A smart driving system scene risk assessment method, device, medium and equipment

Abstract

The application provides a kind of intelligent driving system scene risk assessment method, device, medium and equipment, based on multiple influence factors in target scene, the coupling coefficient of target scene is calculated;Based on the coupling coefficient, the basic risk weight vector is corrected, and the composite weight vector is obtained;Based on the composite weight vector and the failure rate vector of each risk source, the comprehensive damage degree of target scene is calculated;Based on the comprehensive damage degree, the scene frequency of target scene, the test confidence coefficient and the test pass rate of intelligent driving system under target scene, the scene risk value of target scene is calculated;Based on the scene risk value of all target scenes, the scene risk level of intelligent driving system is determined;By setting the influence factor of scene, the classification of scene is refined, and the coupling coefficient is calculated by using influence factor to correct the risk weight, so as to improve the weight of dominant risk source in high-risk scene, thereby improving the accuracy of risk assessment.

Description

Technical Field

[0001] This application relates to the field of intelligent driving assessment technology, specifically to a method, device, medium, and equipment for assessing scenario risks in an intelligent driving system. Background Technology

[0002] End-to-end intelligent driving systems directly map raw sensor inputs into vehicle control commands through deep neural networks. Compared to traditional modular architectures (perception-decision-planning-control), this reduces cascading errors and information transmission losses, demonstrating significant advantages in complex driving scenarios. However, integrating the perception, decision-making, and control processes into a highly nonlinear black-box model results in a lack of explicit interpretability in its internal feature representation and logical reasoning, posing a fundamental challenge to system safety assessment.

[0003] To address the evaluation needs of end-to-end intelligent driving systems, various safety and scenario testing methods have been developed. However, existing evaluation methods still have the following significant shortcomings in practical applications: First, the environmental description is static and lacks granularity. Existing methods typically abstract environmental conditions into discrete labels, such as treating rainy days as a single scene category, but cannot distinguish the different impacts of light rain and heavy rain on sensor physical characteristics (such as camera transmittance and lidar point cloud noise) and control output quality. This coarse-grained environmental characterization is difficult to support the safe assessment of the performance degradation boundary of end-to-end systems under continuous environmental gradients.

[0004] Second, risk calculations often assume linear independence and ignore coupling effects. Current risk assessment models generally assume that risk sources such as technical failures, environmental interference, and operational deviations are independent of each other, and use a linear superposition method to calculate the comprehensive risk value. However, in real driving environments, risk factors often exhibit strong coupling characteristics: severe weather leads to a decrease in the signal-to-noise ratio of sensors, which in turn causes errors in feature extraction by the perception module. These errors, when propagated to the control layer, may cause planning decisions to deviate significantly from expectations.

[0005] Third, the scenario definition is vague and fails to distinguish between normal environments and temporary anomalies. Existing scenario classification methods treat road usage as static or quasi-static, lacking explicit modeling of temporary abnormal events (such as road construction, traffic accident debris, temporary traffic control, etc.). Because existing assessment methods fail to clearly distinguish the boundary between normal and abnormal scenarios, they are unable to systematically cover such high-risk real road conditions.

[0006] Fourth, the weighting allocation is rigid and lacks dynamic adaptability. Current assessment systems generally use expert scoring or the analytic hierarchy process (AHP) to predetermine fixed weights for each assessment indicator and apply them uniformly across all scenarios. However, the dominant risk factors differ significantly across different scenarios: for example, when visibility suddenly decreases, perceived reliability should be given higher weight; while at less structured urban intersections, the assessment weights for behavioral prediction and interaction logic need to be increased accordingly. Fixed weighting mechanisms cannot adjust the assessment focus in real time according to dynamic changes in the scenario, limiting the sensitivity of the assessment results to real risks.

[0007] In summary, existing end-to-end intelligent driving system safety assessment methods have significant shortcomings in terms of environmental representation granularity, risk coupling modeling, scene dynamic classification, and weight adaptation. There is an urgent need for a risk assessment technology solution that can balance model interpretability and real-time dynamic response. Summary of the Invention

[0008] To address the aforementioned technical problems, this application is proposed. Embodiments of this application provide a method, apparatus, medium, and device for assessing scenario risks in an intelligent driving system.

[0009] According to one aspect of this application, a method for scenario risk assessment of an intelligent driving system is provided, comprising: calculating a coupling coefficient of the target scenario based on multiple influencing factors in the target scenario; wherein the influencing factors are scenario factors that affect the performance of the intelligent driving system, and the coupling coefficient represents the degree of influence of the target scenario on the intelligent driving system; modifying a basic risk weight vector based on the coupling coefficient to obtain a composite weight vector; wherein each element in the basic risk weight vector represents the risk weight of each risk source of the intelligent driving system; calculating the comprehensive damage degree of the target scenario based on the composite weight vector and the failure rate vector of each risk source; wherein the failure rate vector represents the probability that the corresponding risk source causes the intelligent driving system to fail; calculating a scenario risk value of the target scenario based on the comprehensive damage degree, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario; and determining the scenario risk level of the intelligent driving system based on the scenario risk values ​​of all the target scenarios.

[0010] In one embodiment, calculating the coupling coefficient of the target scene based on multiple influencing factors in the target scene includes: calculating the coupling coefficient based on the comprehensive baseline strength of multiple influencing factors in the target scene and the sensitivity correlation coefficient between each influencing factor and the interactive behavior in the target scene.

[0011] In one embodiment, calculating the coupling coefficient based on the comprehensive baseline strength of multiple influencing factors in the target scene, the sensitivity correlation coefficient between each influencing factor and the interactive behavior in the target scene, includes: the formula for calculating the coupling coefficient is: ;in, For the first s The coupling coefficient of a target scenario. For the first k The overall baseline strength of each influencing factor For the first k The first impact factor and the first s Sensitivity correlation coefficient of interactive behaviors in a target scenario K This represents the total number of influence factors.

[0012] In one embodiment, the step of modifying the basic risk weight vector based on the coupling coefficient to obtain the composite weight vector includes: using an expert matrix to score the basic risk weight vector; using the coupling coefficient to modify each element in the basic risk weight vector and performing a normalization operation to obtain the composite weight vector.

[0013] In one embodiment, calculating the overall damage level of the target scenario based on the composite weight vector and the failure rate vectors of each risk source includes: the formula for calculating the overall damage level is: ;in, For the first s The overall degree of damage in each target scenario For the first s A composite weight vector for each target scenario. For the first s Failure rate vector of each risk source in a target scenario For the first s The interaction matrix of each risk source in a target scenario represents the probability of coupling failure between each risk source.

[0014] In one embodiment, calculating the scenario risk value of the target scenario based on the overall damage level, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario includes: the calculation formula for the scenario risk value is: ;in, For the first s The scenario risk value for each target scenario. For the first s Scene frequency of each target scenario For intelligent driving systems in the first s The pass rate for tests in each target scenario For the first sRobustness adjustment factor for each target scenario For the first s The overall degree of damage in each target scenario For the first s The test confidence coefficient for each target scenario.

[0015] In one embodiment, the formula for calculating the test confidence coefficient of the target scene is: ;in, This refers to the actual test mileage. The target baseline mileage.

[0016] According to another aspect of this application, a scenario risk assessment device for an intelligent driving system is provided, comprising: a coupling coefficient calculation module, used to calculate the coupling coefficient of the target scenario based on multiple influencing factors in the target scenario; wherein the influencing factors are scenario factors that affect the performance of the intelligent driving system, and the coupling coefficient represents the degree of influence of the target scenario on the intelligent driving system; a risk weight correction module, used to correct a basic risk weight vector based on the coupling coefficient to obtain a composite weight vector; wherein each element in the basic risk weight vector represents the risk weight of each risk source of the intelligent driving system; a damage degree calculation module, used to calculate the comprehensive damage degree of the target scenario based on the composite weight vector and the failure rate vector of each risk source; wherein the failure rate vector represents the probability that the corresponding risk source causes the intelligent driving system to fail; a scenario risk calculation module, used to calculate the scenario risk value of the target scenario based on the comprehensive damage degree, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario; and a risk level determination module, used to determine the risk level of the target scenario based on the scenario risk value of the target scenario.

[0017] According to another aspect of this application, a computer-readable storage medium is provided, the storage medium storing a computer program for performing any of the methods described above.

[0018] According to another aspect of this application, an electronic device is provided, comprising: a processor; a memory for storing processor-executable instructions; the processor being configured to perform any of the methods described above.

[0019] This application provides a method, apparatus, medium, and device for scene risk assessment of intelligent driving systems. Based on multiple influencing factors in the target scene, the coupling coefficient of the target scene is calculated. The influencing factors are scene factors that affect the performance of the intelligent driving system, and the coupling coefficient represents the degree of influence of the target scene on the intelligent driving system. Based on the coupling coefficient, the basic risk weight vector is corrected to obtain a composite weight vector. Each element in the basic risk weight vector represents the risk weight of each risk source in the intelligent driving system. Based on the composite weight vector and the failure rate vector of each risk source, the comprehensive damage level of the target scene is calculated. The failure rate vector represents the probability that the corresponding risk source will cause the intelligent driving system to fail. Based on the comprehensive damage level, the scene frequency and test confidence coefficient of the target scene, and the test pass rate of the intelligent driving system in the target scene, the scene risk value of the target scene is calculated. Based on the scene risk values ​​of all target scenes, the scene risk level of the intelligent driving system is determined. By setting the influencing factors of the scene, the scene classification is refined. The coupling coefficient calculated using the influencing factors is used to correct the risk weights, thereby increasing the weight of the dominant risk source in high-risk scenes and improving the accuracy of risk assessment. Attached Figure Description

[0020] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0021] Figure 1 This is a flowchart illustrating an exemplary embodiment of the intelligent driving system scenario risk assessment method provided in this application.

[0022] Figure 2 This is a schematic diagram of the structure of an intelligent driving system scenario risk assessment device provided in an exemplary embodiment of this application.

[0023] Figure 3 This is a structural diagram of an electronic device provided in an exemplary embodiment of this application. Detailed Implementation

[0024] Hereinafter, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0025] Figure 1 This is a flowchart illustrating an exemplary embodiment of the intelligent driving system scenario risk assessment method provided in this application. Figure 1 As shown, the scenario risk assessment method for this intelligent driving system includes the following steps: Step 110: Calculate the coupling coefficient of the target scene based on multiple influencing factors in the target scene.

[0026] In this application, the influencing factors are the scene factors that affect the performance of the intelligent driving system, and the coupling coefficient represents the degree of influence of the target scene on the intelligent driving system. The target scene is constructed by creating multiple influencing factors and calculating the coupling coefficient based on these factors. The target scene comprises a four-layer orthogonal architecture: a basic geometry layer, a physical environment layer, a temporary anomaly layer, and a traffic participant layer. The basic geometry layer defines the road structure, including straight roads, curves, intersections, and ramps. The physical environment layer defines normal natural attributes, including weather, lighting, and road surface adhesion. The temporary anomaly layer defines sudden or temporary states, including road construction, road flooding, traffic accidents, and traffic congestion. The traffic participant layer defines interactive objects, including large vehicles, small vehicles, pedestrians, two-wheeled vehicles, and their interactive behaviors.

[0027] Step 120: Based on the coupling coefficient, the basic risk weight vector is modified to obtain the composite weight vector.

[0028] In this application, each element of the basic risk weight vector represents the risk weight of each risk source in the intelligent driving system. The basic risk weight vector is modified using a coupling coefficient to increase the risk weight of the dominant risk source in the risk scenario.

[0029] Step 130: Calculate the overall damage level of the target scenario based on the composite weight vector and the failure rate vector of each risk source.

[0030] The failure rate vector represents the probability that a corresponding risk source will cause the intelligent driving system to fail. This application combines the composite weight vector and the failure rate vectors of each risk source to calculate the overall damage level of the target scenario.

[0031] Step 140: Calculate the scenario risk value of the target scenario based on the comprehensive damage level, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario.

[0032] This application calculates the scenario risk value of the target scenario based on the comprehensive degree of damage, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario.

[0033] Step 150: Determine the scenario risk level of the intelligent driving system based on the scenario risk values ​​of all target scenarios.

[0034] Based on the scenario risk values ​​of all target scenarios, this application ultimately determines the scenario risk level of the intelligent driving system.

[0035] This application provides a method for scenario risk assessment of intelligent driving systems. Based on multiple influencing factors in the target scenario, the coupling coefficient of the target scenario is calculated. The influencing factors are scenario factors that affect the performance of the intelligent driving system, and the coupling coefficient represents the degree of influence of the target scenario on the intelligent driving system. Based on the coupling coefficient, the basic risk weight vector is corrected to obtain a composite weight vector. Each element in the basic risk weight vector represents the risk weight of each risk source in the intelligent driving system. Based on the composite weight vector and the failure rate vector of each risk source, the comprehensive damage level of the target scenario is calculated. The failure rate vector represents the probability that the corresponding risk source will cause the intelligent driving system to fail. Based on the comprehensive damage level, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario, the scenario risk value of the target scenario is calculated. Based on the scenario risk values ​​of all target scenarios, the scenario risk level of the intelligent driving system is determined. By setting scenario influencing factors, the scenario classification is refined. The coupling coefficient calculated using the influencing factors is used to correct the risk weights, thereby increasing the weight of the dominant risk source in high-risk scenarios and improving the accuracy of risk assessment.

[0036] In one embodiment, step 110 can be implemented by calculating the coupling coefficient based on the comprehensive baseline strength of multiple influencing factors in the target scenario and the sensitivity correlation coefficient between each influencing factor and the interactive behavior in the target scenario.

[0037] This application calculates the coupling coefficient based on the comprehensive baseline intensity of multiple influencing factors in the target scenario and the sensitivity correlation coefficient between each influencing factor and the interactive behavior in the target scenario. The comprehensive baseline intensity is obtained by multi-level probability weighted summation. For example, taking the scenario of "pedestrians crossing an intersection in rainy weather" as an example, rainfall is divided into light rain (corresponding to an intensity of 0.2), moderate rain (corresponding to an intensity of 0.4), and heavy rain (corresponding to an intensity of 0.7). The weights are determined based on the probability of rainfall in natural driving data, and the comprehensive baseline intensity of rainfall is obtained by weighted summation. The correlation coefficient between each influencing factor and the sensitivity of interactive behavior in the target scenario indicates the degree of influence of that influencing factor on the interactive behavior in the target scenario. For example, under rainy conditions, pedestrians crossing an intersection are extremely sensitive to visual obstruction. .

[0038] In one embodiment, step 110 can be implemented as follows: the formula for calculating the coupling coefficient is: ;in, For the first s The coupling coefficient of a target scenario. For the first k The overall baseline strength of each influencing factor For the first k The first impact factor and the first s Sensitivity correlation coefficient of interactive behaviors in a target scenario K This represents the total number of influence factors.

[0039] This application calculates the comprehensive baseline strength and corresponding sensitivity correlation coefficient of each influencing factor in the target scenario, and then calculates the coupling coefficient of the target scenario using the above formula. For example, the coupling coefficient of the above scenario "pedestrians crossing an intersection in rainy weather" is calculated using only rainfall as an influencing factor. 0.317 × 1.8 = 1.57.

[0040] In one embodiment, step 120 can be implemented as follows: using an expert matrix to score and obtain a basic risk weight vector; using a coupling coefficient to correct each element in the basic risk weight vector and perform a normalization operation to obtain a composite weight vector.

[0041] This application employs multiple experts to assign risk weights to various risk sources, synthesizing the scores from multiple experts to obtain a basic risk weight. Then, a coupling coefficient is used to modify each element in the basic risk weight vector, followed by normalization, to obtain a composite weight vector. Specifically, the modification formula is as follows: ,in, The normalization function is expressed as follows: , for n The 1st dimension of the vector i One element, Based on the basic risk weight vector, For example, a composite weight vector. These correspond to four risk sources: technology, operation, data, and environment. For example, the basic risk weight vector obtained from expert matrix scoring... Taking the scenario of "pedestrians crossing the road in the rain" as an example, the composite weight vector obtained after modifying each element in the basic risk weight vector using a coupling coefficient and performing a normalization operation is as follows: This shows that the weight of environmental risk sources has increased from 0.4 to 0.44, reflecting dynamic stress response capabilities.

[0042] In one embodiment, step 130 can be specifically implemented as follows: the formula for calculating the overall degree of damage is: ;in, For the first s The overall degree of damage in each target scenario For the firsts A composite weight vector for each target scenario. For the first s Failure rate vector of each risk source in a target scenario For the first s The interaction matrix of each risk source in a target scenario represents the probability of coupling failure between each risk source.

[0043] This application addresses the problem of independent risk sources in traditional risk assessment by introducing a risk source interaction matrix to characterize the coupling failure probability among four types of risk sources: technology, operation, data, and environment. Specifically, in the scenario... Below, risk source interaction matrix It is Square array (usually) (corresponding to technology, operation, data, environment): ; diagonal elements Indicates risk source Its own strength (usually set to 1 or based on independent evaluation), off-diagonal elements Indicates risk source For risk sources Influencing factors (amplification or suppression coefficients), for example, rainy weather increases noise in sensing sensors, which may increase the probability of sensing algorithm failure. Therefore, it is determined that environmental risks have an amplifying effect on technological risks. .

[0044] In one embodiment, step 140 can be specifically implemented as follows: the formula for calculating the scenario risk value is: ;in, For the first s The scenario risk value for each target scenario. For the first s Scene frequency of each target scene For intelligent driving systems in the first s The pass rate of tests in each target scenario For the first s Robustness adjustment factor for each target scenario For the first s The overall degree of damage in each target scenario For the first s The test confidence coefficient for each target scenario.

[0045] This application calculates the... sAfter determining the overall damage level of each target scenario, the scenario risk value for each target scenario is calculated using the above formula, taking into account the overall damage level, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario.

[0046] In one embodiment, step 140 can be specifically implemented as follows: the formula for calculating the test confidence coefficient of the target scenario is: ;in, This refers to the actual test mileage. The target baseline mileage.

[0047] This application calculates the test confidence coefficient for the target scenario by setting a target baseline mileage and measuring the actual test mileage. For example, the actual test mileage... kilometers, target mileage kilometers, then confidence level As the test mileage increases, The value should be close to 1 to ensure the reliability of the evaluation results.

[0048] After calculating the scenario risk value for each target scenario, this application can also sum the scenario risk values ​​of all target scenarios to obtain the overall risk index of the intelligent driving system. And according to the preset threshold The risk level of the intelligent driving system is determined by classifying it.

[0049] Figure 2 This is a schematic diagram of the structure of an intelligent driving system scenario risk assessment device provided in an exemplary embodiment of this application. Figure 2 As shown, the intelligent driving system scenario risk assessment device 20 includes: a coupling coefficient calculation module 21, used to calculate the coupling coefficient of the target scenario based on multiple influencing factors in the target scenario; wherein, the influencing factors are scenario factors that affect the performance of the intelligent driving system, and the coupling coefficient represents the degree of influence of the target scenario on the intelligent driving system; a risk weight correction module 22, used to correct the basic risk weight vector based on the coupling coefficient to obtain a composite weight vector; wherein, each element in the basic risk weight vector represents the risk weight of each risk source of the intelligent driving system; a damage degree calculation module 23, used to calculate the comprehensive damage degree of the target scenario based on the composite weight vector and the failure rate vector of each risk source; wherein, the failure rate vector represents the probability that the corresponding risk source causes the intelligent driving system to fail; a scenario risk calculation module 24, used to calculate the scenario risk value of the target scenario based on the comprehensive damage degree, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario; and a risk level determination module 25, used to determine the risk level of the target scenario based on the scenario risk value of the target scenario.

[0050] This application provides a scenario risk assessment device for an intelligent driving system. A coupling coefficient calculation module 21 calculates the coupling coefficient of the target scenario based on multiple influencing factors. The influencing factors are scenario factors that affect the performance of the intelligent driving system, and the coupling coefficient represents the degree of influence of the target scenario on the intelligent driving system. A risk weight correction module 22 corrects the basic risk weight vector based on the coupling coefficient to obtain a composite weight vector. Each element in the basic risk weight vector represents the risk weight of each risk source in the intelligent driving system. A damage degree calculation module 23 calculates the target... The overall damage level of the scenario is assessed; the failure rate vector represents the probability that the corresponding risk source will cause the intelligent driving system to fail; the scenario risk calculation module 24 calculates the scenario risk value of the target scenario based on the overall damage level, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario; the risk level determination module 25 determines the scenario risk level of the intelligent driving system based on the scenario risk values ​​of all target scenarios; the scenario classification is refined by setting the scenario influence factors, and the risk weight is corrected by using the influence factors to calculate the coupling coefficient, so as to improve the weight of the dominant risk source in high-risk scenarios, thereby improving the accuracy of risk assessment.

[0051] In one embodiment, the coupling coefficient calculation module 21 can be further configured to calculate the coupling coefficient based on the comprehensive baseline strength of multiple influencing factors in the target scene and the sensitivity correlation coefficient between each influencing factor and the interactive behavior in the target scene.

[0052] In one embodiment, the coupling coefficient calculation module 21 can be further configured such that the formula for calculating the coupling coefficient is: ;in, For the first s The coupling coefficient of a target scenario. For the first k The overall baseline strength of each influencing factor For the first k The first impact factor and the first s Sensitivity correlation coefficient of interactive behaviors in a target scenario K This represents the total number of influence factors.

[0053] In one embodiment, the risk weight correction module 22 can be further configured to: use an expert matrix to score and obtain a basic risk weight vector; use a coupling coefficient to correct each element in the basic risk weight vector and perform a normalization operation to obtain a composite weight vector.

[0054] In one embodiment, the damage degree calculation module 23 can be further configured such that the formula for calculating the overall damage degree is: ;in, For the first s The overall degree of damage in each target scenario For the first s A composite weight vector for each target scenario. For the first s Failure rate vector of each risk source in a target scenario For the first s The interaction matrix of each risk source in a target scenario represents the probability of coupling failure between each risk source.

[0055] In one embodiment, the scenario risk calculation module 24 can be further configured such that the calculation formula for the scenario risk value is: ;in, For the first s The scenario risk value for each target scenario. For the first s Scene frequency of each target scene For intelligent driving systems in the first s The pass rate of tests in each target scenario For the first s Robustness adjustment factor for each target scenario For the first s The overall degree of damage in each target scenario For the first s The test confidence coefficient for each target scenario.

[0056] In one embodiment, the scenario risk calculation module 24 can be further configured such that the calculation formula for the test confidence coefficient of the target scenario is: ;in, This refers to the actual test mileage. The target baseline mileage.

[0057] Below, for reference Figure 3 This application describes an electronic device according to embodiments thereof. The electronic device may be either or both of a first device and a second device, or a standalone device independent of them, which may communicate with the first device and the second device to receive acquired input signals from them.

[0058] Figure 3 A block diagram of an electronic device according to an embodiment of this application is illustrated.

[0059] like Figure 3 As shown, the electronic device 10 includes one or more processors 11 and memory 12.

[0060] The processor 11 may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0061] The memory 12 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 11 may execute the program instructions to implement the methods of the various embodiments of this application described above and / or other desired functions. Various contents such as input signals, signal components, and noise components may also be stored in the computer-readable storage medium.

[0062] In one example, the electronic device 10 may also include an input device 13 and an output device 14, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0063] When the electronic device is a standalone device, the input device 13 can be a communication network connector for receiving the collected input signals from the first device and the second device.

[0064] In addition, the input device 13 may also include, for example, a keyboard, a mouse, etc.

[0065] The output device 14 can output various information to the outside, including determined distance information, direction information, etc. The output device 14 may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0066] Of course, for the sake of simplicity, Figure 3 Only some of the components of the electronic device 10 relevant to this application are shown in this illustration; components such as buses, input / output interfaces, etc., are omitted. In addition, the electronic device 10 may include any other suitable components depending on the specific application.

[0067] In addition to the methods and apparatus described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0068] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0069] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0070] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0071] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0072] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0073] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0074] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0075] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A method for scenario risk assessment in an intelligent driving system, characterized in that, include: Based on multiple influencing factors in the target scenario, the coupling coefficient of the target scenario is calculated; wherein, the influencing factors are scenario factors that affect the performance of the intelligent driving system, and the coupling coefficient represents the degree of influence of the target scenario on the intelligent driving system; Based on the coupling coefficient, the basic risk weight vector is modified to obtain a composite weight vector; wherein, each element in the basic risk weight vector represents the risk weight of each risk source of the intelligent driving system. Based on the composite weight vector and the failure rate vector of each risk source, the overall damage level of the target scenario is calculated; wherein, the failure rate vector represents the probability that the corresponding risk source causes the intelligent driving system to fail; Based on the overall damage level, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario, the scenario risk value of the target scenario is calculated. Based on the scenario risk values ​​of all the target scenarios, the scenario risk level of the intelligent driving system is determined.

2. The intelligent driving system scenario risk assessment method according to claim 1, characterized in that, The calculation of the coupling coefficient of the target scene based on multiple influencing factors includes: The coupling coefficient is calculated based on the comprehensive baseline strength of multiple influencing factors in the target scenario, the sensitivity correlation coefficient between each influencing factor and the interactive behavior in the target scenario.

3. The intelligent driving system scenario risk assessment method according to claim 2, characterized in that, The calculation of the coupling coefficient, based on the comprehensive baseline strength of multiple influencing factors in the target scenario and the sensitivity correlation coefficient between each influencing factor and the interactive behavior in the target scenario, includes: The formula for calculating the coupling coefficient is: ； in, For the first s The coupling coefficient of a target scenario. For the first k The overall baseline strength of each influencing factor For the first k The first impact factor and the first s Sensitivity correlation coefficient of interactive behaviors in a target scenario K This represents the total number of influence factors.

4. The intelligent driving system scenario risk assessment method according to claim 1, characterized in that, The process of modifying the basic risk weight vector based on the coupling coefficient to obtain the composite weight vector includes: The basic risk weight vector is obtained by using an expert matrix scoring method. The coupling coefficient is used to modify each element in the basic risk weight vector and then normalized to obtain the composite weight vector.

5. The intelligent driving system scenario risk assessment method according to claim 1, characterized in that, The calculation of the overall damage level of the target scenario based on the composite weight vector and the failure rate vector of each risk source includes: The formula for calculating the overall degree of damage is as follows: ； in, For the first s The overall degree of damage in each target scenario For the first s A composite weight vector for each target scenario. For the first s Failure rate vector of each risk source in a target scenario For the first s The interaction matrix of each risk source in a target scenario represents the probability of coupling failure between each risk source.

6. The intelligent driving system scenario risk assessment method according to claim 1, characterized in that, The calculation of the scenario risk value of the target scenario based on the overall damage level, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario includes: The formula for calculating the risk value of the scenario is as follows: ； in, For the first s The scenario risk value for each target scenario. For the first s Scene frequency of each target scenario For intelligent driving systems in the first s The pass rate for tests in each target scenario For the first s Robustness adjustment factor for each target scenario For the first s The overall degree of damage in each target scenario For the first s The test confidence coefficient for each target scenario.

7. The intelligent driving system scenario risk assessment method according to claim 6, characterized in that, The formula for calculating the test confidence coefficient of the target scenario is as follows: ； in, This refers to the actual test mileage. The target baseline mileage.

8. A scenario risk assessment device for an intelligent driving system, characterized in that, include: The coupling coefficient calculation module is used to calculate the coupling coefficient of the target scenario based on multiple influencing factors in the target scenario; wherein, the influencing factors are scenario factors that affect the performance of the intelligent driving system, and the coupling coefficient represents the degree of influence of the target scenario on the intelligent driving system; The risk weight correction module is used to correct the basic risk weight vector based on the coupling coefficient to obtain a composite weight vector; wherein each element in the basic risk weight vector represents the risk weight of each risk source of the intelligent driving system. The damage severity calculation module is used to calculate the comprehensive damage severity of the target scenario based on the composite weight vector and the failure rate vector of each risk source; wherein, the failure rate vector represents the probability that the corresponding risk source causes the intelligent driving system to fail; The scenario risk calculation module is used to calculate the scenario risk value of the target scenario based on the comprehensive damage level, the scenario frequency and test confidence coefficient of the target scenario, and the test pass rate of the intelligent driving system in the target scenario. The risk level determination module is used to determine the risk level of the target scenario based on the scenario risk value of the target scenario.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program for performing the method described in any one of claims 1-7.

10. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is used to execute the method described in any one of claims 1-7.

Images