A method, device, electronic device and storage medium for generating an autonomous driving simulation scene

Abstract

The present application provides a method, device, electronic device and storage medium for generating an autonomous driving simulation scenario, including: calculating the failure probability of multiple sample autonomous driving simulation scenarios based on a probability function, determining a sample autonomous driving dangerous simulation scenario in multiple sample autonomous driving simulation scenarios; determining a dangerous element set and a safe element set based on multiple sample autonomous driving dangerous simulation scenarios; selecting multiple first target dangerous elements and multiple target safe elements from the dangerous element set and the safe element set respectively based on the simulation scenario setting requirements, performing simulation scenario design based on the multiple first target dangerous elements and the multiple target safe elements, and generating an autonomous driving dangerous simulation scenario. By designing a simulation test of a dangerous scenario based on dangerous elements, it is more in line with the actual state of the autonomous driving simulation scenario, and can effectively ensure the accuracy of the scenario design and implementation results.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving simulation technology, and in particular to a method, device, electronic device and storage medium for generating an autonomous driving simulation scene. Background Art

[0002] Autonomous driving simulation testing is a test verification method that uses effective test cases to quantify parameters and confirm the effects of the autonomous driving system or system components (perception unit, decision unit, and execution unit) under test in a simulation environment. The simulation test scenario includes static and dynamic scenarios. Static scenarios refer to roads, traffic signs, road infrastructure, etc.; dynamic scenarios refer to targets, changing weather, etc. At present, based on simulation tools, the construction and configuration of static and dynamic scenarios for simulation testing can reproduce real traffic test scenarios, saving a lot of manpower and material resources in the development and test verification of autonomous driving systems. However, for some dangerous scenarios, edge scenarios, and complex scenarios that are difficult to test in actual road tests, especially some "suddenly appearing scene elements", existing simulation test tools do not have good scenario design and construction solutions, making them difficult to reproduce. Therefore, how to effectively construct autonomous driving simulation scenarios has become a technical issue that cannot be underestimated. Summary of the invention

[0003] In view of this, the purpose of the present application is to provide a method, device, electronic device and storage medium for generating an autonomous driving simulation scene. By designing a simulation test of a dangerous scene based on dangerous elements, it is more in line with the actual state of the autonomous driving simulation scene, can effectively ensure the accuracy of the scene design and implementation results, and can more easily support the setting of emergencies, dangerous events or extreme events, so as to detect the recognition and response of the autonomous driving algorithm under test.

[0004] The present application provides a method for generating an autonomous driving simulation scene, the method comprising:

[0005] Acquire multiple sample autonomous driving simulation scenarios, calculate the failure probabilities of the multiple sample autonomous driving simulation scenarios based on a probability function, and determine a sample autonomous driving dangerous simulation scenario among the multiple sample autonomous driving simulation scenarios;

[0006] Based on the plurality of sample autonomous driving danger simulation scenarios, determining a set of danger elements and a set of safety elements;

[0007] Based on the simulation scenario setting requirements, multiple first target dangerous elements and multiple target safety elements are respectively selected from the dangerous element set and the safety element set, and a simulation scenario is designed based on the multiple first target dangerous elements and the multiple target safety elements to generate an autonomous driving dangerous simulation scenario.

[0008] In a possible implementation manner, the calculating the failure probability of the plurality of sample autonomous driving simulation scenarios based on the probability function to determine a sample autonomous driving dangerous simulation scenario among the plurality of sample autonomous driving simulation scenarios includes:

[0009] Calculating the scene evaluation index in each of the sample autonomous driving simulation scenes based on the probability function to determine the scene failure probability of each of the sample autonomous driving simulation scenes;

[0010] Based on the scenario failure probability and a preset failure probability threshold, a sample autonomous driving dangerous simulation scenario is determined among multiple sample autonomous driving simulation scenarios.

[0011] In a possible implementation manner, determining a set of dangerous elements and a set of safe elements based on a plurality of sample autonomous driving dangerous simulation scenarios includes:

[0012] Extracting the danger simulation parameter information in the sample autonomous driving danger simulation scene based on a plurality of preset danger categories to determine the danger elements, and forming the danger element set based on the danger elements under each danger category; wherein the danger category includes a weather category, a target category, and an obstacle category;

[0013] Based on multiple preset safety categories, the safety simulation parameter information in the sample autonomous driving danger simulation scenario is extracted to determine the safety elements, and the safety element set is formed based on the safety elements under each safety category; wherein the safety categories include road categories, road surface categories, lane feature categories, and road edge categories.

[0014] In a possible implementation manner, the simulation scene setting requirements include:

[0015] For the first target dangerous element, at least one of the first target dangerous elements is selected from the dangerous element set and one dangerous element of the same category is selected, and the sum of the first target dangerous elements accounts for less than or equal to the first proportion in all elements of the autonomous driving simulation scene;

[0016] For the target safety element, at least one target safety element is selected from the safety elements corresponding to each safety category, and the sum of the target safety elements accounts for a proportion greater than or equal to a second proportion in all elements of the autonomous driving simulation scene; wherein the second proportion is greater than the first proportion, and the sum of the second proportion and the first proportion is 1.

[0017] In a possible implementation manner, the performing simulation scenario design based on the plurality of the first target dangerous elements and the plurality of the target safe elements to generate an autonomous driving dangerous simulation scenario includes:

[0018] Based on the target safety element, the autonomous driving algorithm is controlled to perform a simulation test. After the autonomous driving algorithm controls the vehicle to be in a stable operating state, the first target danger element is added to the simulation test to generate the autonomous driving danger simulation scenario.

[0019] In a possible implementation manner, after the autonomous driving algorithm is controlled to perform a simulation test based on the target safety element and the autonomous driving algorithm controls the vehicle to be in a stable operating state, the first target dangerous element is added to the simulation test to generate the autonomous driving dangerous simulation scenario, the generation method further includes:

[0020] Continue to screen out a second target dangerous element from the dangerous element set, and add the second target dangerous element to the current simulation test in real time.

[0021] In a possible implementation manner, after performing simulation scenario design based on the plurality of first target dangerous elements and the plurality of target safety elements to generate an autonomous driving dangerous simulation scenario, the generating method further includes:

[0022] The information corresponding to the first target dangerous element in the autonomous driving dangerous simulation scene and the test results of the autonomous driving dangerous simulation scene are stored.

[0023] The embodiment of the present application also provides a device for generating a dangerous simulation scenario for autonomous driving, the generating device comprising:

[0024] a dangerous scenario determination module, configured to obtain a plurality of sample autonomous driving simulation scenarios, calculate the failure probabilities of the plurality of sample autonomous driving simulation scenarios based on a probability function, and determine a sample autonomous driving dangerous simulation scenario among the plurality of sample autonomous driving simulation scenarios;

[0025] An element splitting module, used to determine a dangerous element set and a safe element set based on a plurality of sample autonomous driving dangerous simulation scenarios;

[0026] A design module is used to select a plurality of first target dangerous elements and a plurality of target safety elements from the dangerous element set and the safety element set respectively based on the simulation scenario setting requirements, perform simulation scenario design based on the plurality of first target dangerous elements and the plurality of target safety elements, and generate an autonomous driving dangerous simulation scenario.

[0027] An embodiment of the present application also provides an electronic device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the method for generating an autonomous driving simulation scene as described above are performed.

[0028] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for generating an autonomous driving simulation scene as described above are executed.

[0029] The embodiment of the present application provides a method, device, electronic device and storage medium for generating an autonomous driving simulation scene, the generating method comprising: obtaining multiple sample autonomous driving simulation scenes, calculating the failure probability of the multiple sample autonomous driving simulation scenes based on a probability function, and determining a sample autonomous driving dangerous simulation scene in the multiple sample autonomous driving simulation scenes; determining a dangerous element set and a safe element set based on the multiple sample autonomous driving dangerous simulation scenes; selecting multiple first target dangerous elements and multiple target safe elements from the dangerous element set and the safe element set respectively based on the simulation scene setting requirements, performing simulation scene design based on the multiple first target dangerous elements and the multiple target safe elements, and generating an autonomous driving dangerous simulation scene. By designing a simulation test of a dangerous scene based on dangerous elements, it is more in line with the real state of the autonomous driving simulation scene, can effectively ensure the accuracy of the scene design and implementation results, and can more easily support the setting of emergencies, dangerous events or extreme events, so as to detect the recognition and response of the tested autonomous driving algorithm.

[0030] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 A flowchart of a method for generating an autonomous driving simulation scene provided in an embodiment of the present application;

[0033] Figure 2One of the structural schematic diagrams of a method and device for generating an autonomous driving simulation scene provided in an embodiment of the present application;

[0034] Figure 3 A second structural diagram of a method and device for generating an autonomous driving simulation scene provided in an embodiment of the present application;

[0035] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0036] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art, under the guidance of the content of the present application, can add one or more other operations to the flowchart, or remove one or more operations from the flowchart.

[0037] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0038] In order to enable those skilled in the art to use the contents of this application, the following implementation method is given in combination with the specific application scenario of "generating an autonomous driving simulation scenario". For those skilled in the art, the general principles defined here can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application.

[0039] First, the application scenarios to which this application is applicable are introduced. This application can be applied in the field of autonomous driving simulation technology.

[0040] According to research, autonomous driving simulation testing is a test verification method that uses effective test cases to quantify parameters and confirm the effects of the autonomous driving system or system components (perception unit, decision unit, and execution unit) under test. The simulation test scenario includes static and dynamic scenarios. Static scenarios refer to roads, traffic signs, road infrastructure, etc.; dynamic scenarios refer to target objects, changing weather, etc. At present, based on simulation tools, the construction and configuration of static and dynamic scenarios for simulation tests can reproduce real traffic test scenarios, saving a lot of manpower and material resources in the development and test verification of autonomous driving systems. However, for some dangerous scenarios, edge scenarios, and complex scenarios that are difficult to test in actual road tests, especially some "suddenly appearing scene elements", the existing simulation test tools do not have a good scenario design and construction solution, and it is difficult to reproduce. Therefore, how to effectively construct autonomous driving simulation scenarios has become a technical issue that cannot be underestimated.

[0041] Based on this, an embodiment of the present application provides a method for generating an autonomous driving simulation scenario. By designing a simulation test of a dangerous scenario based on dangerous elements, it is more in line with the actual state of the autonomous driving simulation scenario, can effectively ensure the accuracy of the scenario design and implementation results, and can more easily support the setting of emergencies, dangerous events or extreme events, so as to detect the recognition and response of the autonomous driving algorithm under test.

[0042] See also Figure 1 , Figure 1 This is a flow chart of a method for generating an autonomous driving simulation scene provided in an embodiment of the present application. Figure 1 As shown in, the generation method provided in the embodiment of the present application includes:

[0043] S101: Acquire multiple sample autonomous driving simulation scenarios, calculate the failure probabilities of the multiple sample autonomous driving simulation scenarios based on a probability function, and determine a sample autonomous driving dangerous simulation scenario among the multiple sample autonomous driving simulation scenarios.

[0044] In this step, multiple sample autonomous driving simulation scenarios are obtained, the failure probabilities of the multiple sample autonomous driving simulation scenarios are calculated according to a probability function, and sample autonomous driving dangerous simulation scenarios among the multiple sample autonomous driving simulation scenarios are determined.

[0045] Here, based on a large amount of real natural driving data, several sample autonomous driving simulation scenes are obtained through scene slicing, scene extraction and scene restoration. 。

[0046] In a possible implementation manner, the calculating the failure probability of the plurality of sample autonomous driving simulation scenarios based on the probability function to determine a sample autonomous driving dangerous simulation scenario among the plurality of sample autonomous driving simulation scenarios includes:

[0047] A: Calculate the scenario evaluation index in each of the sample autonomous driving simulation scenarios based on the probability function to determine the scenario failure probability of each of the sample autonomous driving simulation scenarios.

[0048] Here, the scenario evaluation index in each sample autonomous driving simulation scenario is calculated according to the probability function to determine the scenario failure probability of each sample autonomous driving simulation scenario.

[0049] Among them, scene evaluation indicators include collision time, relative distance between vehicles, etc.

[0050] Here, it is assumed that the probability density function of the sample autonomous driving simulation scenario X is expressed as To express. represents an evaluation index of the autonomous driving system in the simulation scenario X, and without loss of generality, the failure of the system is defined as , so the failure probability of the autonomous driving system in scenario X can be expressed as:

[0051]

[0052] in, is the scenario failure probability of scenario X, Prob is the probability function, fail scenario The simulation scenario failed.

[0053] B: Based on the failure probability of the scenario and a preset failure probability threshold, determine a sample autonomous driving dangerous simulation scenario from among multiple sample autonomous driving simulation scenarios.

[0054] Here, the failure probability threshold is customized. , when the scenario fails Greater than the set threshold , The autonomous driving system or system component under test does not meet the safety requirements under the simulation test scenario, which is defined as a sample autonomous driving danger simulation scenario, also known as a "dangerous scenario."

[0055] S102: Based on the plurality of sample autonomous driving danger simulation scenarios, determine a set of danger elements and a set of safety elements.

[0056] In this step, a set of dangerous elements and a set of safe elements are determined based on multiple sample autonomous driving danger simulation scenarios.

[0057] In a possible implementation manner, determining a set of dangerous elements and a set of safe elements based on a plurality of sample autonomous driving dangerous simulation scenarios includes:

[0058] (1): Based on a plurality of preset hazard categories, the hazard simulation parameter information in the sample autonomous driving hazard simulation scene is extracted to determine the hazard elements, and the hazard element set is formed based on the hazard elements under each hazard category; wherein the hazard category includes a weather category, a target object category, and an obstacle category.

[0059] Here, the danger simulation parameter information in the sample autonomous driving danger simulation scene is extracted according to the preset multiple danger categories to determine the danger elements, and a danger element set is formed according to the danger elements under each danger category.

[0060] Among them, the definition m Dangerous elements A and n Security Elements B The simulation test scenario X is expressed as: That is, dangerous elements are used A To represent, including 3 categories of elements: 1) Subcategory - Weather elements A 天 : Heavy rain, heavy snow, heavy fog, thunderstorms and lightning, glare, etc. A 天 ={heavy rain, heavy snow, heavy fog, thunderstorm lightning, glare}; 2) Subcategory-target element A 交 : The target vehicle cuts in extremely, brakes suddenly, etc., i.e. A 交 ={target vehicle extreme cut-in, emergency braking}; 3) Subcategory - obstacle element A 障 : A billboard that suddenly falls, a stone that suddenly falls, an uncovered manhole cover, etc. A 障 ={suddenly fallen billboards, suddenly fallen stones, uncovered manhole covers}.

[0061] (2): Based on a plurality of preset safety categories, the safety simulation parameter information in the sample autonomous driving danger simulation scenario is extracted to determine the safety elements, and the safety element set is formed based on the safety elements under each safety category; wherein the safety categories include road categories, road surface categories, lane feature categories, and road edge categories.

[0062] Here, the safety simulation parameter information in the sample autonomous driving danger simulation scenario is extracted according to multiple preset safety categories to determine the safety elements, and a safety element set is formed according to the safety elements under each safety category.

[0063] Among them, the security element is B It is represented by 6 category elements: 1) Subcategory-Road Type B 路类 : Urban roads, rural roads, highways, parking lots, etc., i.e. B 路类 ={Urban Road, Rural Road, Highway, Parking Lot}; 2) Subcategory - Road Surface B 路表 : Asphalt concrete pavement, cement concrete pavement, etc., namely B 路表 ={asphalt concrete pavement, cement concrete pavement}; 3) Subcategory-lane characteristics B 车道 : Motor vehicle lane, emergency lane, tidal lane, non-motor vehicle lane, bus lane, etc. B 车道 ={motor vehicle lane, emergency lane, tidal lane, non-motor vehicle lane, bus lane}; 4) Subcategory - road edge B 标线 : solid white line, dashed white line, solid yellow line, dashed yellow line, etc., i.e. B 标线 ={solid white line, dashed white line, solid yellow line, dashed yellow line}; 5) Subcategory - road edge B 路边 : kerbstones, hard shoulders, guardrails, etc., i.e. B 路边 ={kerb, hard shoulder, guardrail}; 6) Subcategory - Traffic Control Facilities B 交控 (If applicable): road traffic signs, traffic lights, etc., B 交控 ={road traffic signs, traffic lights}.

[0064] S103: Based on the simulation scenario setting requirements, a plurality of first target dangerous elements and a plurality of target safety elements are respectively selected from the dangerous element set and the safety element set, and a simulation scenario is designed based on the plurality of the first target dangerous elements and the plurality of the target safety elements to generate an autonomous driving dangerous simulation scenario.

[0065] In this step, multiple first target dangerous elements and multiple target safety elements are respectively selected from the dangerous element set and the safety element set according to the simulation scene setting requirements, and a simulation scene is designed based on the multiple first target dangerous elements and the multiple target safety elements to generate an autonomous driving dangerous simulation scene.

[0066] In a possible implementation manner, the simulation scene setting requirements include:

[0067] For the first target dangerous element, at least one of the first target dangerous elements is selected from the dangerous element set and one dangerous element of the same category is selected, and the sum of the first target dangerous elements accounts for less than or equal to the first proportion in all elements of the autonomous driving simulation scene;

[0068] For the target safety element, at least one target safety element is selected from the safety elements corresponding to each safety category, and the sum of the target safety elements accounts for a proportion greater than or equal to a second proportion in all elements of the autonomous driving simulation scene; wherein the second proportion is greater than the first proportion, and the sum of the second proportion and the first proportion is 1.

[0069] Among them, the first accounts for 20% and the second accounts for 80%.

[0070] Here, assuming a and b Dangerous element collection A and security element collection B The elements in , then a ∈ A , b ∈ B . Dangerous elements in simulation test scenario X m and security elements n At least the following conditions must be met:

[0071] 1) Contain at least one dangerous element, that is, m≥1; 2) You can select any subcategory of dangerous elements to set, that is A 天 , A 目 and A 障 You can choose any one, that is, a∈ A 天 or a∈ A 目 or a∈ A 障 ; 3) Only one dangerous element of the same subcategory can be selected, but dangerous elements of different subcategories can be superimposed; 4) Safety elements must be in the subcategory B 路类 ，B路表 ，B 车道 ， B 路边 ，B 标线 Select at least one from each subcategory, that is, n≥5; 5) The sum of dangerous elements in all elements of the scene does not exceed 20%, that is:

[0072]

[0073] 6) The sum of safety elements accounts for no less than 80% of all elements in the scene, that is:

[0074]

[0075] In a possible implementation, the designing of simulation scenarios based on the plurality of the first target dangerous elements and the plurality of the target safe elements to generate an autonomous driving dangerous simulation scenario includes:

[0076] Based on the target safety element, the autonomous driving algorithm is controlled to perform a simulation test. After the autonomous driving algorithm controls the vehicle to be in a stable operating state, the first target danger element is added to the simulation test to generate the autonomous driving danger simulation scenario.

[0077] Here, the autonomous driving algorithm is controlled according to the target safety element to perform simulation testing. After the autonomous driving algorithm controls the vehicle to be in a stable operating state, the first target danger element is added to the simulation test to generate an autonomous driving danger simulation scenario.

[0078] In the specific embodiment, a sufficiently long road is reserved to ensure that the autonomous driving system is activated and the test vehicle speed reaches a stable state. When the conditions are met, the simulation tool adds dangerous elements in real time at a certain distance in the direction of the test vehicle. A If dangerous elements A Include A 天 and A 障 Class, then set the weather elements and obstacle elements in real time. If the dangerous elements A Include A 交 For example, when the test vehicle continues to drive after reaching a stable state, a target vehicle traveling at a constant speed of 30 km / h along the center line of the lane can be set in real time 30 meters away from the test vehicle, and the subsequent running trajectory of the target vehicle can be set [for example, the target vehicle maintains a constant speed in the lane, suddenly cuts into the lane where the test vehicle is located and continues to travel at a constant speed of 30 km / h] to simulate dangerous scenarios.

[0079] In a possible implementation manner, after the autonomous driving algorithm is controlled to perform a simulation test based on the target safety element and the autonomous driving algorithm controls the vehicle to be in a stable operating state, the first target dangerous element is added to the simulation test to generate the autonomous driving dangerous simulation scenario, the generation method further includes:

[0080] Continue to screen out a second target dangerous element from the dangerous element set, and add the second target dangerous element to the current simulation test in real time.

[0081] Here, a second target dangerous element is added to the current simulation test in real time to ensure the test interaction and test effectiveness.

[0082] In a possible implementation manner, after performing simulation scenario design based on the plurality of first target dangerous elements and the plurality of target safety elements to generate an autonomous driving dangerous simulation scenario, the generating method further includes:

[0083] The information corresponding to the first target dangerous element in the autonomous driving dangerous simulation scene and the test results of the autonomous driving dangerous simulation scene are stored.

[0084] Here, the information corresponding to the first target dangerous element in the autonomous driving dangerous simulation scenario and the test results of the autonomous driving dangerous simulation scenario are stored to ensure that all data in the simulation test process are traceable.

[0085] In this application, based on a large amount of real natural driving data, by calculating the failure probability of the scene and setting the threshold acceptable to the enterprise, the customized test scene is a dangerous test scene. The dangerous scene has high authenticity and the enterprise has high flexibility in setting the threshold of the dangerous scene. The dangerous scene based on the probability function is disassembled to determine the dangerous elements and safety elements. The elements are screened at multiple levels and categories to accurately identify the most critical and core elements that make the autonomous driving system fail. The real-time setting of scene elements based on dangerous elements has clear test purposes, increases the danger of the scene, has strong test interactivity, and ensures the effectiveness of the test. And all data in the simulation test process are traceable. It can more easily support the setting of emergencies, dangerous events or extreme events (Cornercases), and can detect the recognition and response of the tested autonomous driving system. For example, during the test, a boulder suddenly falls 15m ahead. In the safe environment of simulation testing, the recognition and response of the tested autonomous driving system to emergencies, dangerous events or extreme events (Corner cases) are examined to ensure the coverage of the test scene setting, thereby improving the safety of the autonomous driving system.

[0086] The embodiment of the present application provides a method for generating an autonomous driving simulation scene, the method comprising: obtaining multiple sample autonomous driving simulation scenes, calculating the failure probability of the multiple sample autonomous driving simulation scenes based on a probability function, and determining a sample autonomous driving dangerous simulation scene in the multiple sample autonomous driving simulation scenes; determining a dangerous element set and a safe element set based on the multiple sample autonomous driving dangerous simulation scenes; selecting multiple first target dangerous elements and multiple target safe elements from the dangerous element set and the safe element set respectively based on the simulation scene setting requirements, performing simulation scene design based on the multiple first target dangerous elements and the multiple target safe elements, and generating an autonomous driving dangerous simulation scene. By designing a simulation test of a dangerous scene based on dangerous elements, it is more in line with the actual state of the autonomous driving simulation scene, can effectively ensure the accuracy of the scene design and implementation results, and can more easily support the setting of emergencies, dangerous events or extreme events, so as to detect the recognition and response of the tested autonomous driving algorithm.

[0087] See also Figure 2 , Figure 3 , Figure 2 One of the structural schematic diagrams of a device for generating an autonomous driving simulation scene provided in an embodiment of the present application; Figure 3 This is a second structural diagram of a method and device for generating an autonomous driving simulation scene provided in an embodiment of the present application. Figure 3 As shown in , the generating device 200 includes:

[0088] The dangerous scene determination module 210 is used to obtain a plurality of sample autonomous driving simulation scenes, calculate the failure probability of the plurality of sample autonomous driving simulation scenes based on a probability function, and determine a sample autonomous driving dangerous simulation scene among the plurality of sample autonomous driving simulation scenes;

[0089] An element splitting module 220, configured to determine a dangerous element set and a safe element set based on a plurality of sample autonomous driving dangerous simulation scenarios;

[0090] The design module 230 is used to select a plurality of first target hazardous elements and a plurality of target safety elements from the hazardous element set and the safety element set respectively based on the simulation scenario setting requirements, perform simulation scenario design based on the plurality of first target hazardous elements and the plurality of target safety elements, and generate an autonomous driving hazardous simulation scenario.

[0091] Further, when the dangerous scene determination module 210 is used to calculate the failure probability of the plurality of sample autonomous driving simulation scenes based on the probability function and determine the sample autonomous driving dangerous simulation scene among the plurality of sample autonomous driving simulation scenes, the dangerous scene determination module 210 is specifically used to:

[0092] Calculating the scene evaluation index in each of the sample autonomous driving simulation scenes based on the probability function to determine the scene failure probability of each of the sample autonomous driving simulation scenes;

[0093] Based on the scenario failure probability and a preset failure probability threshold, a sample autonomous driving dangerous simulation scenario is determined among multiple sample autonomous driving simulation scenarios.

[0094] Further, when the element splitting module 220 is used to determine the dangerous element set and the safe element set based on the plurality of sample autonomous driving dangerous simulation scenarios, the element splitting module 220 is specifically used to:

[0095] Extracting the danger simulation parameter information in the sample autonomous driving danger simulation scene based on a plurality of preset danger categories to determine the danger elements, and forming the danger element set based on the danger elements under each danger category; wherein the danger category includes a weather category, a target category, and an obstacle category;

[0096] Based on multiple preset safety categories, the safety simulation parameter information in the sample autonomous driving danger simulation scenario is extracted to determine the safety elements, and the safety element set is formed based on the safety elements under each safety category; wherein the safety categories include road categories, road surface categories, lane feature categories, and road edge categories.

[0097] Furthermore, the simulation scene setting requirements include:

[0098] For the first target dangerous element, at least one of the first target dangerous elements is selected from the dangerous element set and one dangerous element of the same category is selected, and the sum of the first target dangerous elements accounts for less than or equal to the first proportion in all elements of the autonomous driving simulation scene;

[0099] For the target safety element, at least one target safety element is selected from the safety elements corresponding to each safety category, and the sum of the target safety elements accounts for a proportion greater than or equal to a second proportion in all elements of the autonomous driving simulation scene; wherein the second proportion is greater than the first proportion, and the sum of the second proportion and the first proportion is 1.

[0100] Further, when the design module 230 is used to perform simulation scenario design based on the plurality of first target dangerous elements and the plurality of target safety elements to generate an autonomous driving dangerous simulation scenario, the design module 230 is specifically used to:

[0101] Based on the target safety element, the autonomous driving algorithm is controlled to perform a simulation test. After the autonomous driving algorithm controls the vehicle to be in a stable operating state, the first target danger element is added to the simulation test to generate the autonomous driving danger simulation scenario.

[0102] Furthermore, the design module 230 is used to:

[0103] Continue to screen out a second target dangerous element from the dangerous element set, and add the second target dangerous element to the current simulation test in real time.

[0104] Further, such as Figure 3 As shown, the generating device 200 further includes a storage module 240, and the storage module 240 is used to:

[0105] The information corresponding to the first target dangerous element in the autonomous driving dangerous simulation scene and the test results of the autonomous driving dangerous simulation scene are stored.

[0106] The embodiment of the present application provides a device for generating a dangerous simulation scenario for autonomous driving, the generating device comprising: a dangerous scenario determination module, used to obtain multiple sample autonomous driving simulation scenarios, calculate the failure probability of the multiple sample autonomous driving simulation scenarios based on a probability function, and determine the sample dangerous simulation scenario for autonomous driving in the multiple sample autonomous driving simulation scenarios; an element splitting module, used to determine a dangerous element set and a safe element set based on the multiple sample dangerous simulation scenarios for autonomous driving; a design module, used to select multiple first target dangerous elements and multiple target safe elements from the dangerous element set and the safe element set respectively based on the simulation scenario setting requirements, perform simulation scenario design based on the multiple first target dangerous elements and the multiple target safe elements, and generate a dangerous simulation scenario for autonomous driving. By designing a simulation test of a dangerous scenario based on dangerous elements, it is more in line with the real state of the autonomous driving simulation scenario, can effectively ensure the accuracy of the scenario design and implementation results, can more easily support the setting of emergencies, dangerous events or extreme events, so as to detect the recognition and response of the tested autonomous driving algorithm.

[0107] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 4 As shown in , the electronic device 400 includes a processor 410 , a memory 420 and a bus 430 .

[0108] The memory 420 stores machine-readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, the above-mentioned Figure 1 The steps of the method for generating the autonomous driving simulation scene in the method embodiment shown, the specific implementation method can be found in the method embodiment, and will not be repeated here.

[0109] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the computer program can execute the above-mentioned Figure 1 The steps of the method for generating the autonomous driving simulation scene in the method embodiment shown, the specific implementation method can be found in the method embodiment, and will not be repeated here.

[0110] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0111] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0112] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0113] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0114] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0115] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-mentioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-mentioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for generating an autonomous driving simulation scene, characterized in that: The generation method comprises: Acquire multiple sample autonomous driving simulation scenarios, calculate the failure probabilities of the multiple sample autonomous driving simulation scenarios based on a probability function, and determine a sample autonomous driving dangerous simulation scenario among the multiple sample autonomous driving simulation scenarios; Based on the plurality of sample autonomous driving danger simulation scenarios, determining a set of danger elements and a set of safety elements; Based on the simulation scenario setting requirements, a plurality of first target dangerous elements and a plurality of target safety elements are respectively selected from the dangerous element set and the safety element set, and a simulation scenario is designed based on the plurality of first target dangerous elements and the plurality of target safety elements to generate an autonomous driving dangerous simulation scenario; The calculating the failure probability of the plurality of sample autonomous driving simulation scenarios based on the probability function to determine a sample autonomous driving dangerous simulation scenario among the plurality of sample autonomous driving simulation scenarios includes: Calculating the scene evaluation index in each of the sample autonomous driving simulation scenes based on the probability function to determine the scene failure probability of each of the sample autonomous driving simulation scenes; The probability of scenario failure is determined by the following formula: in, is the scenario failure probability of scenario X, Prob is the probability function, fail scenario is the failed simulation scenario, represents an evaluation index of the autonomous driving system in the simulation scenario X. is the probability density function of the dynamic driving simulation scenario X; Based on the scenario failure probability and a preset failure probability threshold, determining a sample autonomous driving dangerous simulation scenario among a plurality of sample autonomous driving simulation scenarios; The simulation scene setting requirements include: For the first target dangerous element, at least one of the first target dangerous elements is selected from the dangerous element set and one dangerous element of the same category is selected, and the sum of the first target dangerous elements accounts for less than or equal to the first proportion in all elements of the autonomous driving simulation scene; For the target safety element, at least one target safety element is selected from the safety elements corresponding to each safety category, and the sum of the target safety elements accounts for a proportion greater than or equal to a second proportion in all elements of the autonomous driving simulation scene; wherein the second proportion is greater than the first proportion, and the sum of the second proportion and the first proportion is 1.

2. The generation method according to claim 1, characterized in that: The determining of a set of dangerous elements and a set of safe elements based on the plurality of sample autonomous driving dangerous simulation scenarios includes: Extracting the danger simulation parameter information in the sample autonomous driving danger simulation scene based on a plurality of preset danger categories to determine the danger elements, and forming the danger element set based on the danger elements under each danger category; wherein the danger category includes a weather category, a target category, and an obstacle category; Based on multiple preset safety categories, the safety simulation parameter information in the sample autonomous driving danger simulation scenario is extracted to determine the safety elements, and the safety element set is formed based on the safety elements under each safety category; wherein the safety categories include road categories, road surface categories, lane feature categories, and road edge categories.

3. The generation method according to claim 1, characterized in that: The designing of simulation scenarios based on the plurality of the first target dangerous elements and the plurality of the target safety elements to generate an autonomous driving dangerous simulation scenario includes: Based on the target safety element, the autonomous driving algorithm is controlled to perform a simulation test. After the autonomous driving algorithm controls the vehicle to be in a stable operating state, the first target danger element is added to the simulation test to generate the autonomous driving danger simulation scenario.

4. The generation method according to claim 3, characterized in that: In the simulation test of controlling the automatic driving algorithm based on the target safety element, after the automatic driving algorithm controls the vehicle to be in a stable operating state, adding the first target dangerous element to the simulation test to generate the automatic driving dangerous simulation scenario, the generation method further includes: Continue to screen out a second target dangerous element from the dangerous element set, and add the second target dangerous element to the current simulation test in real time.

5. The generation method according to claim 1, characterized in that: After designing a simulation scenario based on the plurality of first target dangerous elements and the plurality of target safety elements to generate an autonomous driving dangerous simulation scenario, the generating method further includes: The information corresponding to the first target dangerous element in the autonomous driving dangerous simulation scene and the test results of the autonomous driving dangerous simulation scene are stored.

6. A device for generating a dangerous simulation scene for autonomous driving, characterized in that: The generating device comprises: a dangerous scenario determination module, configured to obtain a plurality of sample autonomous driving simulation scenarios, calculate the failure probabilities of the plurality of sample autonomous driving simulation scenarios based on a probability function, and determine a sample autonomous driving dangerous simulation scenario among the plurality of sample autonomous driving simulation scenarios; An element splitting module, used to determine a dangerous element set and a safe element set based on a plurality of sample autonomous driving dangerous simulation scenarios; A design module is used to select a plurality of first target dangerous elements and a plurality of target safety elements from the dangerous element set and the safety element set respectively based on the simulation scenario setting requirements, perform simulation scenario design based on the plurality of first target dangerous elements and the plurality of target safety elements, and generate an autonomous driving dangerous simulation scenario; When the dangerous scene determination module is used to calculate the failure probability of the plurality of sample autonomous driving simulation scenes based on the probability function and determine the sample autonomous driving dangerous simulation scene among the plurality of sample autonomous driving simulation scenes, the dangerous scene determination module is specifically used to: Calculating the scene evaluation index in each of the sample autonomous driving simulation scenes based on the probability function to determine the scene failure probability of each of the sample autonomous driving simulation scenes; The probability of scenario failure is determined by the following formula: in, is the scenario failure probability of scenario X, Prob is the probability function, fail scenario is the failed simulation scenario, represents an evaluation index of the autonomous driving system in the simulation scenario X. is the probability density function of the dynamic driving simulation scenario X; Based on the scenario failure probability and a preset failure probability threshold, determining a sample autonomous driving dangerous simulation scenario among a plurality of sample autonomous driving simulation scenarios; The simulation scene setting requirements include: For the first target dangerous element, at least one of the first target dangerous elements is selected from the dangerous element set and one dangerous element of the same category is selected, and the sum of the first target dangerous elements accounts for less than or equal to the first proportion in all elements of the autonomous driving simulation scene; For the target safety element, at least one target safety element is selected from the safety elements corresponding to each safety category, and the sum of the target safety elements accounts for a proportion greater than or equal to a second proportion in all elements of the autonomous driving simulation scene; wherein the second proportion is greater than the first proportion, and the sum of the second proportion and the first proportion is 1.

7. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the method for generating an autonomous driving simulation scene as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for generating an autonomous driving simulation scene as described in any one of claims 1 to 5 are executed.

Images