Automatic driving simulation test model matching method, device and equipment
By constructing a driver behavior model and automatically calling it in simulation tests, the instability and inaccuracy of autonomous driving simulation tests are solved, achieving higher test stability and safety.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- AUTOMOTIVE INTELLIGENCE & CONTROL OF CHINA CO LTD
- Filing Date
- 2022-12-19
- Publication Date
- 2026-06-02
AI Technical Summary
Existing autonomous driving simulation testing methods suffer from poor stability and accuracy due to algorithm instability and uncertainty, thus failing to guarantee the safety of autonomous driving.
By constructing multiple driver behavior models, and based on the parameter information of the algorithm module and the interface information of the test vehicle, a driver behavior model is established, and the corresponding driver behavior model is automatically called during the simulation test, thereby improving the stability and accuracy of the simulation test.
This improves the stability and accuracy of autonomous driving simulation testing, ensuring the safety of autonomous driving.
Smart Images
Figure CN115963738B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of autonomous driving technology, and in particular to an autonomous driving simulation test model matching method, apparatus and equipment. Background Technology
[0002] With the continuous development of technologies such as the Internet, autonomous driving technology is also becoming increasingly sophisticated. Because the development cycle of autonomous driving systems is long and the operating environment places high demands on system performance, simulation testing of autonomous driving systems has become an unavoidable part of the technology process to ensure safety and stability.
[0003] In related technologies, data mining and deep learning are typically used to simulate and analyze driver behavior, forming corresponding algorithms. Simulation testing usually focuses on testing these algorithms. However, due to the instability and uncertainty of the algorithms, this simulation testing method suffers from poor stability and low accuracy, failing to guarantee the safety of autonomous driving. Summary of the Invention
[0004] This application provides a method, apparatus, and equipment for matching autonomous driving simulation test models, which improves the stability and accuracy of autonomous driving simulation testing and ensures the safety of autonomous driving.
[0005] In a first aspect, embodiments of this application provide an autonomous driving simulation test model matching method, including:
[0006] Multiple driver behavior models required for autonomous driving functionality are determined; the autonomous driving functionality includes multiple algorithm modules.
[0007] Based on the parameter information of the algorithm module and the interface information of the test vehicle, the multiple driver behavior models are established.
[0008] During the simulation test, at least one of the driver behavior models is invoked based on the output parameters of the algorithm module.
[0009] In one possible implementation, establishing the multiple driver behavior models based on the parameter information of the algorithm module and the interface information of the test vehicle includes:
[0010] Determine the target function corresponding to the driver behavior model; the target function is associated with at least one of the algorithm modules.
[0011] Obtain the interface information of the test vehicle, and obtain the parameter information output by at least one of the algorithm modules for the target function;
[0012] The driver behavior model is established based on the interface information and the parameter information.
[0013] In one possible implementation, invoking at least one of the driver behavior models based on the output parameters of the algorithm module includes:
[0014] Obtain the output parameters and determine the state changes of the test vehicle based on the output parameters;
[0015] Determine the driving behavior corresponding to the state change, and call the driver behavior model corresponding to the driving behavior.
[0016] In one possible implementation, the multiple driver behavior models required to determine the autonomous driving function include:
[0017] Based on the actual driving needs corresponding to the autonomous driving function, the multiple driver behavior models are determined.
[0018] In one possible implementation, the driver behavior model includes at least one of the following: a lane driving model, a lane changing model, an overtaking model, and a parking model.
[0019] In one possible implementation, the algorithm module includes at least one of the following: a perception module system, a fusion module system, a decision module system, a planning module system, a prediction module system, a lane driving system, a lane changing system, and a parking system.
[0020] Secondly, embodiments of this application provide an autonomous driving simulation test model matching device, comprising:
[0021] A determination module is used to determine multiple driver behavior models required for the autonomous driving function; the autonomous driving function includes multiple algorithm modules.
[0022] A module is established to build the multiple driver behavior models based on the parameter information of the algorithm module and the interface information of the test vehicle.
[0023] The calling module is used to call at least one of the driver behavior models during the simulation test, based on the output parameters of the algorithm module.
[0024] In one possible implementation, the establishment module is specifically used for:
[0025] Determine the target function corresponding to the driver behavior model; the target function is associated with at least one of the algorithm modules.
[0026] Obtain the interface information of the test vehicle, and obtain the parameter information output by at least one of the algorithm modules for the target function;
[0027] The driver behavior model is established based on the interface information and the parameter information.
[0028] In one possible implementation, the calling module is specifically used for:
[0029] Obtain the output parameters and determine the state changes of the test vehicle based on the output parameters;
[0030] Determine the driving behavior corresponding to the state change, and call the driver behavior model corresponding to the driving behavior.
[0031] In one possible implementation, the determining module is specifically used for:
[0032] Based on the actual driving needs corresponding to the autonomous driving function, the multiple driver behavior models are determined.
[0033] In one possible implementation, the driver behavior model includes at least one of the following: a lane driving model, a lane changing model, an overtaking model, and a parking model.
[0034] In one possible implementation, the algorithm module includes at least one of the following: a perception module system, a fusion module system, a decision module system, a planning module system, a prediction module system, a lane driving system, a lane changing system, and a parking system.
[0035] Thirdly, embodiments of this application provide a device, including: a processor and a memory;
[0036] The memory stores computer-executed instructions;
[0037] The processor executes computer execution instructions stored in the memory to implement the method as described in any of the first aspects.
[0038] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed, are used to implement the method described in any of the first aspects.
[0039] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed, implements the method described in any of the first aspects.
[0040] The autonomous driving simulation test model matching method, apparatus, and equipment provided in this application determine multiple driver behavior models required for an autonomous driving function. The autonomous driving function includes multiple algorithm modules. Multiple driver behavior models are established based on the parameter information of the algorithm modules and the interface information of the test vehicle. During simulation testing, at least one driver behavior model is invoked based on the output parameters of the algorithm modules. Thus, by constructing multiple driver behavior models, the corresponding driver behavior model can be automatically invoked during simulation testing, improving the stability and accuracy of autonomous driving simulation testing, thereby ensuring the safety of autonomous driving. Attached Figure Description
[0041] Figure 1 This is a schematic diagram illustrating an application scenario provided in the embodiments of this application;
[0042] Figure 2 A flowchart illustrating an autonomous driving simulation test model matching method provided in this application embodiment;
[0043] Figure 3 A flowchart illustrating another method for matching autonomous driving simulation test models provided in this application embodiment;
[0044] Figure 4 A schematic diagram of the structure of an autonomous driving simulation test model matching device provided in this application embodiment;
[0045] Figure 5 This is a schematic diagram of the structure of an autonomous driving simulation test model matching device provided in an embodiment of this application. Detailed Implementation
[0046] To enable those skilled in the art to better understand the technical solutions of this application, the application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments and drawings described herein are merely for explaining this application and are not intended to limit this application.
[0047] In recent years, with the rapid development of autonomous driving and advanced driver assistance systems (ADAS), simulation testing of autonomous driving and ADAS systems and their related algorithm modules has become an unavoidable part of autonomous driving technology due to the long development cycle of autonomous driving systems and their related algorithm modules and the high performance requirements of the operating environment. Simulating the actual driving environment of the system on a real-world platform is an effective way to verify the functionality and performance of autonomous driving systems and their related algorithm modules. To address how to meet the closed-loop simulation testing requirements of scenario-based virtual simulation testing of autonomous vehicles, current methods mostly focus on system-level closed loops, i.e., directly controlling the throttle and steering of the controlled vehicle in the simulator, thereby achieving closed-loop simulation between the simulator, the test object, the transfer tool, and the controlled vehicle.
[0048] In related technologies, methods such as data mining, temporal convolutional networks, and deep learning are typically used to simulate and analyze driver behavior, resulting in various algorithms. These algorithms are designed for the test vehicle, while the algorithm modules of autonomous driving functions are also algorithms. Thus, simulating and testing the algorithm modules of autonomous driving functions based on the algorithms of the test vehicle is inherently unstable and uncertain, leading to low stability and accuracy in autonomous driving simulation testing, and ultimately failing to guarantee the safety of autonomous driving.
[0049] Figure 1 This is a schematic diagram illustrating an application scenario provided in an embodiment of this application. Please refer to [link / reference]. Figure 1 This includes a simulation testing system 101 and a model library 102. The simulation testing system 101 can be deployed on a terminal device, such as a computer or mobile terminal; the model library 102 can be deployed on the terminal device where the simulation testing system 101 is located, or it can be deployed on other terminal devices; this application embodiment does not limit this. Figure 1 As shown, based on the requirements of autonomous driving functions, multiple driver behavior models are constructed and form a model library 102; subsequently, during the simulation test, the functional simulation test system 101 can call the corresponding driver behavior model from the model library 102 to perform simulation tests.
[0050] In this embodiment, multiple driver behavior models required for the autonomous driving function are determined. The autonomous driving function includes multiple algorithm modules. Multiple driver behavior models are established based on the parameter information of the algorithm modules and the interface information of the test vehicle. During simulation testing, at least one driver behavior model is invoked based on the output parameters of the algorithm modules. Thus, by constructing multiple driver behavior models, the corresponding model can be automatically invoked during simulation testing, improving the stability and accuracy of autonomous driving simulation testing, thereby ensuring the safety of autonomous driving.
[0051] The solution presented in this application will now be described in detail through specific embodiments. It should be noted that the following embodiments may exist independently or in combination with each other; identical or similar content will not be repeated in different embodiments.
[0052] The following combination Figure 2 The illustrated embodiment explains the process of matching autonomous driving simulation test models.
[0053] Figure 2 This is a flowchart illustrating an autonomous driving simulation test model matching method provided in an embodiment of this application. Please refer to... Figure 2 The method may include:
[0054] S201. Determine the multiple driver behavior models required for the autonomous driving function; the autonomous driving function includes multiple algorithm modules.
[0055] In this embodiment, the autonomous driving function may include multiple algorithm modules for performing various functions. The driver behavior model can refer to a simulation model corresponding to the driver's actual behavior. Specifically, in the autonomous driving function, multiple driver behavior models required by the autonomous driving function can be determined based on the different driving behaviors required by the tested vehicle during actual driving.
[0056] S202. Based on the parameter information of the algorithm module and the interface information of the test vehicle, establish multiple driver behavior models.
[0057] In this embodiment, parameter information may refer to the relevant parameter information of the parameter module in the autonomous driving function, specifically including the specific types of input parameters and / or output parameters, the units of the parameters, etc. The interface information of the test vehicle may refer to the various interfaces (Application Program Interface, API) of the test vehicle, which can be provided by the simulator.
[0058] Specifically, in this embodiment of the application, when constructing a driver behavior model, at least one algorithm module associated with the driver behavior model can be determined first. Then, based on the parameter information of these algorithm modules and the interface information of the test vehicle provided by the simulator, a corresponding algorithm is selected to construct the driver behavior model. For example, if it is necessary to model lane-keeping behavior in an autonomous driving function, the corresponding algorithm module includes a control algorithm module. In this case, a trajectory following algorithm or similar algorithm can be selected to establish a driver behavior model corresponding to lane-keeping behavior based on the parameter information of the control algorithm module and the API interface information of the test vehicle.
[0059] S203. During the simulation test, at least one driver behavior model is called according to the output parameters of the algorithm module.
[0060] In this embodiment, the output parameters can refer to specific parameter values of the algorithm module, such as road signs or road numbers (Identity Document, ID), speed, and acceleration. After constructing multiple driver behavior models, these models can be grouped into a model library. Subsequently, during simulation testing, when the output parameters of a certain algorithm module representing driving behavior change, the simulation testing system can call the driver behavior model from the model library to simulate driving behavior. This enables automatic matching of driver behavior models during the simulation testing of autonomous driving functions and various algorithm modules, ensuring the determinism and stability of the simulation test.
[0061] The autonomous driving simulation test model matching method provided in this application determines multiple driver behavior models required for an autonomous driving function. This autonomous driving function includes multiple algorithm modules. Multiple driver behavior models are established based on the parameter information of the algorithm modules and the interface information of the test vehicle. During the simulation test, at least one driver behavior model is invoked based on the output parameters of the algorithm modules. In this way, by constructing multiple driver behavior models, the corresponding driver behavior model can be automatically invoked during simulation testing, improving the stability and accuracy of autonomous driving simulation testing, thereby ensuring the safety of autonomous driving.
[0062] Based on the above embodiments, the following is combined with Figure 3 The illustrated embodiment provides a detailed explanation of the process of matching autonomous driving simulation test models.
[0063] Figure 3 This is a flowchart illustrating another method for matching autonomous driving simulation test models provided in this application. Please refer to... Figure 3 The method may include:
[0064] S301. Based on the actual driving needs corresponding to the autonomous driving function, determine multiple driver behavior models.
[0065] In this embodiment, actual driving needs refer to the driving needs that the tested vehicle may have during normal driving in a real environment, such as driving within a lane or changing lanes. The simulation testing system can identify and determine multiple driver behavior models required for autonomous driving functions based on actual driving needs, thereby improving the comprehensiveness of driver behavior model determination.
[0066] In one possible implementation, the driver behavior model includes at least one of the following: lane driving model, lane changing model, overtaking model, and parking model.
[0067] In this embodiment, multiple driver behavior models can be combined into a model library, which is categorized according to different types. This model library can contain four main types (first-level types): lane-keeping models, lane-changing models, overtaking models, and parking models. These main types can be further subdivided into different subcategories (second-level types), enabling unified management and scheduling of driver behavior models, facilitating retrieval, and thus improving the speed of subsequent model invocation. Examples of specific subcategories are as follows:
[0068] Lane-based driving models can include: lane cruise, lane following, lane obstacle avoidance, lane emergency braking, etc.; lane-changing models can include: left lane change, right lane change, continuous left lane change, continuous right lane change, lane change cancellation, etc.; overtaking models can include: overtaking in the fast lane, overtaking in the slow lane, overtaking failure, etc.; parking models can include: reversing into parking from the left, reversing into parking from the right, parking with the front of the vehicle in, parking with the front of the vehicle in from the right, reversing out from the left, reversing out from the right, parking with the front of the vehicle in from the left, parking with the front of the vehicle in from the right, etc.
[0069] It should be noted that the above main types and subclasses are merely examples and are not intended to limit the embodiments of this application. Other types of main types and subclasses may also be included in the model library, and the embodiments of this application do not limit this.
[0070] S302, The autonomous driving function includes multiple algorithm modules; the target function corresponding to the driver behavior model is determined; the target function is associated with at least one algorithm module.
[0071] In this embodiment, the target function can refer to the specific function that the driver behavior model wants to achieve, such as emergency braking within the lane or changing lanes to the left. The target function can be associated with one or more algorithm modules; that is, in the autonomous driving function, the target function may require one or more algorithm modules to implement.
[0072] In one possible implementation, the algorithm module includes at least one of the following: a perception module system, a fusion module system, a decision module system, a planning module system, a prediction module system, a lane driving system, a lane changing system, and a parking system.
[0073] In this embodiment, the autonomous driving function may include multiple algorithm modules that work together. The perception module system can be used to perceive surrounding road conditions, etc. The fusion module system can be used to fuse data from multiple sensors to improve data accuracy. The decision module system can be used to output decisions. The planning module system can be used for path planning. The prediction module system can be used for collision detection, etc. The lane-keeping system, lane-changing system, and parking system correspond to driving behaviors such as lane-keeping, lane-changing, and parking, respectively.
[0074] It should be noted that the above algorithm modules are merely examples and are not intended to limit the embodiments of this application. Based on actual needs, other algorithm modules may also be included in the autonomous driving function, and the embodiments of this application do not limit this.
[0075] S303. Obtain the interface information of the test vehicle and the parameter information output by at least one algorithm module for the target function.
[0076] S304. Based on the interface information and parameter information, establish a driver behavior model.
[0077] In this embodiment of the application, after determining the target function of the driver behavior model and at least one algorithm module associated with the target function, the API interface information of the test vehicle in the control simulation test scenario provided by the simulator to the user can be determined, and the parameter information of at least one algorithm module associated with the target function can be obtained. The corresponding algorithm is then selected to model the driver behavior model.
[0078] Specifically, for each driver behavior model, when building the model, the output parameters of the algorithm module and the input parameters received by the API interface need to be combined. The output parameters of the algorithm module can be used as the input of the driver behavior model, and the API interface can be used to receive the output of the driver behavior model, ensuring that the specific driving behavior determined by the algorithm module can be accurately sent to the corresponding API interface. In this way, it can be ensured that the input and output of the finally built driver behavior model correspond to the output parameters of the algorithm module and the API interface information, respectively, ensuring smooth subsequent model calls and improving the stability of model calls.
[0079] S305. Obtain the output parameters of the algorithm module, and determine the state changes of the test vehicle based on the output parameters.
[0080] S306. Determine the driving behavior corresponding to the state change and call the driver behavior model corresponding to the driving behavior.
[0081] In this embodiment, the change in the state of the test vehicle can refer to changes in the test vehicle's path, lane, speed, etc. Different algorithm modules have various types of output parameters. Taking the decision module as an example, the output parameters may include a road ID. If the road ID output this time is different from the road ID output last time, it can be determined that the state of the test vehicle has changed. The driving behavior corresponding to this state change is a lane change. The simulation test system can call the driver behavior model corresponding to the lane change to implement the lane change process.
[0082] Specifically, whether the test vehicle experiences state changes can be determined using a state judgment algorithm, such as a finite state machine algorithm or a decision tree, which is not limited in this embodiment. For example, the decision algorithm module may involve some or all sub-type models of four types of driver behavior models: lane changing, driving within a lane, overtaking, and parking. Using a finite state machine algorithm, the state changes of the test vehicle can be determined, and then the driver behavior model can be invoked based on the driving behavior corresponding to the state change, achieving automatic matching of various driver behavior models.
[0083] exist Figure 3 In the illustrated embodiment, multiple driver behavior models are determined based on the actual driving requirements corresponding to the autonomous driving function. The autonomous driving function includes multiple algorithm modules. A target function corresponding to the driver behavior model is determined. The target function is associated with at least one algorithm module. The interface information of the test vehicle and the parameter information output by at least one algorithm module for the target function are obtained. A driver behavior model is established based on the interface information and parameter information. The output parameters of the algorithm module are obtained, and the state changes of the test vehicle are determined based on the output parameters. The driving behavior corresponding to the state change is determined, and the driver behavior model corresponding to the driving behavior is called. In this way, the simulation and matching of the driver behavior model can be realized, thereby achieving a closed loop between the autonomous driving function and its algorithm modules and the simulation test system, simulation tools, and simulation test environment. By fully utilizing the relevant API interface information and the parameter information of the algorithm modules to simulate driver behavior, the stability of the driver behavior model is improved, providing a stable model for the simulation test of the autonomous driving function and algorithm modules, improving the accuracy of the simulation test, and thus improving the safety of autonomous driving.
[0084] Figure 4 This is a schematic diagram of the structure of an autonomous driving simulation test model matching device provided in an embodiment of this application. Please refer to... Figure 4 The autonomous driving simulation test model matching device 10 may include:
[0085] Module 11 is used to determine multiple driver behavior models required for the autonomous driving function; the autonomous driving function includes multiple algorithm modules.
[0086] Module 12 is established to build multiple driver behavior models based on the parameter information of the algorithm module and the interface information of the test vehicle.
[0087] Module 13 is used to invoke at least one driver behavior model during the simulation test, based on the output parameters of the algorithm module.
[0088] In one possible implementation, module 12 is specifically used for:
[0089] Determine the target function corresponding to the driver behavior model; the target function is associated with at least one algorithm module.
[0090] Obtain the interface information of the test vehicle, and obtain the parameter information output by at least one algorithm module for the target function;
[0091] A driver behavior model is established based on the interface information and parameter information.
[0092] In one possible implementation, module 13 is invoked, specifically for:
[0093] Obtain the output parameters and determine the state changes of the test vehicle based on the output parameters;
[0094] Determine the driving behavior corresponding to the state change, and call the driver behavior model corresponding to the driving behavior.
[0095] In one possible implementation, the determining module 11 is specifically used for:
[0096] Based on the actual driving needs corresponding to autonomous driving functions, multiple driver behavior models are determined.
[0097] In one possible implementation, the driver behavior model includes at least one of the following: lane driving model, lane changing model, overtaking model, and parking model.
[0098] In one possible implementation, the algorithm module includes at least one of the following: a perception module system, a fusion module system, a decision module system, a planning module system, a prediction module system, a lane driving system, a lane changing system, and a parking system.
[0099] The autonomous driving simulation test model matching device 10 provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be described again here.
[0100] Figure 5 This is a schematic diagram of the structure of an autonomous driving simulation test model matching device provided in an embodiment of this application. Please refer to... Figure 5 The autonomous driving simulation test model matching device 20 may include a memory 21 and a processor 22. For example, the memory 21 and the processor 22 are interconnected via a bus 23.
[0101] Memory 21 is used to store program instructions;
[0102] The processor 22 is used to execute the program instructions stored in the memory to implement the autonomous driving simulation test model matching method shown in the above embodiment.
[0103] Figure 5The autonomous driving simulation test model matching device 20 shown in the embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be repeated here.
[0104] This application provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the above-described autonomous driving simulation test model matching method when executed by a processor.
[0105] This application embodiment may also provide a computer program product, including a computer program, which, when executed by a processor, can implement the above-described autonomous driving simulation test model matching method.
[0106] It should be noted that the processor mentioned in the embodiments of this application can be a central processing unit (CPU), or it can be 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. A general-purpose processor can be a microprocessor or any conventional processor.
[0107] It should be understood that the memory mentioned in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DR RAM). It should be noted that when the processor is a general-purpose processor, DSP, ASIC, FPGA, or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, the memory (storage module) is integrated into the processor. It should be noted that the memory described herein is intended to include, but is not limited to, these and any other suitable types of memory.
[0108] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes 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.
[0109] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processing unit of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processing unit of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0110] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0111] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0112] Regarding the modules / units included in the various devices and products described in the above embodiments, they can be software modules / units, hardware modules / units, or a combination of both. Each device and product can be applied to or integrated into a chip, chip module, or terminal device. For example, for devices and products applied to or integrated into a chip, each included module / chip can be implemented entirely using hardware methods such as circuits, or at least some modules / units can be implemented using software programs running on a processor integrated within the chip, while the remaining modules / units can be implemented using hardware methods such as circuits.
[0113] In this application, the term "comprising" and its variations can refer to non-limiting inclusion; the term "or" and its variations can refer to "and / or". The terms "first", "second", etc., in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. In this application, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.
[0114] The above are only some embodiments of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.
Claims
1. A method for matching simulation test models for autonomous driving, characterized in that, include: Multiple driver behavior models required for autonomous driving functionality are determined; the autonomous driving functionality includes multiple algorithm modules. Based on the parameter information of the algorithm module and the interface information of the test vehicle, multiple driver behavior models are established. Each driver behavior model is a simulation model corresponding to the driver's actual behavior. During simulation testing, at least one driver behavior model is invoked based on the output parameters of the algorithm module. Each driver behavior model is an execution model independent of the autonomous driving function algorithm module. For each driver behavior model, when establishing the model, the output parameters of the algorithm module and the input parameters received from the API interface are combined. The output parameters of the algorithm module serve as the input to the driver behavior model, and the API interface is used to receive the output of the driver behavior model, ensuring that the specific driving behavior determined by the algorithm module is accurately sent to the corresponding API interface. The step of establishing the multiple driver behavior models based on the parameter information of the algorithm module and the interface information of the test vehicle includes: Determine the target function corresponding to the driver behavior model; the target function is associated with at least one of the algorithm modules. Obtain the interface information of the test vehicle, and obtain the parameter information output by at least one of the algorithm modules for the target function; The driver behavior model is established based on the interface information and the parameter information. The step of invoking at least one of the driver behavior models based on the output parameters of the algorithm module includes: Obtain the output parameters and determine the state changes of the test vehicle based on the output parameters; Determine the driving behavior corresponding to the state change, and call the driver behavior model corresponding to the driving behavior.
2. The method according to claim 1, characterized in that, The multiple driver behavior models required to determine the autonomous driving function include: Based on the actual driving needs corresponding to the autonomous driving function, the multiple driver behavior models are determined.
3. The method according to claim 1 or 2, characterized in that, The driver behavior model includes at least one of the following: lane driving model, lane changing model, overtaking model, and parking model.
4. The method according to claim 1 or 2, characterized in that, The algorithm module includes at least one of the following: a perception module system, a fusion module system, a decision module system, a planning module system, a prediction module system, a lane driving system, a lane changing system, and a parking system.
5. An autonomous driving simulation test model matching device, characterized in that, include: A determination module is used to determine multiple driver behavior models required for the autonomous driving function; the autonomous driving function includes multiple algorithm modules. A module is established to build multiple driver behavior models based on the parameter information of the algorithm module and the interface information of the test vehicle. Each driver behavior model is a simulation model corresponding to the driver's actual behavior. The driver behavior model is an execution model independent of the autonomous driving function algorithm module. For each driver behavior model, during model building, the output parameters of the algorithm module and the input parameters received from the API interface are combined. The output parameters of the algorithm module serve as the input to the driver behavior model, and the API interface is used to receive the output of the driver behavior model, ensuring that the specific driving behavior determined by the algorithm module is accurately sent to the corresponding API interface. The calling module is used to call at least one of the driver behavior models during the simulation test, based on the output parameters of the algorithm module. The establishment module is specifically used to determine the target function corresponding to the driver behavior model; the target function is associated with at least one of the algorithm modules; obtain the interface information of the test vehicle, and obtain the parameter information output by at least one of the algorithm modules for the target function; and establish the driver behavior model based on the interface information and the parameter information. The calling module is specifically used to obtain the output parameters, and determine the state changes of the test vehicle based on the output parameters; determine the driving behavior corresponding to the state changes, and call the driver behavior model corresponding to the driving behavior.
6. An autonomous driving simulation test model matching device, characterized in that, include: Processor, memory; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 4.
7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed, are used to implement the method described in any one of claims 1 to 4.
8. A computer program product, characterized in that, Includes a computer program that, when executed, implements the method described in any one of claims 1 to 4.