Automatic driving expected function safety scene library construction method, device and equipment

Abstract

The invention discloses an automatic driving expected function safety scene library construction method, device and equipment, and relates to the technical field of automatic driving, and the method comprises the steps: constructing an initial expected function safety scene library; determining a logic scene parameter boundary according to a design operation domain and a hazardous behavior of the automatic driving system; generating a simulation automatic driving scene according to the logic scene parameter boundary; and performing boundary value test analysis on the simulated automatic driving scene to obtain a simulated automatic driving scene meeting a preset condition, and adding the simulated automatic driving scene meeting the preset condition into the initial expected function safety scene library to obtain an automatic driving safety scene library. According to the method, the expected function safety simulation scene library is established, the scene library is continuously reused and updated in the development process of a plurality of series of vehicle types so as to perform vehicle safety verification, and the working efficiency can be improved.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a method, apparatus, and device for constructing a library of expected functional safety scenarios for autonomous driving. Background Technology

[0002] While autonomous driving technology brings immense convenience to drivers and passengers, it also poses significant challenges to vehicle safety. This is especially true for advanced autonomous driving systems (Level 3 and above), where the expected functional safety issues are particularly prominent compared to lower-level driver assistance systems. To address problems such as insufficient definition of autonomous driving functions and potential misuse by drivers, automotive development teams must establish a comprehensive expected functional safety system. Through multiple rounds of iterative development and testing, they must ensure that the risks associated with autonomous driving functions are kept at a reasonably acceptable level.

[0003] Simulation is the primary method for testing and evaluating intended functional safety. Since extensive road testing requires significant time and resources, and many scenarios are difficult to encounter in actual road testing, simulation technology is a good alternative. While some methods for automatically generating simulation scenarios exist in the field of autonomous driving, there is a lack of scenario libraries specifically designed for testing during the intended functional safety development process. This results in insufficient simulation test scenarios, requiring the creation of separate simulation test scenarios for different vehicle models, leading to low work efficiency. Summary of the Invention

[0004] In view of the above, the present invention aims to provide a method, apparatus and device for constructing a library of expected functional safety scenarios for autonomous driving, so as to solve the aforementioned technical problems.

[0005] The technical solution adopted in this invention is as follows:

[0006] This invention provides a method for constructing a library of expected functional safety scenarios for autonomous driving, including:

[0007] Build an initial library of expected functional safety scenarios;

[0008] Determine the boundaries of logical scenario parameters based on the design operating domain and hazardous behaviors of the autonomous driving system;

[0009] Based on the boundaries of the logical scenario parameters, a simulated autonomous driving scenario is generated;

[0010] Boundary value testing and analysis are performed on the simulated autonomous driving scenario to obtain a simulated autonomous driving scenario that meets the preset conditions. The simulated autonomous driving scenario that meets the preset conditions is added to the initial expected functional safety scenario library to obtain the autonomous driving safety scenario library.

[0011] Optionally, build an initial library of expected functional safety scenarios, including:

[0012] Based on existing low-level assisted driving nominal scenarios and regulatory scenarios, an initial expected functional safety scenario library is constructed.

[0013] If a low-level driver assistance scenario library has not been established, construct a library of nominal and regulatory scenarios.

[0014] The nominal and regulatory scenario library includes at least one of the following scenarios: vehicle cutting out, vehicle cutting in, emergency braking, and stable following.

[0015] Optionally, based on the design operating domain and hazardous behaviors of the autonomous driving system, the boundaries of logical scenario parameters are determined, including:

[0016] Based on the design operating domain of the autonomous driving system, determine the autonomous driving scenarios required by the vehicle;

[0017] Add the hazardous behaviors and their triggering conditions to the autonomous driving scenarios required by the vehicle to determine the boundaries of the logical scenario parameters.

[0018] Optionally, based on the logical scene parameter boundaries, a simulated autonomous driving scenario is generated, including:

[0019] The real image data and location data of the target driving scene are collected by drone aerial photography. The real image data includes at least the light intensity and the clarity of the lane lines.

[0020] Based on 3D reconstruction algorithms and scene boundary values, a BIM model containing road structure and environmental elements is generated.

[0021] The model is converted to OpenX format to obtain a simulated autonomous driving scenario.

[0022] Optionally, generating a simulated autonomous driving scenario based on the logical scenario parameter boundaries further includes:

[0023] The key parameter values ​​are obtained by randomly selecting values ​​from the real image data according to a Poisson distribution;

[0024] If the value of a key parameter exceeds the boundary of the logical scene parameter, then the boundary of the logical scene parameter is set as the scene boundary value.

[0025] Optionally, the key parameter values ​​include:

[0026] The clarity of the first lane line is >30%, the clarity of the second lane line is >10%, and the light intensity is >3000 lux.

[0027] Optionally, boundary value testing and analysis are performed on the simulated autonomous driving scenario to obtain a simulated autonomous driving scenario that meets preset conditions, including:

[0028] Input the simulated autonomous driving scenario into the simulation test software, set the safe driving conditions for autonomous driving, and filter out the simulated autonomous driving scenarios that meet the preset conditions.

[0029] The present invention also provides an apparatus for constructing a library of expected functional safety scenarios for autonomous driving, comprising:

[0030] The simulation scenario generation module is used to construct an initial expected functional safety scenario library; determine the logical scenario parameter boundaries based on the design operating domain and hazardous behaviors of the autonomous driving system; and generate simulated autonomous driving scenarios based on the logical scenario parameter boundaries.

[0031] The scenario library construction module is used to perform boundary value test analysis on the simulated autonomous driving scenario to obtain simulated autonomous driving scenarios that meet preset conditions, and add the simulated autonomous driving scenarios that meet preset conditions to the initial expected functional safety scenario library to obtain the autonomous driving safety scenario library.

[0032] The present invention also provides a computing device, comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above.

[0033] The present invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above.

[0034] The above-described solution of the present invention has at least the following beneficial effects:

[0035] The above-described solution of the present invention involves: constructing an initial expected functional safety scenario library; determining logical scenario parameter boundaries based on the design operating domain and hazardous behaviors of the autonomous driving system; generating simulated autonomous driving scenarios based on the logical scenario parameter boundaries; performing boundary value testing and analysis on the simulated autonomous driving scenarios to obtain simulated autonomous driving scenarios that meet preset conditions; and adding the simulated autonomous driving scenarios that meet the preset conditions to the initial expected functional safety scenario library to obtain an autonomous driving safety scenario library. The solution of the present invention, by establishing an expected functional safety simulation scenario library, allows for continuous reuse and updating of this scenario library during the development of multiple vehicle series, thereby improving work efficiency. Attached Figure Description

[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below with reference to the accompanying drawings, wherein:

[0037] Figure 1 A flowchart illustrating the method for constructing a library of expected functional safety scenarios for autonomous driving provided in this embodiment of the invention.

[0038] Figure 2A schematic diagram of the module of the autonomous driving expected functional safety scenario library construction device provided in the embodiments of the present invention. Detailed Implementation

[0039] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0040] This invention proposes an embodiment of a method for constructing a library of expected functional safety scenarios for autonomous driving. Specifically, as follows: Figure 1 As shown, it includes:

[0041] Step 11: Construct an initial library of expected functional safety scenarios;

[0042] Step 12: Determine the logical scenario parameter boundaries based on the design operating domain and hazardous behaviors of the autonomous driving system;

[0043] Step 13: Generate a simulated autonomous driving scenario based on the logical scenario parameter boundaries;

[0044] Step 14: Perform boundary value test analysis on the simulated autonomous driving scenario to obtain a simulated autonomous driving scenario that meets the preset conditions, and add the simulated autonomous driving scenario that meets the preset conditions to the initial expected functional safety scenario library to obtain the autonomous driving safety scenario library.

[0045] In this embodiment, based on low-level assisted driving scenarios, the product design operation domain (ODD) is analyzed to fully consider the characteristics of scenarios that should be covered, such as insufficient perception, insufficient control, and misuse by personnel. By combining random parameter generalization and real-world modeling methods, the expected functional safety scenario library is gradually built for the enterprise's autonomous driving system development and testing team.

[0046] When a company first undertakes the design of a high-level autonomous driving system, it begins to build a library of expected functional safety scenarios based on existing low-level assisted driving nominal scenarios and regulatory scenarios. If an enterprise has not established a low-level assisted driving scenario library, it should establish a nominal scenario and regulatory scenario library based on the current intelligent driving standard system and commonly used domestic and international traffic regulations. This library should cover common driving scenarios such as cutting out from the front, cutting in, emergency braking, and stable following. The enterprise should determine the product design and operating domain of the autonomous driving system and identify the scope of simulation test scenarios that should be covered based on this domain. It should conduct analysis activities such as defining expected functional safety, identifying hazard identification, triggering condition identification, and verification and confirmation strategies to identify and summarize functional safety deficiencies and potential human misuse. Based on the above analysis, it should identify hazardous scenarios. Furthermore, it should extract hazardous scenarios from data collected from real-world open road tests of other similar vehicles and define the categories of functional scenarios containing hazards. Through data mining and expert analysis, it should determine the key parameters and parameter ranges of logical scenarios. Based on the parameter ranges of logical scenarios, it should conduct boundary value test analysis. By establishing specific scenarios and conducting tests, it should confirm whether the scenario belongs to the scenario required for product development. If it does, it should be included in the expected functional safety scenario library; otherwise, the parameter range should be redefined and the above tests repeated. Methods for establishing scenarios include random parameter generalization, direct modeling using simulation software, and technical re-conversion formats based on Building Information Modeling (BIM) + real-world modeling. The expected functional safety analysis of the above process must be assessed to determine if it meets release requirements. If it does, the product development is complete; otherwise, the entire development process should be iterated until release requirements are met. When developing the next autonomous driving system product, the previously accumulated expected functional safety scenario library can be directly used as the initial scenario library, with new scenarios added as needed during the development of the new product.

[0047] Specifically, in generating a simulated autonomous driving scenario, real image data and location data of the target driving scene are collected by drone aerial photography. The real image data includes at least the light intensity and the clarity of lane lines. Based on the 3D reconstruction algorithm and scene boundary values, a BIM model containing road structure and environmental elements is generated. The model is then converted into OpenX format to obtain the simulated autonomous driving scenario.

[0048] The simulated autonomous driving scenario is input into the simulation test software, and the conditions for safe autonomous driving are set, such as light intensity parameters and lane line clarity parameters. Simulated autonomous driving scenarios that meet the preset conditions are then selected and included in the scenario library.

[0049] The proposed method for constructing a simulated scenario library for autonomous driving with expected functional safety in this embodiment establishes such a library. This library can be continuously reused and updated during the development of multiple vehicle series. It can continuously identify new hazardous scenarios that need to be tested in the autonomous driving development of automotive companies, reduce the space of unknown scenarios, improve product safety, and increase work efficiency.

[0050] The following description uses two specific embodiments as examples.

[0051] Example 1:

[0052] A company is developing a new passenger vehicle equipped with an autonomous driving system. The process of building its expected functional safety scenario library is as follows.

[0053] Step 1: The autonomous driving development team has experience in developing L2-level assisted driving and has established a library of nominal scenarios and traffic regulation scenarios, covering the content required by traffic regulations and standards such as the Road Traffic Safety Law, GB / T39323, GB / T38186, and GB / T39265.

[0054] Step 2: The ODD of this autonomous driving system is limited to highways. The function can only be activated under clear weather, daytime conditions, and when road facilities are in good condition. Therefore, the test scenario is also limited to the above conditions.

[0055] Step 3: Through the analysis of expected functional safety, multiple triggering conditions for functional deficiencies and human misuse were identified, and the residual risk acceptance criterion was determined to be: with a 95% confidence level, the safety of the autonomous driving system is no less than that of a qualified human driver.

[0056] Step 4: Conduct scenario analysis on the hazardous behaviors and triggering conditions identified in Step 3. Taking a lack of perception as an example: "The autonomous driving system relies on the forward-facing camera to identify lane lines and achieve lane keeping function. When the lane lines are unclear, there is a risk of lane departure." For this functional deficiency, the scenario for testing is determined to be: highways with insufficient lane line clarity.

[0057] Step 5: Taking into account expert experience and real-world road test results of similar models, the identified logical scenario is: "The clarity of lane lines transitions from being clearly visible on both sides to being clear on one side and only 10% clear on the other side, and finally to being clear on both sides with 10% clarity."

[0058] Step 6: Create several specific scenarios to conduct boundary value testing and analysis. First, create a highway scenario. To enhance the realism of the simulation, select a real highway section for drone aerial photography and real-world modeling. The selected drone is equipped with high-definition cameras, LiDAR, and other sensors. The preset flight route is to conduct aerial photography along the starting point to the end point of the highway section. The drone's flight direction is kept consistent with the route direction in real time, collecting image data and location information of the highway and its surroundings.

[0059] To avoid potential data loss or errors from a single acquisition, three aerial flights were conducted. The acquired image data was then examined, and the highest quality images from each flight period were selected. Computer vision and 3D reconstruction algorithms were used to process and analyze the selected image data, calculating the position of each pixel in 3D space to construct a 3D model of the highway segment. The model included elements such as straight sections, curves, traffic signs, guardrails, and surrounding trees. The output was a model file directly readable by BIM software, which was then converted to OpenX format and imported into the simulation testing software.

[0060] Then, in the highway scenario model, the illumination intensity and the clarity of the lane lines on both sides are randomly selected according to a Poisson distribution. If the parameter boundaries defined in step 5 are exceeded, they are modified to boundary values, generating multiple simulation scenarios. Three are selected for illustration. Scenario 1: Both lane lines are clear, illumination intensity is 3600 lux; Scenario 2: One side is clear, the other side has only 10% clarity, illumination intensity is 5200 lux; Scenario 3: Both lane lines have 10% clarity, illumination intensity is 12000 lux. Tests were conducted on each scenario. The test results showed that vehicles in scenarios 1 and 2 could maintain their lanes, while in scenario 3, the vehicle deviated significantly, prompting the driver to take over. After adjustments and testing, it was determined that "one lane line clarity greater than 30%, the other side clarity greater than 10%, and illumination intensity greater than 3000 lux" are required for the autonomous driving system to maintain its lane normally. This logical scenario was included in the company's expected functional safety logical scenario library.

[0061] Step 7: After the first round of expected functional safety analysis, the residual risk acceptance criteria could not be met. Therefore, subsequent iterative analysis was carried out. After the final completion and release of the SOTIF analysis, a total of 23 new expected functional safety scenarios were added.

[0062] Step 8: In the initial stage of developing the next product, the initial scenario library is analyzed using the original nominal scenario, traffic regulation scenario, and the 23 new expected functional safety scenarios added to this product, which effectively improves development efficiency.

[0063] Example 2:

[0064] A company is developing a new MPV equipped with an autonomous driving system. The process of building its expected functional safety scenario library is as follows.

[0065] Step 1: The autonomous driving development team has not yet established a low-level assisted driving scenario library. First, based on the requirements of traffic regulations, they established a total of 420 traffic regulation test scenarios, including vehicle starting, stopping, speed limit sign recognition, pedestrian crossing recognition, traffic light recognition, following other vehicles, and sudden appearance of obstacles in front.

[0066] Step 2: The ODD of this autonomous driving system is limited to urban expressways. The function can only be activated under sunny, daytime conditions and when the road facilities are in good condition. Therefore, the test scenario is also limited to the above conditions.

[0067] Step 3: Through the analysis of expected functional safety, multiple triggering conditions for functional deficiencies and human misuse were identified, and the residual risk acceptance criterion was determined to be: with a 99% confidence level, the safety of the autonomous driving system is no less than that of a qualified human driver.

[0068] Step 4: Perform scenario analysis on the hazardous behaviors and triggering conditions identified in Step 3. For example, consider a perceived deficiency: "At dusk, the forward and lateral vision sensors are affected by illumination levels, causing them to miss identifying other vehicles in the forward or lateral direction, posing a collision risk." For this deficiency, the scenario for testing is determined to be: urban expressways with insufficient light intensity at dusk.

[0069] Step 5: The main parameter of this functional scenario is the time when the vehicle is running, which affects the perception performance through the light intensity (denoted as I). The weather is limited to "sunny day, horizontal visibility greater than 500m". After analysis, the logical scenario identified is: "within the range of 100°~120° east longitude in the Northern Hemisphere, time 17:00-21:00".

[0070] Step 6: Create several specific scenarios to conduct boundary value testing and analysis. First, create an urban expressway scenario. Select a real urban expressway segment. A BIM-based real-world model of this segment already exists; convert it to an OpenX format and import it into the simulation testing software. Then, in the above scenario model, randomly select timestamps according to a uniform distribution, record the illumination intensity at each moment, add traffic flow, and generate multiple simulation scenarios. Three scenarios are selected for illustration: Scenario 1: Time 18:40, illumination intensity 2300 lux; Scenario 2: Time 19:50, illumination intensity 560 lux; Scenario 3: Time 17:10, illumination intensity 5100 lux. Tests were conducted on each scenario. The results showed that Scenario 1 allowed normal driving, while scenarios 2 and 3 resulted in vehicle collisions. Further analysis revealed that in Scenario 2, the overall low illumination intensity prevented the detection of lateral vehicles. In Scenario 3, the collision area was obstructed by a high-rise building, resulting in an illumination intensity of less than 600 lux in that local area, causing the collision. After multiple adjustments and tests, it was determined that only an area with light intensity greater than 800 lux and no obstructions at dusk can enable the autonomous driving system to maintain its lane normally. This logical scenario was then incorporated into the company's expected functional safety logical scenario library.

[0071] Step 7: After the first round of expected functional safety analysis, the residual risk acceptance criteria could not be met. Therefore, subsequent iterative analysis was carried out. After the final completion and release of the SOTIF analysis, a total of 34 new expected functional safety scenarios were added.

[0072] Step 8: At the beginning of the development of the next product, the initial traffic regulations scenario, plus the 34 new expected functional safety scenarios added to this product, will be used as the initial scenario library for analysis, which will effectively improve development efficiency.

[0073] like Figure 2 As shown, this embodiment of the invention also provides an apparatus 20 for constructing a library of expected functional safety scenarios for autonomous driving, comprising:

[0074] The simulation scenario generation module 21 is used to construct an initial expected functional safety scenario library; determine the logical scenario parameter boundaries based on the design operating domain and hazardous behaviors of the autonomous driving system; and generate simulated autonomous driving scenarios based on the logical scenario parameter boundaries.

[0075] The scenario library construction module 22 is used to perform boundary value test analysis on the simulated autonomous driving scenario to obtain the simulated autonomous driving scenario that meets the preset conditions, and add the simulated autonomous driving scenario that meets the preset conditions to the initial expected functional safety scenario library to obtain the autonomous driving safety scenario library.

[0076] Optionally, build an initial library of expected functional safety scenarios, including:

[0077] Based on existing low-level assisted driving nominal scenarios and regulatory scenarios, an initial expected functional safety scenario library is constructed.

[0078] If a low-level driver assistance scenario library has not been established, construct a library of nominal and regulatory scenarios.

[0079] The nominal and regulatory scenario library includes at least one of the following scenarios: vehicle cutting out, vehicle cutting in, emergency braking, and stable following.

[0080] Optionally, based on the design operating domain and hazardous behaviors of the autonomous driving system, the boundaries of logical scenario parameters are determined, including:

[0081] Based on the design operating domain of the autonomous driving system, determine the autonomous driving scenarios required by the vehicle;

[0082] Add the hazardous behaviors and their triggering conditions to the autonomous driving scenarios required by the vehicle to determine the boundaries of the logical scenario parameters.

[0083] Optionally, based on the logical scene parameter boundaries, a simulated autonomous driving scenario is generated, including:

[0084] The real image data and location data of the target driving scene are collected by drone aerial photography. The real image data includes at least the light intensity and the clarity of the lane lines.

[0085] Based on 3D reconstruction algorithms and scene boundary values, a BIM model containing road structure and environmental elements is generated.

[0086] The model is converted to OpenX format to obtain a simulated autonomous driving scenario.

[0087] Optionally, generating a simulated autonomous driving scenario based on the logical scenario parameter boundaries further includes:

[0088] The key parameter values ​​are obtained by randomly selecting values ​​from the real image data according to a Poisson distribution;

[0089] If the value of a key parameter exceeds the boundary of the logical scene parameter, then the boundary of the logical scene parameter is set as the scene boundary value.

[0090] Optionally, the key parameter values ​​include:

[0091] The clarity of the first lane line is >30%, the clarity of the second lane line is >10%, and the light intensity is >3000 lux.

[0092] Optionally, boundary value testing and analysis are performed on the simulated autonomous driving scenario to obtain a simulated autonomous driving scenario that meets preset conditions, including:

[0093] Input the simulated autonomous driving scenario into the simulation test software, set the safe driving conditions for autonomous driving, and filter out the simulated autonomous driving scenarios that meet the preset conditions.

[0094] It should be noted that this device is the same as the method described above. All implementations in the above method embodiments are applicable to the embodiments of this device and can achieve the same technical effect.

[0095] An embodiment of the present invention also provides a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described in the above embodiments. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0096] In this embodiment of the invention, a computer-readable storage medium is also provided, storing instructions that, when executed on a computer, cause the computer to perform the method described in the above embodiments. All implementations of the methods described in the above embodiments are applicable to this embodiment and can achieve the same technical effect.

[0097] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0098] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0099] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0100] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0101] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0102] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0103] Furthermore, it should be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Moreover, the steps performing the above series of processes can naturally be executed in the order described, but are not necessarily required to be executed in chronological order; some steps can be executed in parallel or independently of each other. Those skilled in the art will understand that all or any step or component of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or a combination thereof. This is something that those skilled in the art can achieve by using their basic programming skills after reading the description of the present invention.

[0104] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing device. The computing device can be a known general-purpose device. Therefore, the object of the present invention can also be achieved simply by providing a program product containing program code implementing the method or apparatus. That is, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent to the present invention. Furthermore, the steps performing the above series of processes can naturally be performed in the order described, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel or independently of each other.

[0105] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for constructing a library of expected functional safety scenarios for autonomous driving, characterized in that, include: Build an initial library of expected functional safety scenarios; Determine the boundaries of logical scenario parameters based on the design operating domain and hazardous behaviors of the autonomous driving system; Based on the boundaries of the logical scenario parameters, a simulated autonomous driving scenario is generated; Boundary value testing and analysis are performed on the simulated autonomous driving scenario to obtain a simulated autonomous driving scenario that meets the preset conditions. The simulated autonomous driving scenario that meets the preset conditions is added to the initial expected functional safety scenario library to obtain the autonomous driving safety scenario library.

2. The method for constructing a library of expected functional safety scenarios for autonomous driving according to claim 1, characterized in that, Construct an initial library of expected functional safety scenarios, including: Based on existing low-level assisted driving nominal scenarios and regulatory scenarios, an initial expected functional safety scenario library is constructed. If a low-level driver assistance scenario library has not been established, construct a library of nominal and regulatory scenarios. The nominal and regulatory scenario library includes at least one of the following scenarios: vehicle cutting out, vehicle cutting in, emergency braking, and stable following.

3. The method for constructing a library of expected functional safety scenarios for autonomous driving according to claim 1, characterized in that, Based on the design operating domain and hazardous behaviors of the autonomous driving system, determine the boundaries of logical scenario parameters, including: Based on the design operating domain of the autonomous driving system, determine the autonomous driving scenarios required by the vehicle; Add the hazardous behaviors and their triggering conditions to the autonomous driving scenarios required by the vehicle to determine the boundaries of the logical scenario parameters.

4. The method for constructing a library of expected functional safety scenarios for autonomous driving according to claim 3, characterized in that, Based on the logical scenario parameter boundaries, a simulated autonomous driving scenario is generated, including: The real image data and location data of the target driving scene are collected by drone aerial photography. The real image data includes at least the light intensity and the clarity of the lane lines. Based on 3D reconstruction algorithms and scene boundary values, a BIM model containing road structure and environmental elements is generated. The model is converted to OpenX format to obtain a simulated autonomous driving scenario.

5. The method for constructing a library of expected functional safety scenarios for autonomous driving according to claim 4, characterized in that, Based on the logical scenario parameter boundaries, generating a simulated autonomous driving scenario further includes: The key parameter values ​​are obtained by randomly selecting values ​​from the real image data according to a Poisson distribution; If the value of a key parameter exceeds the boundary of the logical scene parameter, then the boundary of the logical scene parameter is set as the scene boundary value.

6. The method for constructing a library of expected functional safety scenarios for autonomous driving according to claim 5, characterized in that, The key parameter values ​​include: The clarity of the first lane line is >30%, the clarity of the second lane line is >10%, and the light intensity is >3000 lux.

7. The method for constructing a library of expected functional safety scenarios for autonomous driving according to claim 1, characterized in that, Boundary value testing and analysis are performed on the simulated autonomous driving scenario to obtain a simulated autonomous driving scenario that meets preset conditions, including: Input the simulated autonomous driving scenario into the simulation test software, set the safe driving conditions for autonomous driving, and filter out the simulated autonomous driving scenarios that meet the preset conditions.

8. A device for constructing a library of expected functional safety scenarios for autonomous driving, characterized in that, include: The simulation scenario generation module is used to build an initial expected functional safety scenario library; Determine the boundaries of logical scenario parameters based on the design operating domain and hazardous behaviors of the autonomous driving system; Based on the boundaries of the logical scenario parameters, a simulated autonomous driving scenario is generated; The scenario library construction module is used to perform boundary value test analysis on the simulated autonomous driving scenario to obtain simulated autonomous driving scenarios that meet preset conditions, and add the simulated autonomous driving scenarios that meet preset conditions to the initial expected functional safety scenario library to obtain the autonomous driving safety scenario library.

9. A computing device, characterized in that, include: A processor, a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The system stores instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images