Autonomous traffic system architecture diagnosis method based on rule reasoning

Abstract

The present invention relates to the field of diagnosis of traffic system architecture, and more specifically to a method for diagnosing an autonomous traffic system architecture based on rule reasoning, the method comprising the following steps: S1: tracing abnormal information of the autonomous traffic system architecture based on an attribute graph; S2: classifying the abnormal information according to the diagnosis rules based on the abnormal information tracing results obtained in S1; S3: ending the diagnosis process; S21: the autonomous traffic system architecture diagnosis model detects abnormalities of the autonomous traffic system architecture according to the diagnosis rules; S22: classifying the abnormal information tracing results obtained by the autonomous traffic system architecture diagnosis model; the method splits the knowledge required for diagnosis into attribute graphs and rules, first locates the abnormal information in the architecture through the graph-based autonomous traffic system architecture anomaly detection model, and then obtains the abnormal information classification results through rule reasoning on this basis, thereby improving the efficiency of diagnosis.

Description

Technical Field

[0001] The present invention relates to the field of diagnosis of traffic system architecture, and more particularly to an autonomous traffic system architecture diagnosis method based on rule reasoning. Background Art

[0002] With the development of advanced technologies and evolving transportation needs, intelligent transportation systems are evolving into autonomous transportation systems capable of self-analysis, learning, and decision-making. Based on the overall objectives of the transportation system, the autonomous transportation system architecture comprehensively organizes transportation services, infrastructure, communication networks, and traffic participants, providing macro-level guidance for the transportation system. Architecture diagnosis is key to ensuring the integrity and logical correctness of the autonomous transportation system architecture. By examining the components and structure of the autonomous transportation system architecture, it provides a basis for resolving issues within the autonomous transportation system architecture. Therefore, analyzing, identifying, and processing anomalies based on large-scale autonomous transportation system architecture data is of great significance to the transportation system architecture design industry in terms of event information mining and intelligent transportation system analysis.

[0003] The structure and content of the autonomous transportation system architecture adapt to technological development and changing requirements. Frequent updates and maintenance of the system architecture can introduce potential local errors, leading to system malfunctions, incomplete functionality, and inefficient resource utilization. Architecture diagnosis can identify the source of autonomous transportation system architecture anomalies and classify them, providing a better basis for resolving them. Summary of the Invention

[0004] The purpose of the present invention is to provide an autonomous traffic system architecture diagnosis method based on rule reasoning, which can trace the abnormal information in the autonomous traffic system architecture and realize the classification of the abnormal information in the autonomous traffic system architecture in combination with the diagnosis rules.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] An autonomous traffic system architecture diagnosis method based on rule reasoning, the method comprising the following steps:

[0007] S1: An autonomous traffic system architecture anomaly detection model for tracing abnormal information is provided, and an autonomous traffic system architecture anomaly information tracing based on an attribute graph is provided;

[0008] S11: Construct the autonomous transportation system architecture reference design set R and architecture data S through the autonomous transportation system theoretical framework and logical, functional, and physical views;

[0009] S12: The reference design set R and the architecture dataset S are processed into labeled, directed reference design set attribute graph R' and architecture dataset attribute graph S';

[0010] S13: Autonomous Transportation System Architecture The anomaly detection model will find the largest common part of the attribute graphs R' and S';

[0011] S14: Find the complement of the common parts of the input architecture attribute graph S' and the reference design set attribute graph R', that is, the unmatched nodes and edges in the input architecture attribute graph;

[0012] S2: Based on the abnormal information tracing results obtained in S1, the abnormal information is classified according to the diagnosis rules;

[0013] S21: The autonomous transportation system architecture diagnosis model detects abnormalities in the autonomous transportation system architecture according to the diagnosis rules;

[0014] S22: Classify the abnormal information tracing results obtained by the autonomous traffic system architecture diagnosis model;

[0015] S3: The diagnostic process ends;

[0016] The abnormal information of the autonomous traffic system architecture includes isolated node abnormal information, functional architecture abnormal information, logical architecture abnormal information and physical architecture abnormal information;

[0017] The diagnostic rules include diagnostic rules for isolated nodes, diagnostic rules for functional architecture, diagnostic rules for physical architecture and diagnostic rules for logical architecture;

[0018] The diagnosis rules for the isolated nodes are:

[0019] Rule1:v inNodes but v not exists inEdges

[0020] When this rule is executed, it will traverse the points in the node set. If the point does not appear in the edge set, it will be judged as an isolated point, that is, this type of node has no link relationship with other nodes;

[0021] The diagnostic rules of the functional architecture are:

[0022] Rule2:source.type=servicedomain&&target.type=service

[0023] Rule3:source.type=service&&target.type=subservice

[0024] Rule4:source.type=subservice&&target.type=function

[0025] Rule5:source.type=function&&target.type=subfuntion

[0026] When this rule is executed, all possible edges are matched with the above four types of diagnostic rules in sequence, and the corresponding edges are output if they meet the conditions;

[0027] The diagnostic rules for the physical architecture are:

[0028] Rule7:source.type=phyobject&&link.type=support

[0029] Rule8:source.type=phyobject&&target.type=phyobject

[0030] When this rule is executed, it determines whether the functions provided or undertaken by the physical object are consistent with the autonomous transportation system architecture reference design set;

[0031] The diagnostic rules of the logical architecture are:

[0032] Rule9:source.type=function&&target.type=function

[0033] When this rule is executed, if the input type and function match and the target type and function match, the logic between the functions is considered reasonable, otherwise it is unreasonable;

[0034] This method splits the knowledge required for diagnosis into attribute graphs and rules, and locates abnormal information in the architecture through a graph-based autonomous traffic system architecture anomaly detection model.

[0035] The beneficial effects of the present invention are:

[0036] First, this method breaks down the knowledge required for diagnosis into attribute graphs and rules. First, it uses a graph-based autonomous traffic system architecture anomaly detection model to locate anomalies in the architecture. Based on this, it uses rule-based reasoning to derive anomaly classification results, improving diagnostic efficiency.

[0037] Second, this method first designs an autonomous transportation system architecture anomaly detection model based on the autonomous transportation system architecture reference design set. It locates the abnormal parts of the diagnosed architecture, effectively solves the problem of identifying abnormal information of the autonomous transportation system architecture, and narrows the scope of reasoning diagnosis.

[0038] 3. This method combines the structural characteristics of the autonomous transportation system architecture and designs diagnostic rules from four aspects: functional architecture, logical architecture, physical architecture, and isolated nodes. The results obtained by the autonomous transportation system architecture anomaly detection model are used as the input for the classification of autonomous transportation system architecture anomaly information, realizing a one-to-one correspondence and summary output between the architecture anomaly information and the diagnostic classification results, thereby improving the efficiency of autonomous transportation system architecture diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention will be further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0040] Figure 1 It is the autonomous traffic system architecture diagnosis process based on rule reasoning of the present invention;

[0041] Figure 2 It is a flow chart of abnormal information detection of autonomous traffic system architecture based on attribute graph of the present invention;

[0042] Figure 3 It is the abnormal information tracing result of the autonomous traffic system architecture of the present invention;

[0043] Figure 4 It is the abnormal information classification result of the autonomous traffic system architecture of the present invention;

[0044] Figure 5 It is a diagnostic flow chart of the autonomous traffic system architecture diagnostic model of the present invention. DETAILED DESCRIPTION

[0045] The present invention will be further described in detail below with reference to the accompanying drawings.

[0046] The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention. In the embodiments of the present invention, the data used comes from the vehicle environment perception service under the road vehicle operation service domain; Figures 1 to 4 As shown, in order to solve the technical problem of "how to realize the classification of abnormal information of autonomous traffic system architecture", the steps and functions of the autonomous traffic system architecture diagnosis method based on rule reasoning are described in detail below;

[0047] S1: An autonomous traffic system architecture anomaly detection model for tracing abnormal information is provided, and an autonomous traffic system architecture anomaly information tracing based on an attribute graph is provided;

[0048] First, we constructed an autonomous transportation system architecture reference design set R and an architecture dataset S for vehicle-based environmental awareness services based on the theoretical framework and its logical, functional, and physical aspects. The constructed reference design set S contains 151 relationship pairs, including 59 negative samples and 92 positive samples. The architecture dataset S includes 39 nodes, including one service domain, one service, three sub-service nodes, 18 function nodes, and 16 physical object nodes. It also includes 151 link relationships, including 54 data flows, 50 information interaction pairs, and 47 hierarchical relationships.

[0049] The autonomous transportation system architecture reference design set R and the input architecture dataset S are processed into a labeled, directed reference design set attribute graph R' and an architecture attribute graph S'.

[0050] Based on the data obtained from the above processing, the autonomous transportation system architecture anomaly detection model will find the maximum common part of the reference design set attribute graph R' and the architecture attribute graph S', and finally find the complement of the common part of the reference design set attribute graph R' and the architecture attribute graph S'. In the example of the present invention, a total of 49 anomaly information were obtained, and the anomaly detection output results are as follows: Figure 3 shown.

[0051] S2: Based on the abnormal information tracing results obtained in S1, the abnormal information is classified according to the diagnosis rules;

[0052] This paper summarizes the common anomaly types of autonomous transportation system architecture and sets nine rules for four types of problems: isolated nodes, functional architecture, logical architecture, and physical architecture of autonomous transportation system architecture;

[0053] like Figure 5 As shown in Figure 1, the autonomous traffic system architecture diagnosis model integrates the diagnosis rules, matches the abnormal information obtained by the autonomous traffic system architecture anomaly detection model with the diagnosis rules, and the matching result is the classification result;

[0054] The diagnosis rules for the isolated nodes are:

[0055] Rule1:v inNodes but v not exists inEdges

[0056] When this rule is executed, it will traverse the points in the node set. If the point does not appear in the edge set, it will be judged as an isolated point, that is, this type of node has no link relationship with other nodes;

[0057] The diagnostic rules of the functional architecture are:

[0058] Rule2:source.type=servicedomain&&target.type=service

[0059] Rule3:source.type=service&&target.type=subservice

[0060] Rule4:source.type=subservice&&target.type=function

[0061] Rule5:source.type=function&&target.type=subfuntion

[0062] When this rule is executed, all possible edges are matched with the above four types of diagnostic rules in sequence, and the corresponding edges are output if they meet the conditions;

[0063] The diagnostic rules for the physical architecture are:

[0064] Rule7:source.type=phyobject&&link.type=support

[0065] Rule8:source.type=phyobject&&target.type=phyobject

[0066] When this rule is executed, it determines whether the functions provided or undertaken by the physical object are consistent with the autonomous transportation system architecture reference design set;

[0067] The diagnostic rules of the logical architecture are:

[0068] Rule9:source.type=function&&target.type=function

[0069] When this rule is executed, if the input type and function match and the target type and function match, the logic between the functions is considered reasonable, otherwise it is unreasonable;

[0070] This method splits the knowledge required for diagnosis into attribute graphs and rules, and locates abnormal information in the architecture through a graph-based autonomous traffic system architecture anomaly detection model.

[0071] The architecture anomaly information output by the autonomous transportation system architecture anomaly detection model is used as the input of the autonomous transportation system architecture diagnosis model. The structural characteristics of the architecture anomaly information are used as the classification basis. The node type is used to match the inference rules to achieve the classification of the anomaly information. A total of 49 negative samples were correctly identified in the autonomous transportation system architecture anomaly information traceability. After classification by the diagnosis model, some of the sample results are as follows Figure 3After being processed by the diagnostic model, all 49 negative samples are mapped to the corresponding abnormal information types, reflecting that the current architecture diagnostic rules can well cover the common abnormal information in the architecture, such as Figure 4 As shown;

[0072] S3: The diagnostic process ends;

[0073] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.

Claims

1. A rule-based reasoning-based autonomous traffic system architecture diagnosis method, characterized by: The method comprises the following steps: S1: Abnormal information tracing of autonomous transportation system architecture based on attribute graph; S2: Based on the abnormal information tracing results obtained in S1, the abnormal information is classified according to the diagnosis rules; S3: The diagnostic process ends; Said S1 comprises the following steps: S11: Constructing the reference design set R and architecture dataset S of autonomous transportation system architecture through the autonomous transportation system theoretical framework and logical, functional, and physical views; S12: The reference design set R and the architecture dataset S are processed into labeled, directed reference design set attribute graph R' and architecture dataset attribute graph S'; S13: The autonomous transportation system architecture anomaly detection model will find the largest common part of the reference design set attribute graph R' and the architecture dataset attribute graph S'; S14: Find the complement of the common part of the input architecture dataset attribute graph S' and the reference design set attribute graph R'.

2. The autonomous traffic system architecture diagnosis method based on rule reasoning according to claim 1, characterized in that: The S2 comprises the following steps: S21: The autonomous transportation system architecture diagnosis model detects abnormalities in the autonomous transportation system architecture according to the diagnosis rules; S22: Classify the abnormal information tracing results obtained by the autonomous traffic system architecture diagnosis model.

3. The autonomous traffic system architecture diagnosis method based on rule reasoning according to claim 2, characterized in that: The abnormal information of the autonomous traffic system architecture includes isolated node abnormal information, functional architecture abnormal information, logical architecture abnormal information and physical architecture abnormal information.

4. The autonomous traffic system architecture diagnosis method based on rule reasoning according to claim 3, characterized in that: The diagnosis rules include diagnosis rules for isolated nodes, diagnosis rules for functional architecture, diagnosis rules for physical architecture and diagnosis rules for logical architecture.

5. The autonomous traffic system architecture diagnosis method based on rule reasoning according to claim 4, characterized in that: The diagnosis rules for the isolated nodes are: Rule1:v in Nodes but v not exists in Edges When this rule is executed, the points in the node set will be traversed. If the point has not appeared in the edge set, it will be judged as an isolated point, that is, this type of node has no link relationship with other nodes.

6. The autonomous traffic system architecture diagnosis method based on rule reasoning according to claim 4, characterized in that: The diagnostic rules of the logical architecture are: Rule9:source.type=function&&target.type=function When this rule is executed, if the input type and function match and the target type and function match, the logic between the functions is judged to be reasonable, otherwise it is unreasonable.

7. The autonomous traffic system architecture diagnosis method based on rule reasoning according to claim 1, characterized in that: This method splits the knowledge required for diagnosis into attribute graphs and rules, and locates abnormal information in the architecture through a graph-based autonomous traffic system architecture anomaly detection model.

Images