simulation test system for autonomous driving according to claim 2, wherein The scene simulation module further comprises a scene building submodule, which is configured to: import map information from a preset path, and build a simulation map scene based on the map information; generate a road file based on the simulation map scene; in response to an adding operation of adding vehicle information of the test vehicle and environment information in the road file, add the vehicle information of the test vehicle in the road file, and based on a preset correspondence relationship between different road types and environment information, add corresponding environment information to roads of different road types in the road file to generate the scene file; send the scene file to the analysis submodule.

4. The simulation test system for autonomous driving according to claim 1, wherein The map forwarding module further comprises a configuration submodule; The configuration submodule is configured to configure a simulation environment, task management, a communication interface, and a data interface to obtain configuration information, and to encapsulate the data interface and the simulation environment to obtain encapsulated information; A configuration file is generated based on the configuration information and the encapsulated information, and the configuration file is sent to the data forwarding submodule.

5. The simulation test system for autonomous driving according to claim 1, wherein The vehicle information of the test vehicle includes position information of the test vehicle; the target map data is filtered from the initial map data based on the vehicle information of the test vehicle, including: The position information of the test vehicle is determined as a navigation starting point, and a navigation endpoint corresponding to an automatic driving test task is determined; A navigation map path is determined based on the navigation starting point and the navigation endpoint, including map element information existing in a path within a preset path range centered on the navigation path; The initial map data is filtered based on the navigation map path to obtain target map data.

6. The simulation test system for autonomous driving according to claim 1, wherein The automatic driving controller module includes a map component submodule, a data interface submodule, and a decision control submodule; The map component submodule is configured to receive an Open drive file of a map sent by a third party; The data interface submodule is configured to disconnect the connection with the map component submodule when it is detected that the map component submodule has not received the Open drive file, and receive the target map data and the configuration file, and send the target map data and the configuration file to the decision control submodule; The decision control submodule is configured to generate control parameters of the test vehicle based on the target map data and the configuration file.

7. The simulation test system for autonomous driving according to claim 1, wherein The map forwarding module further includes a time submodule, and the scenario simulation module further includes a visualization submodule; The time submodule is configured to determine a first sending time at which the map forwarding module sends the current target map data to the automatic driving controller module; determine a first receiving time at which the map forwarding module receives the control parameters; send the first sending time and the first receiving time to the visualization submodule; The visualization submodule is configured to control the test vehicle based on the control parameters corresponding to the first receiving time on the target map data corresponding to the first sending time.

8. The simulated test system for autonomous driving of claim 1, wherein, The map forwarding module further includes an update submodule, which is configured to: receive an update instruction sent by the scenario simulation module, the update instruction including map data used to update a simulation scenario; update the latest target map data received by the map forwarding module based on the map data in the update instruction, so that the automatic driving controller module updates the control parameters of the test vehicle based on the updated target map data.

9. An automated driving simulation test method characterized by, A map forwarding module applied to a simulation test system, the simulation test system further including a scenario simulation module and an automatic driving controller module; the simulation test method includes: read initial map data stored by a scene simulation module based on a storage path, wherein the initial map data stored is stored by the scene simulation module based on a data interface protocol of an automatic driving controller module on initial map data parsed based on a scene file, and is shared to the map forwarding module in a shared memory manner, wherein the scene file comprises road environment information and vehicle information of a test vehicle, and the parsed initial map data comprises map element information and vehicle information of the test vehicle; obtain target map data from the initial map data stored based on vehicle information of the test vehicle; value the target map data based on the data interface protocol of the automatic driving controller module, and send a configuration file corresponding to the map forwarding module and the valued target map data to the automatic driving controller module in a calling middleware manner, so that the automatic driving controller module generates control parameters of the test vehicle based on the target map data and the configuration file, and sends the control parameters to the scene simulation module, and the scene simulation module controls the test vehicle based on the control parameters; wherein the configuration file is used to make the configuration of the scene simulation module consistent with that of the automatic driving controller module.

Images