A multi-scene-based intelligent network connected automobile information physical system basic technology confirmation method

Abstract

The application belongs to the technical field of intelligent transportation and intelligent networked vehicles, and discloses a method for confirming the basic technology of an intelligent networked vehicle information physical system based on multiple scenes. In view of the problems of a large number of confirmation elements, a large number of combinations, and great difficulty in improving credibility caused by the diversity of IVCPS basic technology categories and the heterogeneity of objects, the application proposes a verification framework based on a typical reference system prototype. The method obtains operation result data covering multiple scales, multiple levels and multiple subjects by running no less than 16 typical scenes on the prototype system. The method realizes comprehensive confirmation of the basic technology of IVCPS architecture design, modeling, reconstruction integration and testing by using a multi-dimensional conformity evaluation method.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation and intelligent connected vehicle technology, specifically involving a method for confirming the basic technologies of intelligent connected vehicle cyber-physical systems based on multiple scenarios. Background Technology

[0002] With the rapid development of smart cities and intelligent vehicles, intelligent connected vehicle cyber-physical systems (IVCPS) based on vehicle-road-cloud collaboration have become a key technology for improving road safety, traffic efficiency, and achieving high-level autonomous driving. IVCPS is a complex system, and its basic technologies cover multiple stages such as architecture design, system modeling, refactoring and integration, and testing and verification.

[0003] Due to the complex characteristics of IVCPS—multiple stakeholders (vehicle-side, road-side, cloud-side), multiple scales (circuit-vehicle scale, road segment scale, road network scale), and multiple levels (vehicle-road-cloud level, vehicle-road level, vehicle-vehicle level)—the verification of its fundamental technologies faces significant challenges: numerous verification elements, complex combinations, and a lack of a systematic compliance assessment framework. Existing technologies often only verify a single technical point (such as the performance of a certain communication protocol) or a single scenario, making it difficult to comprehensively and objectively evaluate the rationality of the architecture design, the accuracy of the model, the effectiveness of integration, and the adequacy of testing at the system level. This makes it difficult to guarantee the "health" and "reliability" of IVCPS fundamental technologies before actual deployment, increasing the risks and costs of system development and deployment.

[0004] Therefore, there is an urgent need for a method that can systematically and quantitatively verify the core foundational technologies of IVCPS in order to support the reliable research and development and application of IVCPS. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a method for verifying the basic technologies of intelligent connected vehicle cyber-physical systems based on multiple scenarios, aiming to solve the problems of difficulty in verifying and low reliability of IVCPS basic technologies due to the diversity of categories and heterogeneity of objects.

[0006] The present invention solves the above problems by adopting the following technical solution:

[0007] A method for verifying the fundamental technologies of a cyber-physical system for intelligent connected vehicles based on multiple scenarios includes the following steps:

[0008] S1. Based on the understanding of scenarios using the Large Language Model (LLM) and combined with multi-dimensional scenario information, a dynamic weight matching method is established between multiple types of scenarios;

[0009] S2.IVCPS Architecture Technology Confirmation: Through compliance assessment of multi-scale functions, multi-level logic, and multi-entity functions and performance in various scenarios, the architecture technology is confirmed.

[0010] S3.IVCPS Modeling Technology Confirmation: The modeling technology is confirmed by evaluating the effectiveness of multi-level fusion perception and collaborative decision-making in multiple scenarios.

[0011] S4.IVCPS Reconstruction and Integration Technology Confirmation: The effectiveness of multi-agent fusion perception and collaborative decision-making in various scenarios and working conditions is confirmed through compliance assessment.

[0012] S5.IVCPS Testing Technology Validation: The testing technology is validated by evaluating the operational performance of multiple subjects in various scenarios through compliance assessment.

[0013] S6. Based on the operational results data of the typical IVCPS reference system prototype in various operating scenarios, the corresponding multi-dimensional compliance evaluation methods are adopted to confirm the architecture technology, modeling technology, refactoring and integration technology and testing technology of IVCPS. Through the results of the comprehensive technical confirmation method, it is ensured that the basic technologies can achieve the expected functions and performance.

[0014] Furthermore, step S1 includes the following sub-steps:

[0015] S1.1 Based on the understanding of the scene using a large language model, and combined with multi-dimensional information of the scene, it obtains the key standard indicators of the scene.

[0016] Standard metrics include failure severity, frequency of occurrence, system challenge, and scenario complexity;

[0017] Failure Severity: Based on historical accident data and industry safety standards, the system infers the severity of consequences of failure in the corresponding scenario and outputs a failure severity score.

[0018] Frequency of occurrence: Based on real traffic statistics, road type, and time period distribution information, the frequency of occurrence score of the corresponding scenario in actual operation is calculated;

[0019] System Challenge: Analyze the degree of challenge posed by the corresponding scenario to the technical aspects of vehicle-road-cloud collaborative perception, real-time decision-making, multi-agent interaction, and system communication latency in IVCPS, and generate a system challenge score;

[0020] Scene complexity: A scene complexity score is generated by analyzing the number of traffic participants, the randomness of their behavior, the diversity of their interaction relationships, and the size of the parameter combination space in the scene.

[0021] S1.2 Calculate the criticality score of the scenario;

[0022]

[0023] In the formula, i is the scene number, i = 1, 2, ..., n, and n is the total number of scenes; These represent the failure severity, occurrence frequency, system challenge, and scenario complexity of the i-th scenario in the scenario library, respectively. Undetermined weight coefficients are set for utilizing large language models; The scene criticality score for the i-th scene in the scene library;

[0024] S1.3 Key Scenario Weight Matching Formula;

[0025]

[0026] In the formula, For the normalized weights of the i-th scene, =1.

[0027] Furthermore, the specific content of step S2 is as follows:

[0028] Based on IVCPS operational data, the conformity assessment evaluates the effectiveness of adaptive cruise control and collision warning functions at the vehicle-scale, lane-level guidance and green wave traffic functions at the road segment scale, and the effectiveness of collaborative traffic and integrated emergency functions at the road network scale, forming a conformity assessment result of the consistency of multi-scale functions and confirming the multi-scale functional architecture technology; the conformity assessment evaluates the effectiveness of the logical consistency of adaptive cruise control, collision warning, lane-level guidance, green wave traffic, collaborative traffic, and integrated emergency functions at the vehicle-to-vehicle, vehicle-to-infrastructure, and vehicle-to-infrastructure-cloud levels, confirming the multi-level logical architecture design technology; and the conformity assessment evaluates the consistency of vehicle-side functions and performance, road-side functions and performance, and cloud-side functions and performance, confirming the multi-entity cyber-physical architecture design technology.

[0029] Furthermore, the specific content of step S3 is as follows:

[0030] Based on the operational data of IVCPS, the conformity assessment results of vehicle-road-cloud fusion perception and coordinated decision-making are obtained, as well as vehicle-to-vehicle fusion perception and coordinated decision-making. This results in the conformity assessment of multi-level functions and confirms the IVCPS modeling technology.

[0031] Furthermore, the specific content of step S4 is as follows:

[0032] Based on the operational data of IVCPS, the compliance assessment evaluates the effectiveness of cloud-based fusion perception and collaborative decision-making, the roadside fusion perception and collaborative decision-making, and the vehicle-side fusion perception and collaborative decision-making, forming a compliance assessment result for the functions and performance of multiple entities, and confirming the IVCPS reconstruction and integration technology.

[0033] Furthermore, the specific content of step S5 is as follows:

[0034] Based on the operational data of IVCPS, the conformity assessment results are obtained by evaluating the effectiveness of cloud-based testing, roadside testing, and vehicle-based testing, thus forming a multi-entity conformity assessment result and confirming the IVCPS testing technology.

[0035] Furthermore, in steps S2 to S5, the architecture technology confirmation stage evaluates the functionality, logic, and performance using the hierarchical entropy weight-matter extension model, the interactive collaborative relationship neural network algorithm, and the analytic hierarchy process, respectively; the modeling technology confirmation stage focuses on mapping the logical requirements of multi-level functional layers based on the interactive collaborative relationship neural network algorithm; the reconstruction and integration technology confirmation stage coordinates functionality and performance from multiple subject dimensions and reuses the hierarchical entropy weight-matter extension model; and the testing technology confirmation stage uses the hierarchical entropy weight-matter extension model to match test requirements for multi-subject scenarios.

[0036] I. Hierarchical Entropy Weight-Matter-Element Extension Model;

[0037]

[0038] In the formula, is the result of the hierarchical entropy weight-matter extension model for calculating the technical compliance of scenario i; m is the total number of functional / performance indicators selected to evaluate the technology; It is the actual value of the k-th indicator extracted from the running data of scenario i; x represents the actual value of the indicator; Let k be the classical domain of the k-th index; For the k-th index, the section domain is... Represents the matter-element extensional correlation function for the k-th index; This indicates the actual value x of the indicator to the node range. The distance; This indicates the actual value x of the indicator to the node range. The distance;

[0039] II. Interactive Collaborative Relationship Neural Network Algorithm;

[0040]

[0041] In the formula, The result of the interactive collaborative relationship neural network algorithm for calculating the technical compliance of scenario i; This refers to the runtime data of all interactions between entities in scene i; This is the total loss value calculated based on scene interaction data; Loss due to logical consistency; Loss of accuracy in collaborative decision-making; For delay loss; , , Preset weights for each loss term in the network; λ is a scaling factor used to normalize the loss value to the [0,1] interval;

[0042] III. Analytic Hierarchy Process (AHP);

[0043] And satisfy

[0044] In the formula, P represents the technical compliance calculation result of scenario i using the analytic hierarchy process; P is the number of sub-dimensions used for evaluation. For the test requirements corresponding to scenario i, use a coverage matrix or a multi-dimensional evaluation matrix. According to The calculated coverage or compliance rate of the p-th sub-dimension has a value range of... ; For each sub-dimension, there is a weight vector. ; Let be the weight of the p-th sub-dimension; It is a P×P pairwise comparison judgment matrix, reflecting the experts' judgments on the relative importance of each dimension; To determine the matrix The largest eigenvalue; and The above eigenvalue equation is obtained, and the consistency test CR < 0.1 must be satisfied.

[0045] Furthermore, the specific content of step S6 is as follows:

[0046] Based on the operational results data of the i-th scenario on the IVCPS typical reference system prototype, the multi-dimensional compliance assessment method is used to obtain the compliance result of a certain basic technology to be verified in this single scenario. Combined with normalized weights for each scenario The final confirmation score T for this basic technology is calculated using the following formula:

[0047]

[0048] Finally, a confirmation judgment is made: if T ≥ θ, where θ is a preset threshold, then the basic technology is confirmed to be effective; otherwise, it is determined to have failed the confirmation.

[0049] Furthermore, the number of the various operating scenarios is no less than 16, covering complex operating conditions ranging from single vehicle to multiple vehicles and from single function to multi-functional collaboration.

[0050] Beneficial effects:

[0051] 1. Systematic: For the first time, a complete validation framework covering the four fundamental technologies of IVCPS (architecture, modeling, integration, and testing) was proposed, breaking the limitations of traditional single-technology point verification.

[0052] 2. Quantification: The introduction of a multi-dimensional compliance assessment method transforms the original subjective and qualitative technical assessment into objective and comparable quantitative indicators (such as functional consistency percentage, decision accuracy, performance deviation, etc.), thereby improving the credibility and scientific rigor of the confirmation results.

[0053] 3. Based on prototypes and scenarios: By running a variety of scenarios (no less than 16 types) on actual or high-fidelity typical reference system prototypes, real or highly simulated running data is obtained, making the verification process closer to reality and the results more convincing.

[0054] 4. Providing standardized benchmarks: The "typical reference system prototype" on which this invention relies has clearly defined "three-multi" (multi-heterogeneous subjects, multi-temporal and spatial scales, multi-level closed loops) and "three-characteristics" (evolutionary, emergent, and open) features, as well as no less than 16 standardized scenario sets covering typical applications. This provides reproducible and comparable standardized test benchmarks and data foundations for the confirmation of IVCPS basic technologies, which is conducive to promoting the standardization of industry technology evaluation.

[0055] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0056] Figure 1 This is a block diagram of a typical reference system prototype for IVCPS provided in an embodiment of the present invention;

[0057] Figure 2 This is a block diagram illustrating the overall scheme for comprehensive verification of IVCPS basic technologies provided in this embodiment of the invention.

[0058] Figure 3 A flowchart illustrating the verification process for IVCPS basic technologies provided in this embodiment of the invention;

[0059] Figure 4 The IVCPS basic technology verification process based on multiple scenarios provided for embodiments of the present invention;

[0060] Figure 5 This is a schematic diagram illustrating the comprehensive validation of the IVCPS basic technology provided in the embodiments of the present invention to support the design operating domain. Detailed Implementation

[0061] To make the technical solutions, advantages, and objectives of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the protection scope of this application.

[0062] like Figure 1 and Figure 2 As shown, this invention provides a method for verifying the fundamental technologies of a cyber-physical system for intelligent connected vehicles based on multiple scenarios, including the following steps:

[0063] S1. Key Scene Weight Matching Method Based on Large Language Model (LLM): Based on the understanding of scenes by the large language model and combined with multi-dimensional information of scenes, a dynamic weight matching method is established between multiple types of scenes;

[0064] S1.1 Based on the understanding of the scene using a large language model, and combined with multi-dimensional information of the scene, it obtains the key standard indicators of the scene.

[0065] Standard metrics include failure severity, frequency of occurrence, system challenge, and scenario complexity;

[0066] Failure Severity Assessment: Based on historical accident data, industry safety standards, etc., infer the level of consequences that the system may cause if it fails in this scenario (such as from "affecting traffic efficiency" to "causing injury or death"), and output the failure severity score.

[0067] Frequency assessment: By combining real traffic statistics, road types, time period distribution and other information, the frequency score of this scenario in actual operation is automatically calculated.

[0068] System challenge assessment: Analyze the degree of challenge this scenario poses to the technical aspects of IVCPS, such as vehicle-road-cloud collaborative perception, real-time decision-making, multi-agent interaction, and system communication latency, and generate a system challenge score.

[0069] Scene complexity assessment: By analyzing the number of traffic participants, the randomness of their behavior, the diversity of their interaction relationships, and the size of the parameter combination space in the scene, a scene complexity score is generated;

[0070] S1.2 Calculate the criticality score of the scenario;

[0071]

[0072] In the formula, i is the scene number, i = 1, 2, ..., n, and n is the total number of scenes (no less than 16). These represent the failure severity, occurrence frequency, system challenge, and scenario complexity of the i-th scenario in the scenario library, respectively. Undetermined weight coefficients are set for utilizing large language models; The scene criticality score is assigned to the i-th scene in the scene library; the higher the weight assigned to the scene, the higher the criticality score.

[0073] S1.3 Key Scenario Weight Matching Formula;

[0074]

[0075] In the formula, For the normalized weights of the i-th scene, =1, so that all scenarios are matched with the corresponding weight.

[0076] S2.IVCPS Architecture Technology Confirmation: Through compliance assessment of multi-scale functions, multi-level logic, and multi-entity functions and performance in various scenarios, the architecture technology is confirmed.

[0077] The confirmation of IVCPS architecture technology specifically includes: conducting conformity assessments of the effectiveness of adaptive cruise control and collision warning functions at the vehicle scale, lane-level guidance and green wave traffic functions at the road segment scale, and collaborative traffic and integrated emergency functions at the road network scale, through at least 16 scenarios and prototype operation data of typical IVCPS reference systems, to form conformity assessment results of the consistency of multi-scale functions and confirm the multi-scale functional architecture technology; conducting conformity assessments of the logical consistency of adaptive cruise control, collision warning, lane-level guidance, green wave traffic, collaborative traffic, and integrated emergency functions at the vehicle-to-vehicle, vehicle-to-infrastructure, and vehicle-to-infrastructure-cloud levels, to confirm the multi-level logical architecture design technology; and conducting conformity assessments of the consistency of vehicle-side functions and performance, road-side functions and performance, and cloud-side functions and performance, to confirm the multi-entity cyber-physical architecture design technology.

[0078] S3.IVCPS Modeling Technology Confirmation: The modeling technology is confirmed by evaluating the effectiveness of multi-level fusion perception and collaborative decision-making in multiple scenarios.

[0079] The confirmation of IVCPS modeling technology specifically includes: using data from the prototype of a typical IVCPS reference system in no fewer than 16 scenarios to conduct conformity assessments of the effects of vehicle-road-cloud fusion perception and coordinated decision-making, vehicle-road fusion perception and coordinated decision-making, and vehicle-to-vehicle fusion perception and coordinated decision-making, forming conformity assessment results for multi-level functions, and thus confirming the IVCPS modeling technology.

[0080] S4.IVCPS Reconstruction and Integration Technology Confirmation: The effectiveness of multi-agent fusion perception and collaborative decision-making in various scenarios and working conditions is confirmed through compliance assessment.

[0081] The confirmation of IVCPS reconfiguration and integration technology specifically includes: using data from the prototype operation of a typical IVCPS reference system in no fewer than 16 scenarios to conduct conformity assessments of the effects of cloud-based fusion perception and collaborative decision-making, roadside fusion perception and collaborative decision-making, and vehicle-side fusion perception and collaborative decision-making, thereby forming conformity assessment results for the functions and performance of multiple entities and achieving the confirmation of IVCPS reconfiguration and integration technology.

[0082] S5.IVCPS Testing Technology Validation: The testing technology is validated by evaluating the operational performance of multiple subjects in various scenarios through compliance assessment.

[0083] The validation of IVCPS testing technology specifically includes: running data on a typical IVCPS reference system prototype in no fewer than 16 scenarios; evaluating the effectiveness of cloud-based testing; evaluating the effectiveness of roadside testing; evaluating the effectiveness of vehicle-side testing; generating multi-entity testing compliance evaluation results; and thus achieving validation of the IVCPS testing technology. A flowchart for the IVCPS basic technology validation process is attached. Figure 3 .

[0084] To achieve the aforementioned multi-dimensional and quantitative verification objectives, this invention addresses the different dimensions of IVCPS compliance assessment for different underlying technologies, combines typical IVCPS characteristics, employs different mapping methods, and comprehensively utilizes various advanced assessment models and algorithms, primarily including:

[0085] The architecture technology verification phase evaluates functionality, logic, and performance using the hierarchical entropy weight-matter extension model, the interactive collaborative relationship neural network algorithm, and the analytic hierarchy process (AHP). Modeling technology verification focuses on mapping logical requirements across multiple functional levels, also based on the interactive collaborative relationship neural network algorithm. Refactoring and integration technology verification coordinates functionality and performance from multiple perspectives, reusing the hierarchical entropy weight-matter extension model. Testing technology verification targets multi-agent scenarios, using the hierarchical entropy weight-matter extension model for test requirement matching. For a detailed comparison of the evaluation dimensions and methods, please refer to the appendix. Figure 4 :

[0086] I. Hierarchical Entropy Weighting and Matter-Element Extension Model: This method integrates subjective hierarchical analysis with objective entropy weighting, and combines matter-element extension theory to quantify the correlation between technical states and desired goals. It is primarily used to evaluate the functional consistency and integration effectiveness of a system, such as the confirmation of architectural and refactoring integration techniques. The formula for calculating technical compliance using the hierarchical entropy weighting-matter-extension model (e.g., for architectural and refactoring integration techniques) is as follows:

[0087]

[0088] In the formula, is the result of the hierarchical entropy weight-matter extension model for calculating the technical compliance of scenario i; m is the total number of functional / performance indicators selected to evaluate the technology; It is the actual value of the k-th metric extracted from the running data of scenario i (e.g., latency, accuracy, energy consumption); x represents the actual value of the metric. The classic domain (acceptable target range) for the k-th indicator; For the k-th index, the section domain (all possible value ranges); Describes the matter-element extension correlation function for the k-th index. ≥0 indicates that the requirements are met; the higher the value, the better. This indicates the actual value x of the indicator to the node range. The distance; This indicates the actual value x of the indicator to the node range. The distance.

[0089] II. Interactive Collaborative Relationship Neural Network Algorithm: An intelligent evaluation algorithm for complex system interaction relationships. Through learning and reasoning about multi-agent, multi-level interaction patterns in operational data, it assesses the logical consistency of collaborative decision-making and the accuracy of perceptual fusion. It is primarily used for confirming modeling techniques and verifying the consistency of architectural logic. The formula for calculating technical compliance using the interactive collaborative relationship neural network algorithm is as follows (e.g., for modeling techniques).

[0090]

[0091] In the formula, The result of the interactive collaborative relationship neural network algorithm for calculating the technical compliance of scenario i; This refers to the runtime data (time series) of all interactions between the main entities in scenario i. This is the total loss value calculated based on scene interaction data; This is the logical consistency loss (such as the degree of divergence of decision vectors among different subjects). This is due to the loss of accuracy in collaborative decision-making (such as deviation from the benchmark truth). This is due to time delay (delay in information synchronization or decision response). , , The network preset weights for each loss term (fixed during training); λ is a scaling factor used to normalize the loss value to the [0,1] interval;

[0092] III. Analytic Hierarchy Process (AHP): A classic multi-criteria decision-making method used to construct a hierarchical structure of evaluation indicators and quantify the importance weights of each dimension based on expert experience, providing a foundation for comprehensive evaluation. It plays a key role in performance evaluation and determination of comprehensive weights. The formula for calculating technical compliance using AHP (e.g., for testing technology and architecture comprehensive verification) is as follows:

[0093] And satisfy

[0094] In the formula, P represents the technical compliance calculation result of scenario i using the analytic hierarchy process; P is the number of sub-dimensions used for evaluation (e.g., functional coverage, performance coverage, security coverage, etc.). For the test requirements corresponding to scenario i, use a coverage matrix or a multi-dimensional evaluation matrix. According to The calculated coverage or compliance rate of the p-th sub-dimension has a value range of... ; For each sub-dimension, there is a weight vector. ; Let be the weight of the p-th sub-dimension; It is a P×P pairwise comparison judgment matrix, reflecting the experts' judgments on the relative importance of each dimension; To determine the matrix The largest eigenvalue; and The above eigenvalue equation is obtained, and the consistency test CR < 0.1 must be satisfied.

[0095] These methods are applied in targeted combinations based on the different characteristics (such as functionality, logic, performance, and coverage) of the dimensions to be identified in each fundamental technology, collectively forming a system such as... Figure 4 The systematic and intelligent compliance assessment system shown ensures that the confirmation results are comprehensive, objective, and scientific.

[0096] S6. Based on the operational results data of the typical IVCPS reference system prototype in various operating scenarios, the corresponding multi-dimensional compliance evaluation methods are adopted to confirm the architecture technology, modeling technology, refactoring and integration technology and testing technology of IVCPS respectively. Through the results of the comprehensive technical confirmation method, it is ensured that the basic technology can achieve the expected functions and performance.

[0097] Finally, based on the comprehensive verification method using the operational results data of the i-th scenario on the IVCPS typical reference system prototype, the multi-dimensional conformity assessment method is employed to obtain the conformity result of a certain basic technology (such as modeling technology) to be verified in this single scenario. Combined with normalized weights for each scenario The final confirmation score T for this basic technology is calculated using the following formula:

[0098]

[0099] Final confirmation judgment: If T ≥ θ, where θ is a preset threshold, then the basic technology is confirmed to be effective; otherwise, it is determined to have failed the confirmation.

[0100] The aforementioned compliance assessment process uses quantitative methods such as hierarchical entropy weight-matter extension model, interactive collaborative relationship neural network, and analytic hierarchy process to systematically and quantitatively determine the feasibility, applicability, practicality, and usability of various technologies under different conditions, thereby achieving comprehensive confirmation of IVCPS basic technologies from design to operation.

[0101] Furthermore, this method supports technical verification of at least 16 IVCPS design runtime domains through multi-dimensional compliance assessments of at least 16 typical scenarios. See the appendix for details. Figure 5 The IVCPS fundamental technology has been comprehensively validated to support the design of operational domain diagrams. Specific support includes:

[0102] 1. Architecture design technology: Starting from the compliance of multi-scale functional architecture, multi-level logical architecture, and multi-entity cyber-physical architecture, it supports the evaluation of road conditions, traffic environment, roadside facilities, and cloud services, and identifies the feasibility range of multiple temporary modification states.

[0103] 2. Modeling techniques: Starting from the conformity of vehicle-road-cloud, vehicle-road, and vehicle-to-vehicle levels, clarify their applicability to road conditions, traffic environment, temporary modification status, and cloud services to ensure the effectiveness of modeling in different level scenarios.

[0104] 3. Reconstruction and integration technology: Starting from the compliance of cloud, road, and vehicle, confirm its applicability in road conditions, traffic environment, temporary modification status, and cloud services, and ensure that the system still meets the design requirements after integration among different entities.

[0105] 4. Testing and Verification Technology: Through multi-scenario verification, establish a description of the availability range of road conditions, traffic environment, temporary modification status, and cloud services to ensure that the test coverage is realistic and representative.

[0106] Table 1 shows typical operating scenarios of intelligent connected vehicles deployed on prototypes according to embodiments of the present invention.

[0107] Table 1

[0108]

[0109] It is hereby declared that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for verifying the fundamental technologies of a cyber-physical system for intelligent connected vehicles based on multiple scenarios, characterized in that, Includes the following steps: S1. Based on the understanding of scenarios using the Large Language Model (LLM) and combined with multi-dimensional scenario information, a dynamic weight matching method is established between multiple types of scenarios; S2.IVCPS Architecture Technology Confirmation: Through compliance assessment of multi-scale functions, multi-level logic, and multi-entity functions and performance in various scenarios, the architecture technology is confirmed. S3.IVCPS Modeling Technology Confirmation: The modeling technology is confirmed by evaluating the effectiveness of multi-level fusion perception and collaborative decision-making in multiple scenarios. S4.IVCPS Reconstruction and Integration Technology Confirmation: The effectiveness of multi-agent fusion perception and collaborative decision-making in various scenarios and working conditions is confirmed through compliance assessment. S5.IVCPS Testing Technology Validation: The testing technology is validated by evaluating the operational performance of multiple subjects in various scenarios through compliance assessment. S6. Based on the operational results data of the typical IVCPS reference system prototype in various operating scenarios, the corresponding multi-dimensional compliance evaluation methods are adopted to confirm the architecture technology, modeling technology, refactoring and integration technology and testing technology of IVCPS. Through the results of the comprehensive technical confirmation method, it is ensured that the basic technologies can achieve the expected functions and performance.

2. The method for verifying the fundamental technologies of a multi-scenario intelligent connected vehicle cyber-physical system according to claim 1, characterized in that, Step S1 includes the following sub-steps: S1.1 Based on the understanding of the scene using a large language model, and combined with multi-dimensional information of the scene, it obtains the key standard indicators of the scene. Standard metrics include failure severity, frequency of occurrence, system challenge, and scenario complexity; Failure Severity: Based on historical accident data and industry safety standards, the system infers the severity of consequences of failure in the corresponding scenario and outputs a failure severity score. Frequency of occurrence: Based on real traffic statistics, road type, and time period distribution information, the frequency of occurrence score of the corresponding scenario in actual operation is calculated; System Challenge: Analyze the degree of challenge posed by the corresponding scenario to the technical aspects of vehicle-road-cloud collaborative perception, real-time decision-making, multi-agent interaction, and system communication latency in IVCPS, and generate a system challenge score; Scene complexity: A scene complexity score is generated by analyzing the number of traffic participants, the randomness of their behavior, the diversity of their interaction relationships, and the size of the parameter combination space in the scene. S1.2 Calculate the criticality score of the scenario; In the formula, i is the scene number, i = 1, 2, ..., n, and n is the total number of scenes; These represent the failure severity, occurrence frequency, system challenge, and scenario complexity of the i-th scenario in the scenario library, respectively. Undetermined weight coefficients are set for utilizing large language models; The scene criticality score for the i-th scene in the scene library; S1.3 Key Scenario Weight Matching Formula; In the formula, For the normalized weights of the i-th scene, =1.

3. The method for verifying the fundamental technologies of a multi-scenario intelligent connected vehicle cyber-physical system according to claim 2, characterized in that, The specific content of step S2 is as follows: Based on IVCPS operational data, the conformity assessment evaluates the effectiveness of adaptive cruise control and collision warning functions at the vehicle-scale, lane-level guidance and green wave traffic functions at the road segment scale, and the effectiveness of collaborative traffic and integrated emergency functions at the road network scale, forming a conformity assessment result of the consistency of multi-scale functions and confirming the multi-scale functional architecture technology; the conformity assessment evaluates the effectiveness of the logical consistency of adaptive cruise control, collision warning, lane-level guidance, green wave traffic, collaborative traffic, and integrated emergency functions at the vehicle-to-vehicle, vehicle-to-infrastructure, and vehicle-to-infrastructure-cloud levels, confirming the multi-level logical architecture design technology; and the conformity assessment evaluates the consistency of vehicle-side functions and performance, road-side functions and performance, and cloud-side functions and performance, confirming the multi-entity cyber-physical architecture design technology.

4. The method for verifying the fundamental technologies of a multi-scenario intelligent connected vehicle cyber-physical system according to claim 3, characterized in that, The specific content of step S3 is as follows: Based on the operational data of IVCPS, the conformity assessment results of vehicle-road-cloud fusion perception and coordinated decision-making are obtained, as well as vehicle-to-vehicle fusion perception and coordinated decision-making. This results in the conformity assessment of multi-level functions and confirms the IVCPS modeling technology.

5. The method for verifying the fundamental technologies of a multi-scenario intelligent connected vehicle cyber-physical system according to claim 4, characterized in that, The specific content of step S4 is as follows: Based on the operational data of IVCPS, the compliance assessment evaluates the effectiveness of cloud-based fusion perception and collaborative decision-making, the roadside fusion perception and collaborative decision-making, and the vehicle-side fusion perception and collaborative decision-making, forming a compliance assessment result for the functions and performance of multiple entities, and confirming the IVCPS reconstruction and integration technology.

6. The method for verifying the fundamental technologies of a multi-scenario intelligent connected vehicle cyber-physical system according to claim 5, characterized in that, The specific content of step S5 is as follows: Based on the operational data of IVCPS, the conformity assessment results are obtained by evaluating the effectiveness of cloud-based testing, roadside testing, and vehicle-based testing, thus forming a multi-entity conformity assessment result and confirming the IVCPS testing technology.

7. The method for verifying the fundamental technologies of a multi-scenario intelligent connected vehicle cyber-physical system according to claim 6, characterized in that, In steps S2-S5, the architecture technology confirmation stage evaluates the functionality, logic, and performance using the hierarchical entropy weight-matter extension model, the interactive collaborative relationship neural network algorithm, and the analytic hierarchy process, respectively. The modeling technology confirmation stage focuses on mapping logical requirements across multiple functional levels based on the interactive collaborative relationship neural network algorithm. The reconstruction and integration technology confirmation stage coordinates functionality and performance across multiple stakeholders, reusing the hierarchical entropy weight-matter extension model. The testing technology confirmation stage uses the hierarchical entropy weight-matter extension model to match test requirements for multi-stakeholder scenarios. I. Hierarchical Entropy Weight-Matter-Element Extension Model; In the formula, is the result of the hierarchical entropy weight-matter extension model for calculating the technical compliance of scenario i; m is the total number of functional / performance indicators selected to evaluate the technology; It is the actual value of the k-th indicator extracted from the running data of scenario i; x represents the actual value of the indicator; Let k be the classical domain of the k-th index; For the k-th index, the section domain is... Represents the matter-element extensional correlation function for the k-th index; This indicates the actual value x of the indicator to the node range. The distance; This indicates the actual value x of the indicator to the node range. The distance; II. Interactive Collaborative Relationship Neural Network Algorithm; In the formula, The result of the interactive collaborative relationship neural network algorithm for calculating the technical compliance of scenario i; This refers to the runtime data of all interactions between entities in scene i; This is the total loss value calculated based on scene interaction data; Loss due to logical consistency; Loss of accuracy in collaborative decision-making; For delay loss; , , Pre-set weights for each loss term in the network; λ is a scaling factor used to normalize the loss value to the [0,1] interval; III. Analytic Hierarchy Process (AHP); And satisfy In the formula, P represents the technical compliance calculation result of scenario i using the analytic hierarchy process; P is the number of sub-dimensions used for evaluation. For the test requirements corresponding to scenario i, use a coverage matrix or a multi-dimensional evaluation matrix. According to The calculated coverage or compliance rate of the p-th sub-dimension has a value range of... ; For each sub-dimension, there is a weight vector. ; Let be the weight of the p-th sub-dimension; It is a P×P pairwise comparison judgment matrix, reflecting the experts' judgments on the relative importance of each dimension; To determine the matrix The largest eigenvalue; and The above eigenvalue equation is obtained, and the consistency test CR < 0.1 must be satisfied.

8. The method for verifying the fundamental technologies of a multi-scenario intelligent connected vehicle cyber-physical system according to claim 7, characterized in that, The specific content of step S6 is as follows: Based on the operational results data of the i-th scenario on the IVCPS typical reference system prototype, the multi-dimensional compliance assessment method is used to obtain the compliance result of a certain basic technology to be verified in this single scenario. Combined with normalized weights for each scenario The final confirmation score T for this basic technology is calculated using the following formula: Finally, a confirmation judgment is made: if T ≥ θ, where θ is a preset threshold, then the basic technology is confirmed to be effective; Otherwise, it will be considered as a failure to pass confirmation.

9. The method for verifying the fundamental technologies of a multi-scenario intelligent connected vehicle cyber-physical system according to claim 8, characterized in that: The number of operational scenarios is no less than 16, covering complex operational situations ranging from single-vehicle to multi-vehicle and from single-function to multi-function collaboration.

Images