A reliability assessment method for autonomous transportation systems

Through complex network theory, the weighted model is constructed, the network topological characteristics of autonomous transportation systems are analyzed, the importance of nodes is evaluated, and the problem that existing technology is difficult to reflect the reliability of autonomous transportation systems is solved, and the reliability evaluation and adjustment of the system is achieved. Improve the overall safety of the transportation system.

CN114595540BActive Publication Date: 2024-11-15CENT SOUTH UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210206640.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-03
Publication Date
2024-11-15
Estimated Expiration
2042-03-03

AI Technical Summary

Technical Problem

The existing traffic system framework is difficult to reflect the reliability of autonomous traffic systems from the level of system topology, resulting in the out-of-control and overall collapse of the logical architecture, threatening road traffic safety.

Method used

Using complex network theory, by obtaining equipment and data transmission related data of autonomous transportation systems, building a weighted complex network model, analyzing network topology characteristics, evaluating the importance of nodes, judging the logical architecture network type, identifying key nodes and weak links, and targeting them. Sexual adjustment.

Benefits of technology

Analyze the network type from the level of system topology, provide theoretical basis and reliability guidance, improve the overall reliability of the system, identify and adjust important nodes and weak links, and improve the safety of autonomous transportation systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114595540B_ABST
    Figure CN114595540B_ABST
Patent Text Reader

Abstract

The present invention discloses a reliability assessment method for an autonomous transportation system, comprising the following steps: obtaining relevant data of the autonomous transportation system logic architecture; constructing a complex network of the autonomous transportation system logic architecture, and determining the weights of nodes and edges according to the relevant data; selecting complex network characteristic parameters according to the network topology diagram, and analyzing the topological characteristics of the autonomous transportation system logic architecture network; selecting complex network characteristic parameters, and judging the type of the autonomous transportation system logic architecture network according to the characteristic conditions of the complex network; selecting three evaluation methods according to the type and topological characteristics of the architecture network to assess the importance of nodes in the network, and identifying weak nodes that need to be maintained. The present invention can make targeted adjustments to important nodes and weak links in the system, and fundamentally improve the reliability of the system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous traffic systems, and in particular to a reliability assessment method for an autonomous traffic system. Background Art

[0002] With the rapid improvement of the self-organization and autonomous service capabilities of the transportation system, the transportation system has shown more autonomous characteristics, and the existing transportation system framework is difficult to further effectively guide the operation of the new generation of complex transportation systems. The Autonomous Transportation System (ATS) system framework came into being. In the ATS logical architecture system, each information transmission and processing link plays a vital role. Once a link in the logical architecture is damaged, it may cause the loss of information data, resulting in the loss of control of the corresponding link or even the collapse of the entire system, which seriously threatens road traffic safety. Summary of the invention

[0003] The purpose of the present invention is to provide a reliability assessment method for an autonomous traffic system, so as to overcome the problem that the existing traffic system framework assessment method is difficult to reflect its reliability from the level of system topology.

[0004] The technical solution of the present invention to achieve the above-mentioned purpose is: a method for evaluating the reliability of an autonomous traffic system, comprising the following steps:

[0005] Step 1: Obtain relevant data on the physical nodes and data transmission of the devices and management departments of the autonomous transportation system logical architecture;

[0006] Step 2: Determine the edge weights and node weights, build a weighted complex network model, and draw a network topology diagram of the architecture;

[0007] Step 3: According to the complex network model, select indicators to analyze and evaluate the network topology characteristics of the logical architecture;

[0008] Step 4: Select parameters based on complex network judgment conditions to determine the network type characteristics of the logical architecture;

[0009] Step 5: Based on the characteristics of complex networks, three indicators are used to evaluate the importance of nodes;

[0010] The technical effect of the present invention is that the present invention uses complex network theory to analyze and evaluate the reliability of the ATS logical architecture, breaking through the limitations of existing traffic system framework evaluation methods that focus on impact and benefits, analyzing its network type from the level of system topology, and proposing corresponding complexity indicators for nodes and the system as a whole, which can provide a theoretical basis and reliability guidance for the construction of the ATS logical architecture. In addition, the present invention can make targeted adjustments to important nodes and weak links in the system, fundamentally improving the reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a flow chart of the present invention.

[0012] Figure 2 Schematic diagram of the ATS logical architecture of the autonomous and safe automatic driving scenario in the present invention.

[0013] Figure 3 This is a topological diagram of the logical architecture in the present invention.

[0014] Figure 4 It is the cumulative degree distribution curve in the present invention.

[0015] Figure 5 It is the cumulative intensity distribution curve in the present invention. DETAILED DESCRIPTION

[0016] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0017] See also Figure 1 , Figure 1 The flowchart of the present invention. The reliability assessment method of an autonomous traffic system of the present invention takes the ATS logical architecture of the road autonomous safe automatic driving scene as the research object, and the steps are as follows: 1. Obtain the relevant data of the physical nodes and data transmission of the equipment, management departments and other entities of the logical architecture of the autonomous traffic system; 2. Determine the weight of the edge and the weight of the node, and construct a weighted complex network model; 3. Select parameters such as node degree and degree distribution, node strength and strength distribution, clustering coefficient, average path length and network diameter, and global efficiency to analyze the network topology characteristics; 4. Determine the type of complex network according to indicators such as node degree distribution, node strength distribution, average path length, clustering coefficient and so on of the logical architecture network; 5. Evaluate the importance of nodes and the reliability of the logical architecture network according to the node strength method, node betweenness method and mutual information method.

[0018] The specific detailed process of the present invention is as follows:

[0019] 1. Get data

[0020] The ATS logical architecture system of the road autonomous safety autonomous driving scenario collects, stores, transmits and communicates data from user terminals, traffic facilities, management departments, etc. through a unified standard. While ensuring the security of information transmission, it realizes information sharing and information resource sharing to support the self-organization operation of various services. The schematic diagram is as follows Figure 2 shown.

[0021] 2. Build a logical architecture complex network

[0022] 2.1 Building a complex network model

[0023] The ATS logical architecture network G = (V, E) is an undirected weighted network, and the weights are similarity weights, that is, weights that express the coupling intimacy between nodes. The network includes two basic elements: nodes and edges. Among them, nodes are defined as the infrastructure and management departments that complete data flow transmission in the network G, and V is the node set; edges are defined as data flows transmitted between nodes in the network G, and E is the edge set.

[0024] 2.2 Calculating edge weights

[0025] In the ATS logical architecture network of the road autonomous safety automatic driving scenario, the weight of the edge is determined by the number of sub-services served and the importance of the sub-service. The importance of the sub-service can be objectively evaluated based on actual data. The present invention uses the hierarchical analysis method to obtain the weight coefficient P of the sub-service in the network. m The edge weight w ij It can be further defined as the sum of the weight coefficients of all sub-services served by edge (i, j):

[0026]

[0027] Among them, M ij The set of sub-services served by edge (i, j).

[0028] 2.3 Calculating Node Weight

[0029] In the ATS logical architecture network of the road autonomous safety autonomous driving scenario, the node weight q i It is calculated by the sum of the weights of all edges connected to point i:

[0030] q i =∑ j∈V w ij (2)

[0031] Draw the network topology diagram of the architecture (such as Figure 3 ). The width of the line reflects the weight of the edge, and the radius and color of the node reflect the weight of the node.

[0032] 3Analyze the logical architecture network topology characteristics

[0033] According to the ATS logical architecture network topology diagram of the road autonomous safe automatic driving scenario, its network type is determined, and indicators such as degree distribution, intensity distribution, average path length, clustering coefficient, and global efficiency are selected to analyze and evaluate the network topology characteristics of the logical architecture.

[0034] 3.1 Selecting complex network characteristic parameters

[0035] Step 3.1.1, node degree and degree distribution evaluation parameters. Node degree refers to the number of data flows directly transmitted between a node i and its neighboring nodes, that is, k i =∑ j∈V e ij , e ij ∈E. The degree distribution of the network represents the proportion of nodes with degree k in the network, that is, Where N represents the total number of network nodes.

[0036] Step 3.1.2, node strength and strength distribution. Node strength is the sum of the weights of all data flows connected to node i, that is, s i =∑ j∈V w ij The strength distribution of the network represents the proportion of nodes with strength value s in the network, that is,

[0037] Step 3.1.3, clustering parameters. The clustering coefficient is an indicator that reflects the local relationship of nodes. The clustering coefficient C i Represents the degree of aggregation between node i and its neighboring nodes:

[0038]

[0039] Among them, E i Indicates the number of data flows actually connected to node i.

[0040] The average clustering coefficient C of the network is the average of the sum of the clustering coefficients of all nodes in the network G, that is,

[0041]

[0042] Step 3.1.4, average path length and network diameter. Average path length and network diameter are important indicators for measuring network compactness. The average path length L represents the arithmetic mean of the shortest transmission path between any two nodes in the network:

[0043]

[0044] Among them, d ijThe network diameter D represents the maximum value of the shortest transmission path, that is, D = max d ij , i, j∈V, i<j.

[0045] Step 3.1.5, global efficiency. Global efficiency can reflect the strength of the reliability of connectivity between nodes. The global efficiency E of network G is g (G) can be expressed as:

[0046]

[0047] 3.2 Analysis of complex network characteristics

[0048] The basic characteristic values ​​of the ATS logical architecture network for road autonomous safe autonomous driving scenarios are calculated according to the formula, as shown in Appendix 2.

[0049] Appendix 2 Basic characteristic values ​​of the network

[0050]

[0051] It can be seen from Appendix 2 that the nodes in the ATS logical architecture network of the road autonomous safe automatic driving scenario have extensive connections, and a small number of nodes have high connectivity, which undertakes more important tasks in the process of data transmission and function realization. The overall structure of the network is compact, and the transmission of data streams between nodes does not require much processing.

[0052] 4. Determine the logical architecture network type

[0053] 4.1 Determine the complex network judgment conditions

[0054] Determining the type of ATS logical architecture network for road autonomous safety autonomous driving scenarios is the basis for analyzing it using complex network theory. Real networks are generally divided into four categories: regular networks, random networks, scale-free networks, and small-world networks. The complex network judgment conditions are shown in Appendix 1.

[0055] Appendix 1 Complex network judgment conditions

[0056]

[0057]

[0058] N represents the total number of network nodes; K represents the average node degree of the network.

[0059] 4.2 Analysis of complex network types

[0060] According to the complex network judgment conditions and combined with the calculation and analysis of complex network characteristic parameters, the type of ATS logical architecture network for road autonomous safe automatic driving scenarios is determined.

[0061] By calculating the characteristic parameters of a random network of the same size, the average path length L of the random network is obtained. R =2.65, average clustering coefficient C R =0.08. According to Appendix 2, the average path length of the logical architecture network is L = 2.02≈L R , average clustering coefficient C = 0.48>>C R , which meets the judgment criteria of small-world networks.

[0062] In order to eliminate disturbances, the cumulative degree distribution P(k≥k i ) and cumulative intensity P(s≥s i ) distribution. Figure 4 and Figure 5 It can be seen that the cumulative distribution curve can be approximately fitted into a distribution function P(k≥k i )=2.183k -1.033 , determination coefficient R 2 =0.9652, the distribution curve of the cumulative intensity can be approximately fitted into a distribution function of P(s≥s i )=0.29636k -0.7667 , determination coefficient R 2 =0.9373, which meets the judgment criteria of scale-free networks.

[0063] 5. Evaluate Logical Architecture Network Reliability

[0064] Since the ATS logical architecture network of the road autonomous safety automatic driving scenario has a scale-free characteristic, that is, a small number of nodes have a large number of data flows, which has a significant impact on the reliability of the network. Therefore, identifying these nodes is the key to improving network reliability. The present invention uses multiple indicators such as node strength method, node betweenness method and mutual information method to evaluate the importance of nodes.

[0065] 5.1 Node strength method

[0066] Compared with node degree, node strength can better reflect the key position of nodes in data flow transmission in the ATS logical architecture network of road autonomous safety autonomous driving scenarios. The greater the node strength, the more important the node is in data transmission and function realization in the network. It can be calculated that the nodes with higher node strength are autonomous driving vehicles, traffic operation management departments, and emergency vehicles, which are more important in data transmission.

[0067] 5.2 Node Betweenness Method

[0068] In the ATS logical architecture network of the road autonomous safety autonomous driving scenario, the betweenness reflects the key position of the node in the shortest path of data transmission during the data transmission process. The larger the betweenness, the more shortest transmission paths pass through the node. The node is an intermediate node for multiple data stream transmissions, and the greater its influence in data transmission. The betweenness refers to the ratio of the number of paths passing through the node to the total number of shortest paths in the network, expressed as:

[0069]

[0070] Among them, n jk represents the number of shortest paths between nodes j and k, n jk (i) represents the number of nodes i passed through in the shortest path between nodes j and k.

[0071] The calculation shows that the average betweenness value is 0.041, and 88.89% of the node betweenness is less than 0.1, indicating that only a small number of nodes are intermediate nodes for multiple data stream transmissions. The nodes with significantly higher node betweenness are traffic facility monitoring equipment, autonomous driving vehicles, and communication equipment, which are intermediate nodes for multiple data stream transmissions in this architecture network.

[0072] 5.3 Mutual Information Method

[0073] The mutual information method treats a complex network as a complex communication system model, treats edges as information flows, and uses the amount of information contained in each node to represent the importance of the node. In the ATS logical architecture network of the road autonomous safety autonomous driving scenario, the amount of node information represents the amount of data flow information contained in the node during the data transmission process.

[0074] For the ATS logical architecture network of road autonomous safety autonomous driving scenarios, mutual information I ij The calculation formula is as follows:

[0075]

[0076] Among them, P ij =w ij / s i represents the probability of data flow (i, j) for all connected data flows of node i, P ji =w ji / s j It represents the probability of data flow (j, i) for all connected data flows of node j.

[0077] The node information I(i) is the sum of the mutual information between node i and its neighboring nodes, that is, I(i) = ∑ j∈V I ijFrom the above definition, we can see that I ij =I ji , the sum of the information volume of the entire network The information content of all nodes is sorted, and the node with greater information content is more important.

[0078] The calculation shows that only 33.3% of the nodes have a volume of information greater than 0, indicating that a small number of nodes have a large amount of data flow information. The nodes with a significantly larger volume of information are the autonomous driving vehicles and the traffic operation management department. These two nodes have 29.2% of the data flow information volume and are the nodes that need to be protected in this architecture network.

[0079] 5.4 Analyzing Node Importance

[0080] Combined with the ATS logical architecture network of the road autonomous safety automatic driving scenario, the index value of each node is calculated according to the node strength method, node betweenness method and mutual information method. The importance ranking of some nodes under different indicators is shown in Appendix 3. Since the emphasis of each node importance evaluation indicator is different, the obtained node importance ranking is also different.

[0081] The strength values, betweenness values, and information volume of the three nodes of autonomous driving vehicles, traffic operation management departments, and communication base stations are all large. Therefore, these nodes have a large amount of data flow information in the network and are located in the middle nodes of a large amount of data flow transmission in the network. They are of high importance in this architecture network and need to be protected. In addition, the strength values ​​of traveler terminals and emergency vehicles are large, while the betweenness values ​​and information volume are small, but these two nodes are related to road traffic safety, so they are still of high importance in this architecture network and need to be protected. Intelligent highway signs and traffic facility monitoring equipment have large node information volume and node betweenness, but small strength values. Therefore, these two nodes have a large amount of data flow information and are located in the middle nodes of a large amount of data flow transmission, and need a certain degree of protection.

[0082] Appendix 3 Node importance ranking

[0083]

Claims

1. A reliability assessment method for autonomous transportation systems, It is characterized in that Taking the ATS logical architecture of the road autonomous safety automatic driving scenario as the research object, obtain the relevant data of the equipment, management department entity nodes and data transmission of the autonomous traffic system logical architecture; determine the weight of the edge and the weight of the node, and construct a weighted complex network model; select node degree and degree distribution, node strength and strength distribution, clustering coefficient, average path length and network diameter as well as global efficiency parameters to analyze the network topology characteristics; judge the type of complex network according to the node degree distribution, node strength distribution, average path length and clustering coefficient indicators of the logical architecture network; The node importance and logical architecture network reliability are evaluated based on the node strength method, node betweenness method and mutual information method. The specific steps include: Step S1, obtaining data on physical nodes and data transmission of devices and management departments in the logical architecture of the autonomous transportation system, collecting, storing, transmitting and communicating data from user terminals, transportation facilities and management departments through a unified standard, while ensuring the security of information transmission, realizing information sharing and information resource sharing, so as to support the self-organization operation of various services; Step S2, constructing a complex network of the autonomous transportation system logical architecture, and drawing a network topology diagram of the architecture, determining the weights of nodes and edges based on relevant data, where nodes are defined as infrastructure and management departments that complete data flow transmission in the network G, and V is a node set; edges are defined as data flows transmitted between nodes in the network G, and E is an edge set. In the ATS logical architecture network of the road autonomous safety automatic driving scenario, the weight of the edge is determined by the number of sub-services served and the importance of the sub-service; Step S3, selecting complex network characteristic parameters according to the network topology diagram, and analyzing the topological characteristics of the autonomous transportation system logical architecture network; Step S4, select the characteristic parameters of the complex network, including node degree and degree distribution, node strength and strength distribution, clustering coefficient, average path length and network diameter, and global efficiency; node degree refers to the number of data flows directly transmitted between a node i and its neighboring nodes; the degree distribution of the network represents the proportion of nodes with a degree value of k in the network; the node strength is the sum of the weights of all data flows connected to node i; the strength distribution of the network represents the proportion of nodes with a strength value of s in the network; the betweenness reflects the key position of the node in the shortest path of data transmission during the data transmission process; the node information volume represents the amount of data flow information contained in the node during the data transmission process; the type of the autonomous transportation system logical architecture network is determined according to the characteristic conditions of the complex network; Step S5, selecting three evaluation methods according to the type and topological characteristics of the architecture network to evaluate the importance of nodes in the network and identify weak nodes that need to be maintained.

2. The autonomous traffic system reliability assessment method according to claim 1, It is characterized in that Step S2 specifically includes the following steps: Step 2.1: Construct a weighted complex network model; Step 2.2: Determine the weight of the edge based on the importance of each service and its data flow; Step 2.3: Determine the node weight by the sum of the weights of the edges connected to the node.

3. The autonomous traffic system reliability assessment method according to claim 2, It is characterized in that The edge weight calculation formula in step 2.2 is: Among them, P m is the weight coefficient of the sub-service, M ij The set of sub-services served by edge (i, j).

4. The autonomous traffic system reliability assessment method according to claim 3, It is characterized in that The calculation formula for the node weight edge in step 2.3 is: i =∑ j∈V w ij .

5. The autonomous traffic system reliability assessment method according to claim 1, It is characterized in that Step S3 specifically includes the following steps: Step 3.1: Select the characteristic parameters of complex networks, including node degree and degree distribution, node strength and strength distribution, clustering coefficient, average path length and network diameter, and global efficiency; Step 3.2: Perform logical architecture network characteristic analysis based on complex network characteristic parameter values.

6. The autonomous traffic system reliability assessment method according to claim 5, It is characterized in that In step 3.1, the node degree refers to the number of data flows directly transmitted between a node i and its neighboring nodes, that is, k i =∑ j∈V e ij ,e ij ∈E; The degree distribution of the network represents the proportion of nodes with a degree value of k in the network, that is, Where N represents the total number of network nodes; The node strength is the sum of the weights of all data flows connected to node i, that is, s i =∑ j∈V w ij ; The strength distribution of the network represents the proportion of nodes with strength value s in the network, that is, Clustering coefficient C i Represents the degree of aggregation between node i and its neighboring nodes: Among them, E i Indicates the number of data flows actually connected to node i; The average clustering coefficient C of the network is the average of the sum of the clustering coefficients of all nodes in the network G, that is, The average path length L represents the arithmetic mean of the shortest transmission path between any two nodes in the network: Among them, d ij represents the shortest transmission path between any two nodes i and j, and the network diameter D represents the maximum value of the shortest transmission path, that is, D = max d ij ,i,j∈V,i <j; The global efficiency E of the network G g (G) is expressed as:

7. The autonomous traffic system reliability assessment method according to claim 1, It is characterized in that Step S4 specifically includes the following steps: Step 4.1: Select the characteristic parameters of complex networks, including node degree and degree distribution, node strength and strength distribution, clustering coefficient, and average path length; Step 4.2: Determine the network type characteristics of the logical architecture based on the complex network judgment criteria.

8. The autonomous traffic system reliability assessment method according to claim 1, It is characterized in that Step S5 specifically includes the following steps: Step 5.1: Use the node strength method to evaluate the importance of nodes; node strength can better reflect the key position of nodes in data flow transmission in the ATS logical architecture network of the road autonomous safety automatic driving scenario; the greater the node strength, the more important the node is in data transmission and function realization in the network; Step 5.2: Use the node betweenness method to evaluate the importance of the node; the betweenness reflects the key position of the node in the shortest path of data transmission during the data transmission process; Step 5.3: Use the mutual information method to evaluate the importance of nodes; the node information volume represents the amount of data flow information contained in the node during the data transmission process; Step 5.4: Analyze the important nodes in the complex network.

9. The autonomous traffic system reliability assessment method according to claim 8, It is characterized in that Step 5.2 The betweenness number refers to the ratio of the number of paths passing through the node in all the shortest paths in the network to the total number of shortest paths, expressed as: Among them, n jk represents the number of shortest paths between nodes j and k, n jk (i) represents the number of nodes i passing through the shortest path between nodes j and k; Mutual information I in step 5.3 ij The calculation formula is as follows: Among them, P ij =w ij / s i represents the probability of data flow (i, j) for all connected data flows of node i, P ji =w ji / s j It represents the probability of data flow (j,i) for all connected data flows of node j.