Layered decoupling method and device for test-oriented automatic driving vehicle

Abstract

The invention discloses a test-oriented automatic driving vehicle hierarchical decoupling method and device. The method comprises the following steps: constructing a test system behavior map; performing hierarchical decoupling on the test system behavior map to obtain a module total importance score set; and processing the module total importance score set to obtain key evaluation factor information. Through step-by-step decoupling of the physical layer, the algorithm layer and the functional layer, accurate testing and fault positioning in a complex coupling scene are realized, the problems of low testing efficiency and insufficient confidence in automatic driving testing are solved, and the method is suitable for accurate testing and fault tracing of the automatic driving vehicle in a multi-module coupling scene.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving testing and verification technology, and in particular to a layered decoupling method and apparatus for autonomous vehicles oriented towards testing. Background Technology

[0002] Scenario-based virtual simulation testing has a significant driving force and development potential for the development of autonomous driving. However, related research is currently in its early stages, and there is still a significant gap before establishing a systematic theory, method, and platform for autonomous driving testing. Specifically, the shortcomings include: high coupling between different levels of autonomous vehicles, lack of targeted testing; high complexity and strong randomness of vehicle driving environments; current test scenarios are difficult to reflect the real high complexity and strong randomness of traffic environments; the lack of a high-confidence scenario construction method for multi-level and multi-difficulty levels of autonomous driving systems; and the absence of a fast scenario search method corresponding to the test objectives, resulting in low testing efficiency and insufficient confidence.

[0003] This invention achieves accurate testing and fault location in complex coupled scenarios by decoupling the physical layer, algorithm layer, and functional layer step by step. It is suitable for solving the problem of accurate testing and fault tracing in multi-module coupled scenarios. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a layered decoupling method and device for autonomous vehicles for testing. By decoupling the physical layer, algorithm layer and functional layer step by step, it realizes accurate testing and fault location in complex coupled scenarios, solves the problems of low testing efficiency and insufficient confidence in autonomous driving testing, and is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0005] To address the aforementioned technical problems, a first aspect of this invention discloses a layered decoupling method for autonomous vehicles oriented towards testing, the method comprising: S1, Construct a behavioral graph of the test system; S2, perform hierarchical decoupling on the behavior graph of the test system to obtain the total importance score set of the modules; S3, process the evaluation element information set to obtain key evaluation element information.

[0006] As an optional implementation, in the first aspect of the present invention, the construction of the test system behavior graph includes: S11, Obtain the information source node; S12, Based on the information source node, collect the node data timestamp and node spatial location information; S13, using the node data timestamp and the node spatial location information, a system input layer is constructed; S14. Based on the functional modules, the calling relationship of the functional modules, and the data flow relationship, a directed graph structure is constructed. S15. Based on the system function matching of the system input layer and the directed graph structure, a test system behavior graph is constructed.

[0007] As an optional implementation, in the first aspect of the present invention, the step of hierarchically decoupling the behavior graph of the test system to obtain a set of evaluation element information includes: S21, The behavior graph of the test system is identified and processed to obtain a module behavior graph set; The module behavior graph set includes several module behavior graphs; S22, Process the module behavior graph set to obtain the module feature information entropy set; S23, process the module feature information entropy set to obtain the module total importance score set.

[0008] As an optional implementation, in the first aspect of the present invention, processing the module behavior graph set to obtain a module feature information entropy set includes: S221, The module behavior graph set is parsed to obtain all module behavior graphs; S222, construct a feature vector from any of the module behavior graphs to obtain a module feature vector; S223, perform information entropy calculation on the module feature vector to obtain the module feature information entropy; S224, combine all the module feature information entropy in sequence to obtain the module feature information entropy set.

[0009] As an optional implementation, in the first aspect of the present invention, the feature vector construction expression is: , in, Indicates the first Feature vectors of each functional module; Indicates the index of the functional module; The number of dimensions representing the features of the functional module; Indicates the relationship with the functional module In the Statistical or relational features of the behavioral graph dimensions of each of the aforementioned functional modules; The expression for calculating and processing information entropy is: , in, Indicates the functional module Information entropy; Indicates the functional module No. The probability of a feature occurring in historical operation.

[0010] As an optional implementation, in the first aspect of the present invention, processing the module feature information entropy set to obtain the module total importance score set includes: S231, perform mutual information extraction processing on the module feature information entropy set to construct a mutual information matrix; The mutual information extraction and processing expression is: , in, Indicates functional modules With modules Mutual information value; Indicates the functional module With the aforementioned functional modules Joint probability of features; Indicates the functional module Marginal probability distribution of features; Indicates the functional module Marginal probability distribution of features; S232, Based on the mutual information matrix, a module decoupling diagram is constructed; S233, Based on the module feature information entropy set and the module decoupling diagram, the key module information is determined; S234, process the key module information to obtain the total importance score set of the module.

[0011] As an optional implementation, in the first aspect of the present invention, the processing of the key module information to obtain the total importance score set of the modules includes: S2341, Obtain the module heterogeneity feature information of the key module information; S2342, Normalize the heterogeneous feature information of the module to obtain normalized heterogeneous feature information; S2343, Based on the normalized heterogeneous feature information, construct a unified feature matrix; S2344, Process the unified feature matrix to obtain the total importance score set of the module.

[0012] A second aspect of this invention discloses a hierarchical decoupling device for autonomous vehicles oriented towards testing, the device comprising: a behavior graph construction module, an importance score extraction module, and a key evaluation element extraction module; The behavior graph construction module is used to construct the behavior graph of the test system; The evaluation element extraction module is used to perform hierarchical decoupling of the behavior graph of the test system to obtain an evaluation element information set; The importance score extraction module is used to process the evaluation element information set to obtain the module's total importance score information; The behavior graph construction module, the assessment element extraction module, and the key assessment element extraction module are sequentially connected.

[0013] A third aspect of the present invention discloses another layered decoupling device for autonomous vehicles oriented towards testing, the device comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the layered decoupling method for test-oriented autonomous vehicles disclosed in the first aspect of the present invention.

[0014] The fourth aspect of the present invention discloses a computer-readable storage medium storing computer instructions, which, when invoked, execute some or all of the steps in the test-oriented layered decoupling method for autonomous vehicles disclosed in the first aspect of the present invention.

[0015] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: In this embodiment of the invention, by decoupling the physical layer, algorithm layer, and functional layer step by step, accurate testing and fault location in complex coupled scenarios are achieved, solving the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios. Attached Figure Description

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

[0017] Figure 1 This is a schematic diagram of a scenario for an autonomous vehicle testing system provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating a layered decoupling method for autonomous vehicles oriented towards testing, as disclosed in an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of a layered decoupling device for autonomous vehicles oriented towards testing, as disclosed in an embodiment of the present invention. Figure 4 This is a schematic diagram of another layered decoupling device for autonomous vehicles oriented towards testing, disclosed in an embodiment of the present invention. Detailed Implementation

[0018] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0020] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] In this application, the term "exemplary" is used to mean "used as an example, illustration, or description." Any embodiment described as "exemplary" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use this application. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that this application can be made without using these specific details. In other instances, well-known structures and processes are not described in detail to avoid obscuring the description of this application with unnecessary detail. Therefore, this application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0022] It should be noted that since the method in this application embodiment is executed in a computer device, the processing objects of each computer device exist in the form of data or information, such as time, which is essentially time information. It is understood that if size, quantity, position, etc. are mentioned in subsequent embodiments, they are all corresponding data that exist so that the computer device can process them. Specific details will not be elaborated here.

[0023] It should be noted that the artificial intelligence-related technologies that may be involved in this application will be briefly described. Artificial intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. Artificial intelligence is the study of the design principles and implementation methods of various intelligent machines, enabling machines to have the functions of perception, reasoning, and decision-making.

[0024] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies primarily include computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0025] Computer vision (CV) is a science that studies how to enable machines to "see." More specifically, it refers to machine vision, which uses cameras and computers to replace human eyes in recognizing and measuring targets, and then performs image processing to create images more suitable for human observation or transmission to instruments. As a scientific discipline, computer vision studies related theories and technologies, attempting to build artificial intelligence systems capable of extracting information from images or multidimensional data. Computer vision technologies typically include image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content / behavior recognition, 3D object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous localization and mapping (SLAM), and common biometric recognition technologies such as facial recognition and fingerprint recognition.

[0026] This application provides a layered decoupling method, apparatus, computer device, and computer-readable storage medium for autonomous vehicles oriented towards testing, which will be described in detail below.

[0027] Please see Figure 1 , Figure 1 This is a schematic diagram of a scenario for an autonomous vehicle testing system provided in an embodiment of this application. The system may include a computer device 100, which integrates a layered decoupling device for autonomous vehicles designed for testing, such as... Figure 1 Computer equipment in the country.

[0028] In this embodiment, the computer device 100 can be a standalone server, a server network, or a server cluster. For example, the computer device 100 described in this embodiment includes, but is not limited to, a computer, a network host, a single network server, a set of multiple network servers, or a cloud server composed of multiple servers. The cloud server is composed of a large number of computers or network servers based on cloud computing.

[0029] It is understood that the computer device 100 used in the embodiments of this application can be a device that includes both receiving and transmitting hardware, that is, a device having receiving and transmitting hardware capable of performing bidirectional communication on a bidirectional communication link. Such a device may include: cellular or other communication devices having a single-line display, a multi-line display, or a cellular or other communication device without a multi-line display. Specifically, the computer device 100 may be a desktop terminal or a mobile terminal, and may also be one of a mobile phone, tablet computer, laptop computer, etc.

[0030] Those skilled in the art will understand that Figure 1 The application environment shown is merely one application scenario for the solution in this application and does not constitute a limitation on the application scenario of the solution in this application. Other application environments may include those that are more specific to this application. Figure 1 The number of computer devices shown is more or less, for example Figure 1 Only one computer device is shown in the diagram. It is understood that the system may also include one or more other services, which are not limited here.

[0031] In addition, such as Figure 1 As shown, the communication link evaluation system may also include a memory 200 for storing test system parameter data, as well as test system behavior graph data and test result data.

[0032] It should be noted that, Figure 1The schematic diagram of the autonomous vehicle testing system shown is merely an example. The layered decoupling device and scenario for autonomous vehicles oriented to testing described in this application are for the purpose of more clearly illustrating the technical solutions of this application and do not constitute a limitation on the technical solutions provided in this application. As those skilled in the art will know, with the evolution of autonomous vehicle testing systems and the emergence of new business scenarios, the technical solutions provided in this application are also applicable to similar technical problems.

[0033] This invention discloses a layered decoupling method and apparatus for autonomous vehicles in testing scenarios. Through step-by-step decoupling of the physical layer, algorithm layer, and functional layer, it achieves accurate testing and fault location in complex coupled scenarios, solving the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios. Detailed descriptions follow.

[0034] Example 1 Please see Figure 2 , Figure 2 This is a flowchart illustrating a layered decoupling method for autonomous vehicles oriented towards testing, as disclosed in an embodiment of the present invention. Figure 2 The described layered decoupling method for autonomous vehicles oriented towards testing is applied to autonomous vehicle testing systems, such as local servers or cloud servers used in autonomous vehicle testing systems; however, this embodiment of the invention is not limited to these applications. Figure 2 As shown, this layered decoupling method for autonomous vehicles geared towards testing may include the following operations: S1, Construct a behavioral graph of the test system; S2, perform hierarchical decoupling on the behavior graph of the test system to obtain the total importance score set of the modules; S3, process the total importance score set of the module to obtain key evaluation element information.

[0035] As can be seen, the layered decoupling method for autonomous vehicles oriented towards testing described in the embodiments of the present invention obtains key evaluation element information by constructing and processing the behavior graph of the test system. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer. It solves the problems of low testing efficiency and insufficient confidence in autonomous driving testing, and is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0036] In an optional embodiment, constructing the test system behavior graph includes: S11, Obtain the information source node; It should be noted that the information source node is an in-vehicle sensing device, including but not limited to in-vehicle LiDAR, in-vehicle camera, in-vehicle millimeter-wave radar and in-vehicle IMU; S12, Based on the information source node, collect the node data timestamp and node spatial location information; S13, using the node data timestamp and the node spatial location information, a system input layer is constructed; It should be noted that the construction of the system input layer means that the system input layer is constructed using the information source node and its corresponding node data timestamp and node spatial location information; It should be noted that, in this embodiment, for a certain L4 level autonomous driving vehicle system, the communication logs, call chains and data flow paths of its various on-board sensor modules, including but not limited to LiDAR, cameras, millimeter-wave radar and IMU, and control modules, including but not limited to path planning module, behavior decision module and trajectory control module, are collected in typical operating scenarios. S14. Based on the functional modules, the calling relationship of the functional modules, and the data flow relationship, a directed graph structure is constructed. It should be noted that the constructed directed graph structure represents a structure where functional modules are nodes, and the data transmission relationships, control transmission relationships, and attributes of the functional modules are edges. It should be noted that the functional modules include automatic obstacle avoidance function, automatic emergency braking function, intelligent parking function, and vehicle networking function; S15, Based on the system function matching of the system input layer and the directed graph structure, a test system behavior graph is constructed; It should be noted that, in this embodiment, the starting node of the test system behavior graph includes perception source nodes such as LiDAR and cameras, the terminal node includes the vehicle control execution module, and the intermediate nodes of the graph include perception fusion, behavior decision-making and trajectory planning modules; each edge in the test system behavior graph carries attributes such as call frequency and latency information. It should be noted that in the behavior graph of the test system, the starting node and the terminal node correspond to the vehicle-mounted equipment and the actuator, respectively; intermediate nodes include, but are not limited to, the perception processing module, the path planning module, and the motion control module; logical relationships are derived by identifying the data interaction middleware and message routing structure; system functions are extracted through path pattern matching; for example, if there is "target recognition"... risk assessment The "braking control" path determines whether the automatic emergency braking function is supported. The resulting test system behavior graph includes functional module hierarchy, communication structure, operation path and feature dependency information, providing support for subsequent coupling analysis and evaluation element extraction.

[0037] As can be seen, by implementing the layered decoupling method for autonomous vehicles oriented towards testing as described in the embodiments of the present invention, a test system behavior graph is constructed, laying the foundation for subsequent processing. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer. It solves the problems of low testing efficiency and insufficient confidence in autonomous driving testing and is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0038] In an optional embodiment, the hierarchical decoupling of the test system behavior graph to obtain the evaluation element information set includes: S21, The behavior graph of the test system is identified and processed to obtain a module behavior graph set; The module behavior graph set includes several module behavior graphs; S22, Process the module behavior graph set to obtain the module feature information entropy set; S23, process the module feature information entropy set to obtain the module total importance score set.

[0039] As can be seen, implementing the layered decoupling process for autonomous vehicles oriented towards testing as described in the embodiments of the present invention lays the foundation. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer, solving the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0040] In an optional embodiment, processing the module behavior graph set to obtain a module feature information entropy set includes: S221, The module behavior graph set is parsed to obtain all module behavior graphs; It should be noted that the parsing process refers to extracting the module behavior graph according to the field type; S222, construct a feature vector from any of the module behavior graphs to obtain a module feature vector; S223, perform information entropy calculation on the module feature vector to obtain the module feature information entropy; S224, combine all the module feature information entropy in sequence to obtain the module feature information entropy set.

[0041] As can be seen, the hierarchical decoupling method for autonomous vehicles oriented towards testing described in this embodiment of the invention processes the module behavior graph set to obtain the module feature information entropy set, laying the foundation for subsequent processing. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer, solving the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0042] In an optional embodiment, the feature vector construction expression is: , in, Indicates the first Feature vectors of each functional module; Indicates the index of the functional module; The number of dimensions representing the features of the functional module; Indicates the relationship with the functional module In the Statistical or relational features of the behavioral graph dimensions of each of the aforementioned functional modules; The expression for calculating and processing information entropy is: , in, Indicates the functional module Information entropy; Indicates the functional module No. The probability of a feature occurring in historical operation.

[0043] As can be seen, the hierarchical decoupling method for autonomous vehicles oriented towards testing described in this embodiment of the invention utilizes the feature vector to construct expressions and the information entropy calculation processing expressions to process the module behavior graph set, thereby obtaining the module feature information entropy set, which lays the foundation for subsequent processing. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer, solving the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0044] In an optional embodiment, processing the module feature information entropy set to obtain the module total importance score set includes: S231, perform mutual information extraction processing on the module feature information entropy set to construct a mutual information matrix; The mutual information extraction and processing expression is: , in, Indicates functional modules With modules Mutual information value; Indicates the functional module With the aforementioned functional modules Joint probability of features; Indicates the functional module Marginal probability distribution of features; Indicates the functional module Marginal probability distribution of features; It should be noted that the mutual information mentioned The larger the value, the stronger the information dependency between modules and the tighter the coupling relationship. It should be noted that the mutual information matrix obtained by constructing the mutual information matrix means that an information matrix is ​​constructed from the mutual information of all the module pairs. matrix ,in, Indicates the number of functional modules; S232, Based on the mutual information matrix, a module decoupling diagram is constructed; It should be noted that the module decoupling diagram constructed based on the mutual information matrix represents: The mutual information matrix is ​​processed to transform it into an adjacency graph structure, thus constructing a module decoupling graph: The mutual information processing expression is: , in, Indicates the coupling coefficient; It should be noted that when the coupling coefficients are... When, it indicates that the functional module and the functional module It is considered that a strong coupling relationship exists; when the coefficient of the coupling relationship is... When, it indicates the functional module With the aforementioned functional modules Considered as decoupled independent functional modules; It should be noted that, based on the aforementioned module decoupling graph structure, graph connectivity analysis can be performed to identify functional clusters and execute automatic layering. Ultimately, a set of mutually coupled module clusters and a set of independently operable decoupled modules are identified. S233, Based on the module feature information entropy set and the module decoupling diagram, the key module information is determined; S234, process the key module information to obtain the total importance score set of the module.

[0045] As can be seen, by implementing the layered decoupling method for autonomous vehicles oriented towards testing described in the embodiments of the present invention, the module feature information entropy set is processed to obtain the evaluation element information set, laying the foundation for subsequent processing. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer, solving the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0046] In an optional embodiment, determining the key module information based on the module feature information entropy set and the module decoupling graph includes: S2331, Based on the module feature information entropy set, calculate and obtain the average information entropy and the standard deviation of information entropy; S2332, The average information entropy and the standard deviation of information entropy are processed using the recognition threshold calculation model to obtain the module recognition threshold; The expression for the identification threshold calculation model is as follows: , in, This indicates the threshold value for the module to identify; This represents the average value of the information entropy; This represents the standard deviation of the information entropy; Indicates the adjustment coefficient; It should be noted that, in this embodiment, the adjustment coefficient is set to 1; S2333, Based on the module identification threshold and the module decoupling diagram, all functional modules are judged and processed to obtain key module information; It should be noted that the aforementioned judgment process means that when the entropy of the module feature information corresponding to the functional module is greater than the module identification threshold, the functional module is determined to be a key module.

[0047] As can be seen, by implementing the hierarchical decoupling method for autonomous vehicles for testing described in the embodiments of the present invention, the key module information is determined based on the module feature information entropy set, laying the foundation for subsequent processing. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer, solving the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0048] In an optional embodiment, the processing of the key module information and the module decoupling diagram to obtain the total importance score set of the modules includes: S2341, Obtain the module heterogeneity feature information of the key module information; It should be noted that the NetworkX graph structure processing tool is used to process the module behavior graph to obtain the module heterogeneous feature information; S2342, Normalize the heterogeneous feature information of the module to obtain normalized heterogeneous feature information; It should be noted that the normalization expression is as follows: , in, This represents the normalized heterogeneous feature information of functional module i; Indicates functional modules Module heterogeneous feature information; This represents the maximum value among all the heterogeneous feature information of all modules; This represents the minimum value among all the heterogeneous feature information of the modules; S2343, Based on the normalized heterogeneous feature information, construct a unified feature matrix; It should be noted that the construction of the unified feature matrix means that the rows represent functional modules and the columns represent the normalized heterogeneous feature information corresponding to the functional modules, and the unified feature matrix is ​​constructed accordingly. S2344, Process the unified feature matrix to obtain the total importance score set of the module.

[0049] As can be seen, by implementing the layered decoupling method for autonomous vehicles oriented towards testing described in the embodiments of the present invention, the key module information is processed to obtain the evaluation element information set, laying the foundation for subsequent processing. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer, solving the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0050] In an optional embodiment, the processing of the unified feature matrix to obtain the total importance score set of the modules includes: S23441, Based on the unified feature matrix, the rate of change of the dimension-reduced feature space is extracted; The expression for the rate of change of the dimension-reduced feature space is: , in, Indicates functional modules The rate of change of the dimension-reduced feature space; Indicates the length of the sampling time window; Indicates the sampling time; Indicates the functional module In the The uniform eigenvalue at each moment; Indicates the functional module Characteristic mean; S23442, obtain the module propagation depth value; It should be noted that obtaining the module propagation depth value means remapping the functional module to the module behavior graph, identifying its functional path, and statistically obtaining the module propagation depth value. It should be noted that the module propagation depth value is used to represent the shortest path length of the functional module from the starting node to the current node in the functional path; It should be noted that the module propagation depth value is used to quantify its decision chain dependency; the smaller the module propagation depth value, the closer the module is to the decision point in the functional graph, and the greater its influence. S23443, using the module importance calculation model, the structure centrality score, the rate of change of the reduced feature space, and the module propagation depth value are processed to obtain the total module importance score; The expression for the module importance calculation model is as follows: , , in, Indicates the functional module The total importance score of the module; Indicates the functional module The module propagation depth value; Indicates the functional module Structural centrality score; Indicates the weighting coefficient for the rate of change; Indicates the propagation depth weighting coefficient; Indicates the structural centrality weight coefficient; This represents the maximum rate of change of the dimension-reduced feature space across all functional modules; This represents the maximum value among the module propagation depth values; S23444, Combine all the total importance scores of the modules to obtain the set of total importance scores for the modules; It should be noted that the aforementioned combination processing refers to combining according to the ordinal number of the functional modules; As can be seen, by implementing the layered decoupling method for autonomous vehicles oriented towards testing described in the embodiments of the present invention, the unified feature matrix is ​​processed to obtain the total importance score set of the modules, laying the foundation for subsequent processing. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer, solving the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0051] In an optional embodiment, the processing of the total importance score set of the module to obtain key evaluation element information includes: S31, Sort all functional modules in descending order of their total importance score to obtain a set of functional module order; S32, select the first k functional modules as key evaluation elements to obtain key evaluation element information; It should be noted that in this embodiment, the value of k is set to 5; It should be noted that the value of k in this embodiment can be dynamically adjusted according to the total number of modules and the score distribution.

[0052] As can be seen, by implementing the layered decoupling method for autonomous vehicles oriented towards testing described in the embodiments of the present invention, the total importance score set of the modules is processed to obtain key evaluation element information, thus realizing the acquisition of key evaluation element information. This method achieves accurate testing and fault location in complex coupled scenarios through step-by-step decoupling of the physical layer, algorithm layer, and functional layer, and solves the problems of low testing efficiency and insufficient confidence in autonomous driving testing. It is suitable for accurate testing and fault tracing of autonomous vehicles in multi-module coupled scenarios.

[0053] Example 2 Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a layered decoupling device for autonomous vehicles oriented towards testing, as disclosed in an embodiment of the present invention. Figure 3 The described test-oriented layered decoupling device for autonomous vehicles is applied in autonomous vehicle testing systems, such as local servers or cloud servers used in these systems. Figure 3 As shown, the device may include: a behavior graph construction module 101, an importance score extraction module 102, and a key assessment element extraction module 103; The behavior graph construction module 101 is used to construct the behavior graph of the test system; The evaluation element extraction module 102 is used to perform hierarchical decoupling on the behavior graph of the test system to obtain an evaluation element information set. The importance score extraction module 103 is used to process the evaluation element information set to obtain the total importance score information of the module. The behavior graph construction module 101, the assessment element extraction module 102, and the key assessment element extraction module 103 are sequentially connected.

[0054] The device provided in this embodiment has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0055] Example 3 Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a layered decoupling device for autonomous vehicles oriented towards testing, as disclosed in an embodiment of the present invention. Figure 4 The described test-oriented layered decoupling device for autonomous vehicles is applied in autonomous vehicle testing systems, such as local servers or cloud servers used in these systems. Figure 4 As shown, the device may include: Memory 201 storing executable program code; It should be noted that the memory 201 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 various contents such as input multi-source image data may also be stored in the computer-readable storage medium. Processor 202 coupled to memory 201; It should be noted that the processor 202 may be a central processing unit (CPU) or other processing unit with data processing capability and / or instruction execution capability; the processor 202 may run the program instructions to implement the test-oriented layered decoupling method for autonomous vehicles and / or other desired functions as described in the embodiments of this disclosure above. The processor 202 calls the executable program code stored in the memory 201 to execute the steps in the test-oriented layered decoupling method for autonomous vehicles described in Embodiment 1.

[0056] Example 4 This invention discloses a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program causes a computer to execute the steps in the test-oriented layered decoupling method for autonomous vehicles described in Embodiment 1.

[0057] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform the steps in the test-oriented layered decoupling method for autonomous vehicles described in Embodiment 1.

[0058] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0059] It should be noted that all calculation expressions or mathematical functions in the embodiments of the present invention have undergone dimensionless processing of the variables involved before calculation.

[0060] It should be noted that in all the calculation expressions or mathematical functions in the embodiments of the present invention, the values ​​of the input independent variables all meet the reasonable requirements of the input value range of the calculation expression or mathematical function, and can ensure that the calculation expression or mathematical function can be calculated smoothly without violating physical laws or mathematical rules.

[0061] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

[0062] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platform, or of course by hardware. Based on this understanding, the above-mentioned technical solution, or the part that contributes to the existing technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically erasable programmable read-only memory (EEPROM), compact-disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0063] Finally, it should be noted that the layered decoupling method and apparatus for autonomous vehicles oriented to testing disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A layered decoupling method for test-oriented autonomous vehicles, characterized in that, The method includes: S1, Construct a behavioral graph of the test system; S2, perform hierarchical decoupling on the behavior graph of the test system to obtain the total importance score set of the modules; S3, process the total importance score set of the module to obtain key evaluation element information.

2. The layered decoupling method for test-oriented autonomous vehicles according to claim 1, characterized in that, The construction of the test system behavior graph includes: S11, Obtain the information source node; S12, Based on the information source node, collect the node data timestamp and node spatial location information; S13, using the node data timestamp and the node spatial location information, a system input layer is constructed; S14. Based on the functional modules, the calling relationship of the functional modules, and the data flow relationship, a directed graph structure is constructed. S15. Based on the system function matching of the system input layer and the directed graph structure, a test system behavior graph is constructed.

3. The layered decoupling method for test-oriented autonomous vehicles according to claim 1, characterized in that, The hierarchical decoupling of the behavior graph of the test system yields a set of total importance scores for each module, including: S21, The behavior graph of the test system is identified and processed to obtain a module behavior graph set; The module behavior graph set includes several module behavior graphs; S22, Process the module behavior graph set to obtain the module feature information entropy set; S23, process the module feature information entropy set to obtain the module total importance score set.

4. The layered decoupling method for test-oriented autonomous vehicles according to claim 3, characterized in that, The process of processing the module behavior graph set to obtain the module feature information entropy set includes: S221, The module behavior graph set is parsed to obtain all module behavior graphs; S222, construct a feature vector from any of the module behavior graphs to obtain a module feature vector; S223, perform information entropy calculation on the module feature vector to obtain the module feature information entropy; S224, combine all the module feature information entropy in sequence to obtain the module feature information entropy set.

5. The layered decoupling method for test-oriented autonomous vehicles according to claim 4, characterized in that, The expression for constructing the feature vector is: , in, Indicates the first Feature vectors of each functional module; Indicates the index of the functional module; m The number of dimensions representing the features of the functional module; Indicates the relationship with the functional module In the Statistical or relational features of the behavioral graph dimensions of each of the aforementioned functional modules; The expression for calculating and processing information entropy is: ， in, Indicates the functional module Information entropy; Indicates the functional module No. The probability of a feature occurring in historical operation.

6. The layered decoupling method for test-oriented autonomous vehicles according to claim 3, characterized in that, The process of processing the module feature information entropy set to obtain the module's total importance score set includes: S231, perform mutual information extraction processing on the module feature information entropy set to construct a mutual information matrix; The mutual information extraction and processing expression is: ， in, Indicates functional modules With modules Mutual information value; Indicates the functional module With the aforementioned functional modules Joint probability of features; Indicates the functional module Marginal probability distribution of features; Indicates the functional module Marginal probability distribution of features; S232, Based on the mutual information matrix, a module decoupling diagram is constructed; S233, Based on the module feature information entropy set and the module decoupling diagram, the key module information is determined; S234, The key module information is processed to obtain the evaluation element information set.

7. The layered decoupling method for test-oriented autonomous vehicles according to claim 6, characterized in that, The process of processing the key module information to obtain the evaluation element information set includes: S2341, Obtain the module heterogeneity feature information of the key module information; S2342, Normalize the heterogeneous feature information of the module to obtain normalized heterogeneous feature information; S2343, Based on the normalized heterogeneous feature information, construct a unified feature matrix; S2344, Process the unified feature matrix to obtain the total importance score set of the module.

8. A layered decoupling device for autonomous vehicles oriented towards testing, characterized in that, The device includes: a behavior graph construction module, an importance score extraction module, and a key assessment element extraction module; The behavior graph construction module is used to construct the behavior graph of the test system; The evaluation element extraction module is used to perform hierarchical decoupling of the behavior graph of the test system to obtain an evaluation element information set; The importance score extraction module is used to process the evaluation element information set to obtain the module's total importance score information; The behavior graph construction module, the assessment element extraction module, and the key assessment element extraction module are sequentially connected.

9. A layered decoupling device for autonomous vehicles oriented towards testing, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the test-oriented layered decoupling method for autonomous vehicles as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when invoked, are used to execute the test-oriented layered decoupling method for autonomous vehicles as described in any one of claims 1-7.

Images