CN116026612A - Vehicle risk assessment method and vehicle - Google Patents
Vehicle risk assessment method and vehicle Download PDFInfo
- Publication number
- CN116026612A CN116026612A CN202310015425.9A CN202310015425A CN116026612A CN 116026612 A CN116026612 A CN 116026612A CN 202310015425 A CN202310015425 A CN 202310015425A CN 116026612 A CN116026612 A CN 116026612A
- Authority
- CN
- China
- Prior art keywords
- risk assessment
- vehicle
- index
- evaluation
- indexes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012502 risk assessment Methods 0.000 title claims abstract description 158
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000011156 evaluation Methods 0.000 claims abstract description 168
- 238000000513 principal component analysis Methods 0.000 claims abstract description 30
- 238000012911 target assessment Methods 0.000 claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims description 46
- 210000002569 neuron Anatomy 0.000 claims description 23
- 230000006870 function Effects 0.000 claims description 16
- 238000004458 analytical method Methods 0.000 claims description 10
- 238000012549 training Methods 0.000 description 7
- 238000013210 evaluation model Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000012847 principal component analysis method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a vehicle risk assessment method and a vehicle. Wherein the method comprises the following steps: acquiring risk assessment data of a vehicle, wherein the risk assessment data comprises: an evaluation index of a plurality of dimensions; performing principal component analysis on the evaluation indexes of the multiple dimensions to obtain target evaluation indexes in the evaluation indexes of the multiple dimensions; and performing risk assessment on the target assessment index by using a risk assessment model to obtain a risk assessment result of the vehicle, wherein the risk assessment result is used for representing the level of safety risk of the vehicle. The invention solves the technical problem of low evaluation accuracy of risk evaluation on the vehicle in the prior art.
Description
Technical Field
The invention relates to the field of vehicle testing, in particular to a vehicle risk assessment method and a vehicle.
Background
With the development of intelligent networking of automobiles, in-car remote communication Terminals (TBOX), in-car infotainment systems, meters and the like are widely applied, and the interconnectivity of vehicles is greatly improved. Meanwhile, the risk points and the safety access points of the vehicle are also improved, and particularly the functions of wireless network (Wireless Fidelity, WIFI) hot spots, bluetooth, mobile network and the like of the vehicle are improved, so that the attack surface of the vehicle terminal is large and the risk is high, and therefore, risk assessment on the vehicle is an important means for improving the safety of the vehicle.
However, in the prior art, when the risk assessment is performed on the vehicle, the steps are complicated and a lot of manpower and time are required, which results in low assessment accuracy of the risk assessment on the vehicle.
In view of the above problems, no effective solution has been proposed at present.
Disclosure of Invention
The embodiment of the invention provides a vehicle risk assessment method and a vehicle, which at least solve the technical problem of low assessment accuracy of risk assessment on the vehicle in the prior art.
According to an aspect of an embodiment of the present invention, there is provided a vehicle risk assessment method including: acquiring risk assessment data of a vehicle, wherein the risk assessment data comprises: an evaluation index of a plurality of dimensions; performing principal component analysis on the evaluation indexes of the multiple dimensions to obtain target evaluation indexes in the evaluation indexes of the multiple dimensions; and performing risk assessment on the target assessment index by using a risk assessment model to obtain a risk assessment result of the vehicle, wherein the risk assessment result is used for representing the level of safety risk of the vehicle.
Optionally, performing principal component analysis on the evaluation indexes of the multiple dimensions to obtain a target evaluation index of the evaluation indexes of the multiple dimensions, including: generating an index matrix based on the evaluation indexes of the multiple dimensions, wherein elements of each column in the index matrix are used for representing the evaluation indexes of different dimensions; determining characteristic values of the evaluation indexes of the multiple dimensions based on the index matrix; the target evaluation index is determined based on the characteristic values of the evaluation indexes of the multiple dimensions.
Optionally, determining the feature value of the evaluation index of the multiple dimensions based on the index matrix includes: determining a covariance matrix of the index matrix; characteristic values of the evaluation indexes of the multiple dimensions are determined based on the covariance matrix.
Optionally, determining the target evaluation index based on the feature values of the evaluation index of the multiple dimensions includes: determining variance contribution rates of the evaluation indexes of the multiple dimensions based on the characteristic values of the evaluation indexes of the multiple dimensions; descending order sorting is carried out on the evaluation indexes of the multiple dimensions according to the variance contribution ratio, and a sorting result is obtained; and obtaining a target evaluation index from the sorting result, wherein the sum of the variance contribution rates of the target evaluation index is larger than a preset contribution rate.
Optionally, determining the variance contribution ratio of the evaluation index of the multiple dimensions based on the eigenvalues of the evaluation index of the multiple dimensions includes: determining any one of the evaluation indexes of the multiple dimensions as a first evaluation index; obtaining the sum of characteristic values of the evaluation indexes of a plurality of dimensions to obtain a total characteristic value; and obtaining the ratio of the characteristic value of the first evaluation index to the total characteristic value to obtain the variance contribution rate of the first evaluation index.
Optionally, the risk assessment model comprises an input layer, at least one hidden layer and an output layer, and the risk assessment model is used for carrying out risk assessment on target assessment indexes to obtain a risk assessment result of the vehicle, and the risk assessment method comprises the following steps: acquiring a target evaluation index by using an input layer; extracting characteristics of the target evaluation index by utilizing at least one hidden layer to obtain index characteristics; and carrying out risk assessment on the index features by using an output layer to obtain a risk assessment result.
Optionally, the input layer contains the same number of neurons as the target evaluation index, and the hidden layer contains 6 neurons.
Optionally, the method further comprises: acquiring sample data acquired in a real environment, wherein the sample data comprises sample indexes of a plurality of dimensions; performing principal component analysis on the sample indexes of the multiple dimensions to obtain target sample indexes in the sample indexes of the multiple dimensions; performing risk assessment on target sample indexes by using a risk assessment model to obtain a sample assessment result of the vehicle; constructing a loss function of the risk assessment model based on the sample assessment result and the real assessment result of the sample data; and adjusting model parameters of the risk assessment model based on the loss function of the risk assessment model.
Optionally, acquiring risk assessment data of the vehicle includes: carrying out attack feasibility analysis on the vehicle to obtain attack feasibility grades of the vehicle, wherein the attack feasibility grades comprise a plurality of indexes of a first dimension; performing influence grading on the vehicle to obtain a vulnerability influence grade of the vehicle, wherein the vulnerability influence grade comprises a plurality of indexes of a second dimension; and obtaining risk assessment data based on the attack feasibility level and the vulnerability influence level.
According to another aspect of the embodiment of the present invention, there is also provided a vehicle risk assessment apparatus including: the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring risk assessment data of a vehicle, and the risk assessment data comprises: an evaluation index of a plurality of dimensions; the analysis module is used for carrying out principal component analysis on the evaluation indexes of the multiple dimensions to obtain target evaluation indexes in the evaluation indexes of the multiple dimensions; the evaluation module is used for performing risk evaluation on the target evaluation index by using the risk evaluation model to obtain a risk evaluation result of the vehicle, wherein the risk evaluation result is used for representing the level of safety risk of the vehicle.
According to another aspect of an embodiment of the present invention, there is also provided a vehicle including: a memory storing an executable program; and the processor is used for running a program, wherein the program executes the vehicle risk assessment method of any one of the above steps.
According to another aspect of the embodiments of the present invention, there is also provided a non-volatile storage medium including a stored program, wherein the vehicle risk assessment method of any one of the above is executed in a processor of a device where the program is controlled when running.
In the embodiment of the invention, the risk assessment data of the vehicle is acquired, wherein the risk assessment data comprises: an evaluation index of a plurality of dimensions; performing principal component analysis on the evaluation indexes of the multiple dimensions to obtain target evaluation indexes in the evaluation indexes of the multiple dimensions; and performing risk assessment on the target assessment index by using a risk assessment model to obtain a risk assessment result of the vehicle, wherein the risk assessment result is used for representing the mode of the safety risk level of the vehicle. It is easy to notice that the risk assessment result is obtained by carrying out risk assessment on the target assessment index through the risk assessment model, the target assessment index is obtained by carrying out principal component analysis on the assessment indexes with multiple dimensions, misjudgment in the assessment process is avoided through principal component analysis and the risk assessment model, meanwhile, the assessment efficiency is improved, the purpose of accurately obtaining the risk assessment result is achieved, the technical effect of improving the assessment accuracy of the risk assessment is achieved, and the technical problem that the assessment accuracy of the risk assessment on vehicles is low in the prior art is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:
FIG. 1 is a flow chart of a vehicle risk assessment method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an alternative risk assessment model according to an embodiment of the present invention;
FIG. 3 is a flow chart of an alternative vehicle risk assessment method according to an embodiment of the present invention;
fig. 4 is a schematic structural view of a vehicle risk assessment apparatus according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
According to an embodiment of the present invention, there is provided an embodiment of a vehicle risk assessment method, it being noted that the steps shown in the flowchart of the drawings may be performed in a computer system such as a set of computer executable instructions, and, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in an order other than that shown or described herein.
Fig. 1 is a flowchart of a vehicle risk assessment method according to an embodiment of the present invention, as shown in fig. 1, the method including the steps of:
step S102, acquiring risk assessment data of the vehicle, wherein the risk assessment data includes: evaluation indexes of multiple dimensions.
Alternatively, the risk assessment data of the vehicle may be acquired by: carrying out attack feasibility analysis on the vehicle to obtain attack feasibility grades of the vehicle, wherein the attack feasibility grades comprise a plurality of indexes of a first dimension; performing influence grading on the vehicle to obtain a vulnerability influence grade of the vehicle, wherein the vulnerability influence grade comprises a plurality of indexes of a second dimension; and obtaining risk assessment data based on the attack feasibility level and the vulnerability influence level.
The attack feasibility level may be a level of attack difficulty of the vehicle vulnerabilities in the plurality of first dimensions.
The plurality of first dimensions described above may include, but are not limited to: attack time A 1 Experience and knowledge A of attacker 2 Degree of awareness A of attacker of vehicle 3 Attack opportunity A 4 Tool A required for attack 5 。
The indexes of the first dimensions may be indexes obtained by assigning values to the first dimensions by 1-5, where the larger the index value of the first dimension is, the easier the vulnerability of the vehicle corresponding to the first dimension is attacked.
In an alternative embodiment, the attack feasibility level of the vehicle may be obtained by first performing an attack feasibility analysis on the vehicle, where the attack feasibility of the vehicle may include, but is not limited to: a is that 1 ＝{1,2,3,4,5}，A 2 ＝{1,2,3,4,5}，A 3 ＝{1,2,3,4,5}，A 4 ＝{1,2,3,4,5}，A 5 = {1,2,3,4,5}. If A 1 =5, indicate a 1 The more vulnerable the vulnerability of the corresponding vehicle is, if A 1 =1, indicate a 1 The less vulnerable the corresponding vehicle's vulnerability is. If A 2 =5, indicate a 2 The more vulnerable the vulnerability of the corresponding vehicle is, if A 2 =1, indicate a 2 The less vulnerable the corresponding vehicle's vulnerability is.
The vulnerability impact level may be a level of influence of the vehicle vulnerabilities in the plurality of second dimensions.
The plurality of second dimensions described above may include, but are not limited to: personal safety B 1 Property loss B 2 Vehicle running state B 3 Privacy regulations B 4 National society B 5 。
The indexes of the second dimensions may be indexes obtained by assigning values to the second dimensions by 1-5, where the larger the index value of the second dimension is, the deeper the degree of influence of the vulnerability of the vehicle corresponding to the second dimension is.
In another alternative embodiment, the vulnerability impact level of the vehicle may be obtained by rating the impact of the vehicle, where the vulnerability impact level may include, but is not limited to: b (B) 1 ＝{1,2,3,4,5}，B 2 ＝{1,2,3,4,5}，B 3 ＝{1,2,3,4,5}，B 4 ＝{1,2,3,4,5}，B 5 = {1,2,3,4,5}. If B 1 =5, indicate B 1 The deeper the vulnerability of the corresponding vehicle affects, if B 1 =1, indicate B 1 The shallower the degree to which the corresponding vehicle's vulnerability is affected. If B 2 =5, indicate B 2 The greater the degree of influence of the corresponding vehicle's vulnerability, if B 2 =1, indicate B 2 The shallower the degree to which the corresponding vehicle's vulnerability is affected.
In yet another alternative embodiment, after obtaining the attack feasibility level of the vehicle and the vulnerability impact level of the vehicle, all the attack feasibility levels and the vulnerability impact levels may be summarized, so that risk assessment data may be obtained, that is, the risk assessment data may be a 1 ＝{1,2,3,4,5}，A 2 ＝{1,2,3,4,5}，A 3 ＝{1,2,3,4,5}，A 4 ＝{1,2,3,4,5}，A 5 ＝{1,2,3,4,5}，B 1 ＝{1,2,3,4,5}，B 2 ＝{1,2,3,4,5}，B 3 ＝{1,2,3,4,5}，B 4 ＝{1,2,3,4,5}，B 5 ＝{1,2,3,4,5}。
Step S104, performing principal component analysis on the evaluation indexes of the multiple dimensions to obtain target evaluation indexes in the evaluation indexes of the multiple dimensions.
Alternatively, the target evaluation index may be obtained by: generating an index matrix based on the evaluation indexes of the multiple dimensions, wherein elements of each column in the index matrix are used for representing the evaluation indexes of different dimensions; determining characteristic values of the evaluation indexes of the multiple dimensions based on the index matrix; the target evaluation index is determined based on the characteristic values of the evaluation indexes of the multiple dimensions.
The target evaluation index described above may be an index that is an input to the risk evaluation model.
The principal component analysis may be principal component analysis (Principal Component Analysis, PCA), but is not limited thereto.
The index matrix may be a matrix in which a plurality of different dimensions are represented in rows and different indices for each dimension are represented in columns. For example, it may be R m×n . Wherein R represents an index matrix, m represents risk assessment data with m rows and m columns of different dimensions, and n represents assessment indexes with n columns.
In an alternative embodiment, the risk assessment data of different dimensions may be used as rows and the multiple indicators of each dimension may be used as columns by PCA, so as to obtain an indicator matrix X, where X e R m×n 。
Alternatively, the feature value may be obtained by: determining a covariance matrix of the index matrix; characteristic values of the evaluation indexes of the multiple dimensions are determined based on the covariance matrix.
The covariance matrix may be Y, and a specific calculation formula is as follows:
wherein X is an index matrix, X T For the transposed matrix of X, m is the number of risk assessment data of different dimensions, namely the rows of the index matrix, R n*n Risk assessment data representing n different dimensions of a behavior are listed as n assessment indicators.
The characteristic value of the evaluation index of the multiple dimensions may be λ, and the specific calculation formula is:
λ(I-Y)α＝0，
wherein I represents an identity matrix, alpha is a feature vector corresponding to a feature value lambda, and the value can be alpha i ，i＝0,…,n。
In an alternative embodiment, the transpose matrix X of the index matrix X, X may be passed T And the risk evaluation data of m different dimensions determines a covariance matrix Y of the index matrix X, and then the eigenvalue lambda of the evaluation index of the multiple dimensions can be determined based on the identity matrix I, the covariance matrix Y and the eigenvector alpha corresponding to the eigenvalue lambda.
Alternatively, the target evaluation index may be determined by: determining variance contribution rates of the evaluation indexes of the multiple dimensions based on the characteristic values of the evaluation indexes of the multiple dimensions; descending order sorting is carried out on the evaluation indexes of the multiple dimensions according to the variance contribution ratio, and a sorting result is obtained; and obtaining a target evaluation index from the sorting result, wherein the sum of the variance contribution rates of the target evaluation index is larger than a preset contribution rate.
Alternatively, the variance contribution rate may be determined by: determining any one of the evaluation indexes of the multiple dimensions as a first evaluation index; obtaining the sum of characteristic values of the evaluation indexes of a plurality of dimensions to obtain a total characteristic value; and obtaining the ratio of the characteristic value of the first evaluation index to the total characteristic value to obtain the variance contribution rate of the first evaluation index.
The first evaluation index may be λ z Wherein z is less than or equal to n.
The total characteristic value may be lambda 0 +···+λ n 。
The variance contribution ratio may be a value obtained by dividing the eigenvalue λ of a certain eigenvector α by the sum of the eigenvalues λ of all eigenvectors α, and may be expressed as
Wherein z is less than or equal to n.
The preset contribution rate may be a contribution rate set by a user in advance and capable of screening out target evaluation indexes, and when the sum of variance contribution rates of the plurality of evaluation indexes is greater than the preset contribution rate, the plurality of evaluation indexes may be determined to be target evaluation indexes.
In another alternative embodiment, any one of the multiple-dimensional evaluation indexes may be first determined as the first evaluation index λ z Secondly, the sum of the characteristic values of the evaluation indexes of a plurality of dimensions can be obtained to obtain a total characteristic value lambda 0 +···+λ n Finally, the ratio of the characteristic value to the total characteristic value of the first evaluation index can be obtained to obtain the variance contribution rate of the first evaluation index as
Wherein z is less than or equal to n.
In another alternative embodiment, the variance contribution rates of the evaluation indexes of the multiple dimensions may be sorted in descending order to obtain a sorting result, and finally the multiple variance contribution rates may be sequentially added, and in response to the sum of the variance contribution rates being greater than the preset contribution rate, the evaluation index corresponding to each variance contribution rate in the sum of the variance contribution rates may be determined to be the target evaluation index.
And S106, performing risk assessment on the target assessment index by using a risk assessment model to obtain a risk assessment result of the vehicle, wherein the risk assessment result is used for representing the level of safety risk of the vehicle.
Optionally, the risk assessment model includes: the input layer, at least one hidden layer and the output layer utilize the risk assessment model to carry out risk assessment to the target evaluation index, obtain the risk assessment result of vehicle, include: acquiring a target evaluation index by using an input layer; extracting characteristics of the target evaluation index by utilizing at least one hidden layer to obtain index characteristics; and carrying out risk assessment on the index features by using an output layer to obtain a risk assessment result.
Optionally, the input layer contains the same number of neurons as the target evaluation index, and the hidden layer contains 6 neurons.
The risk assessment model described above may be, but is not limited to, a neural network model. The risk assessment model may include, but is not limited to: an input layer, at least one hidden layer, and an output layer. In the present embodiment, one input layer, two hidden layers, and one output layer are described as an example, but not limited thereto.
Note that, the number of neurons included in the input layer is the same as the number of target evaluation indexes, and the hidden layer may include a plurality of neurons, and in this embodiment, 6 neurons are illustrated as an example, but not limited thereto.
FIG. 2 is a schematic diagram of an alternative risk assessment model according to an embodiment of the present invention, as shown in FIG. 2, the risk assessment model includes: an input layer comprising z neurons, wherein the neurons are C 1 ,···,C z Two hidden layers comprising 6 neurons and an output layer, wherein the output layer comprises one neuron SL. W (W) 11 Represent C 1 Connection of a neuron to a first neuron in the hidden layer, W 12 Represent C 1 Connection of a neuron to a second neuron in the hidden layer, and so on, can yield W 16 And (5) connecting wires.
The index feature may be H, and the specific calculation formula is:
wherein H represents index features, w 1i Representing the weight, c i Representing input of hidden layer, b i Representing the bias value of the current layer.
Note that, the function used by the hidden layer is the activation function ReLU, and the expression is f (x) =max (0, x), which indicates that the output result is the maximum value among the multiple values.
The risk assessment result d SL (m) may represent a level of risk of the vehicle, a high risk assessment result representing a greater risk of the vehicle, and a low risk assessment result representing a lower risk of the vehicle.
In an alternative embodiment, the target evaluation index may be obtained through an input layer of the risk evaluation model, the target evaluation index may be extracted through at least one hidden layer to obtain an index feature H, and the risk evaluation may be performed through an output layer to obtain a risk evaluation result d SL (m)。
Optionally, the method further comprises: acquiring sample data acquired in a real environment, wherein the sample data comprises sample indexes of a plurality of dimensions; performing principal component analysis on the sample indexes of the multiple dimensions to obtain target sample indexes in the sample indexes of the multiple dimensions; performing risk assessment on target sample indexes by using a risk assessment model to obtain a sample assessment result of the vehicle; constructing a loss function of the risk assessment model based on the sample assessment result and the real assessment result of the sample data; and adjusting model parameters of the risk assessment model based on the loss function of the risk assessment model.
The sample data may be a plurality of data collected by a user from a plurality of real vehicles in advance, wherein the sample data includes sample indexes of a plurality of dimensions.
The target sample index M may be a sample index obtained by a PCA process.
The sample evaluation result y SL (m) may be an evaluation result obtained after risk evaluation of the target sample index by the risk evaluation model.
The loss function may be L (Θ), and the specific calculation formula is:
in the embodiment of the invention, the risk assessment data of the vehicle is acquired, wherein the risk assessment data comprises: an evaluation index of a plurality of dimensions; performing principal component analysis on the evaluation indexes of the multiple dimensions to obtain target evaluation indexes in the evaluation indexes of the multiple dimensions; and performing risk assessment on the target assessment index by using a risk assessment model to obtain a risk assessment result of the vehicle, wherein the risk assessment result is used for representing the mode of the safety risk level of the vehicle. It is easy to notice that the risk assessment result is obtained by carrying out risk assessment on the target assessment index through the risk assessment model, the target assessment index is obtained by carrying out principal component analysis on the assessment indexes with multiple dimensions, misjudgment in the assessment process is avoided through principal component analysis and the risk assessment model, meanwhile, the assessment efficiency is improved, the purpose of accurately obtaining the risk assessment result is achieved, the technical effect of improving the assessment accuracy of the risk assessment is achieved, and the technical problem that the assessment accuracy of the risk assessment on vehicles is low in the prior art is solved.
According to the vehicle-mounted controller network security risk assessment method based on the PCA and the neural network, the PCA principal component analysis method and the neural network are combined, so that the one-sidedness and subjectivity of the traditional assessment method are overcome, and the problem that a large amount of manual effectiveness and time are required for traditional vehicle network security risk assessment is solved.
The specific implementation steps of the invention include the following:
1. the attack feasibility and the vulnerability impact level jointly determine the vehicle risk level. The five elements for determining the attack feasibility are attack time, experience and knowledge of an attacker, knowledge degree of the attacker on a target, attack opportunity and tools required by the attack respectively. The five elements are assigned with 1-5 as indexes, and attack time A is taken as an example 1 = {1,2,3,4,5}, the larger the value, the shorter the required attack time, the easier the vulnerability is to attack, and similarly, the assignment of the same rule is performed on the other four elements determining the attack feasibility: experience and knowledge A of attacker 2 = {1,2,3,4,5}, degree of knowledge of the target by an attacker a 3 = {1,2,3,4,5}, attack opportunity a 4 = {1,2,3,4,5}, tool a required for attack 5 = {1,2,3,4,5}. The vulnerability impact level determining factors are respectively personal safety B 1 Property loss B 2 Vehicle running state B 3 Privacy regulations B 4 National society B 5 . In addition, since the influence of the vulnerability on the individual is considered, the influence of the country and the society is also considered as one of the vulnerability influence level determining elements, denoted as B 5 . Similarly, five elements are assigned with 1-5 as indexes, and personal safety B 1 = {1,2,3,4,5}, property loss B 2 = {1,2,3,4,5}, vehicle running state B 3 = {1,2,3,4,5}, privacy regulation B 4 = {1,2,3,4,5}, national society B 5 = {1,2,3,4,5}, the greater the value, the deeper the vulnerability is affected.
2. The training set uses a risk assessment dataset of a large number of actual cases, which is derived by an experienced expert. And (3) performing dimension reduction on the training set by using Principal Component Analysis (PCA), and taking the selected principal component as the input of the neural network. The principal component analysis steps are as follows:
(1) The training set has m samples, each sample having dimension n, the samples being formed into a matrix form, each row representing a sample, each column representing a dimension, to obtain an m X n sample matrix X, X e R m×n Calculating covariance matrix of X
Y∈R n×n ；
(2) Each eigenvalue λ and corresponding eigenvector α of the covariance matrix Y is calculated by the formula λ (I-Y) α=0 i I=1,..n, I represents an identity matrix, each eigenvector α representing one principal component;
(3) Calculating the variance contribution rate of each principal component, namely dividing the eigenvalue of a certain eigenvector by the sum of the eigenvalues of all eigenvectors; and calculating the accumulated variance contribution rate, namely, the sum of the variance contribution rates of all the current feature vectors. And extracting the first z principal components according to the order of the variance contribution rate from large to small.
3. The network constructed by the invention is shown in fig. 2, and consists of an input layer, an output layer and a hidden layer. The input layer has z neurons, the hidden layer has 2 layers, each layer has 6 neurons, and the output layer has 1 neuron, namely the output is the vehicle risk level. The output of each layer is the input of the next layer, and the output of each layer is determined by the weight of the current layer and the input of the current layer. The hidden layer first layer neuron output expression is
w 1i As the weight, C i B for input 1 Is the current layer bias value. Wherein the hidden layer uses an activation function of ReLU, f (x) =max (0, x).
4. Network loss function
M is the number of data set samples, y SL (m) is the expected output of the vulnerability risk level, d SL (m) is vulnerability riskActual output of the rank. The loss function L (Θ) is the mean square error of the actual output and the desired output. The smaller the loss function, the better the parameter training, and the closer the actual output of the network to the desired output.
FIG. 3 is a flow chart of an alternative vehicle risk assessment method according to an embodiment of the invention, as shown in FIG. 3, comprising the steps of:
step S301, constructing a training data set by using risk assessment data of actual cases;
step S302, performing dimension reduction on the training set by using a principal component analysis method, and extracting principal components;
step S303, taking the selected principal component as the input of a neural network, performing machine learning on the neural network model, and training to obtain a risk assessment model;
step S304, risk assessment is carried out on the vehicle risk assessment data to be assessed by using a risk assessment model.
The vehicle-mounted controller network security risk assessment method based on the PCA and the neural network provided by the invention has the advantages that the quantification of the vehicle-mounted controller network security risk is completed, the erroneous judgment in the assessment process is avoided, the assessment efficiency is greatly improved, the risk assessment is more objective and comprehensive, and the practicability is higher.
Example 2
According to another aspect of the embodiments of the present invention, there is further provided a vehicle risk assessment device, which may perform the vehicle risk assessment method provided in the foregoing embodiment 1, and specific implementation and preferred application scenarios are the same as those of the foregoing embodiment 1, and are not described herein.
Fig. 4 is a schematic structural view of a vehicle risk assessment apparatus according to an embodiment of the present invention, as shown in fig. 4, the apparatus includes: an obtaining module 40, configured to obtain risk assessment data of the vehicle, where the risk assessment data includes: an evaluation index of a plurality of dimensions; the analysis module 42 is configured to perform principal component analysis on the evaluation indexes of multiple dimensions, so as to obtain a target evaluation index of the evaluation indexes of multiple dimensions; the evaluation module 44 is configured to perform risk evaluation on the target evaluation index by using the risk evaluation model, so as to obtain a risk evaluation result of the vehicle, where the risk evaluation result is used to represent a level of safety risk of the vehicle.
Optionally, the analysis module includes: the generating unit is used for generating an index matrix based on the evaluation indexes of the multiple dimensions, wherein elements of each column in the index matrix are used for representing the evaluation indexes of different dimensions; a first determining unit configured to determine feature values of evaluation indexes of a plurality of dimensions based on the index matrix; and a second determining unit for determining a target evaluation index based on the characteristic values of the evaluation indexes of the plurality of dimensions.
Optionally, the first determining unit includes: a first determining subunit, configured to determine a covariance matrix of the index matrix; and a second determination subunit configured to determine eigenvalues of the evaluation indexes of the multiple dimensions based on the covariance matrix.
Optionally, the second determining unit includes: a third determination subunit, configured to determine a variance contribution ratio of the evaluation indexes of the multiple dimensions based on the feature values of the evaluation indexes of the multiple dimensions; the sorting subunit is used for sorting the evaluation indexes of the multiple dimensions in a descending order according to the variance contribution rate to obtain a sorting result; and the acquisition subunit is used for acquiring target evaluation indexes from the sorting results, wherein the sum of the variance contribution rates of the target evaluation indexes is larger than a preset contribution rate.
Optionally, the third determining subunit is further configured to: determining any one of the evaluation indexes of the multiple dimensions as a first evaluation index; obtaining the sum of the characteristic values of the second evaluation indexes to obtain a total characteristic value, wherein the second evaluation indexes are evaluation indexes except the second evaluation indexes in the evaluation indexes of multiple dimensions; and obtaining the ratio of the characteristic value of the first evaluation index to the total characteristic value to obtain the variance contribution rate of the first evaluation index.
Optionally, the risk assessment model comprises an input layer, at least one hidden layer and an output layer, and the assessment module comprises: the first acquisition unit is used for acquiring target evaluation indexes by utilizing the input layer; the extraction unit is used for carrying out feature extraction on the target evaluation index by utilizing at least one hidden layer to obtain index features; the first evaluation unit is used for performing risk evaluation on the index features by utilizing the output layer to obtain a risk evaluation result.
Optionally, the input layer contains the same number of neurons as the target evaluation index, and the hidden layer contains 6 neurons.
Optionally, the evaluation module further comprises: the second acquisition unit is used for acquiring sample data acquired in a real environment, wherein the sample data comprises sample indexes of multiple dimensions; the first analysis unit is used for carrying out principal component analysis on the sample indexes of the multiple dimensions to obtain target sample indexes in the sample indexes of the multiple dimensions; the second evaluation unit is used for performing risk evaluation on the target sample indexes by using the risk evaluation model to obtain a sample evaluation result of the vehicle; the construction unit is used for constructing a loss function of the risk assessment model based on the sample assessment result and the real assessment result of the sample data; and the adjusting unit is used for adjusting the model parameters of the risk assessment model based on the loss function of the risk assessment model.
Optionally, the acquiring module includes: the second analysis unit is used for carrying out attack feasibility analysis on the vehicle to obtain attack feasibility grades of the vehicle, wherein the attack feasibility grades comprise a plurality of indexes of a first dimension; the grading unit is used for grading the influence of the vehicle to obtain a vulnerability influence grade of the vehicle, wherein the vulnerability influence grade comprises a plurality of indexes of a second dimension; and the processing unit is used for obtaining risk assessment data based on the attack feasibility grade and the vulnerability influence grade.
Example 3
According to another aspect of an embodiment of the present invention, there is also provided a vehicle including: a memory storing an executable program; and the processor is used for running a program, wherein the vehicle risk assessment method is executed when the program runs.
Example 4
According to another aspect of the embodiments of the present invention, there is also provided a non-volatile storage medium including a stored program, wherein the vehicle risk assessment method of any one of the above is executed in a processor of a device where the program is controlled when running.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, for example, may be a logic function division, and may be implemented in another manner, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (10)
1. A vehicle risk assessment method, comprising:
acquiring risk assessment data of the vehicle, wherein the risk assessment data comprises: an evaluation index of a plurality of dimensions;
performing principal component analysis on the evaluation indexes of the multiple dimensions to obtain target evaluation indexes in the evaluation indexes of the multiple dimensions;
and performing risk assessment on the target assessment index by using a risk assessment model to obtain a risk assessment result of the vehicle, wherein the risk assessment result is used for representing the level of safety risk of the vehicle.
2. The vehicle risk assessment method according to claim 1, wherein performing principal component analysis on the multiple-dimensional assessment indices to obtain a target assessment index of the multiple-dimensional assessment indices, comprises:
generating an index matrix based on the evaluation indexes of the multiple dimensions, wherein elements of each column in the index matrix are used for representing the evaluation indexes of different dimensions;
determining characteristic values of the evaluation indexes of the multiple dimensions based on the index matrix;
and determining the target evaluation index based on the characteristic values of the evaluation indexes of the multiple dimensions.
3. The vehicle risk assessment method according to claim 2, wherein determining the characteristic values of the assessment indicators of the plurality of dimensions based on the indicator matrix comprises:
determining a covariance matrix of the index matrix;
and determining characteristic values of the evaluation indexes of the multiple dimensions based on the covariance matrix.
4. The vehicle risk assessment method according to claim 2, wherein determining the target assessment index based on the characteristic values of the assessment indexes of the plurality of dimensions includes:
determining variance contribution rates of the evaluation indexes of the multiple dimensions based on the characteristic values of the evaluation indexes of the multiple dimensions;
descending order sorting is carried out on the evaluation indexes of the multiple dimensions according to the variance contribution ratio, and a sorting result is obtained;
and acquiring the target evaluation index from the sorting result, wherein the sum of the variance contribution rates of the target evaluation index is larger than a preset contribution rate.
5. The vehicle risk assessment method according to claim 4, wherein determining the variance contribution ratio of the evaluation index of the plurality of dimensions based on the feature values of the evaluation index of the plurality of dimensions includes:
determining any one of the evaluation indexes of the multiple dimensions as a first evaluation index;
obtaining the sum of the characteristic values of the evaluation indexes of the multiple dimensions to obtain a total characteristic value;
and obtaining the ratio of the characteristic value of the first evaluation index to the total characteristic value to obtain the variance contribution rate of the first evaluation index.
6. The vehicle risk assessment method according to claim 1, wherein the risk assessment model includes an input layer, at least one hidden layer and an output layer, and the risk assessment is performed on the target assessment index by using the risk assessment model to obtain a risk assessment result of the vehicle, and the risk assessment method includes:
acquiring the target evaluation index by using an input layer;
extracting features of the target evaluation index by utilizing the at least one hidden layer to obtain index features;
and carrying out risk assessment on the index features by using the output layer to obtain the risk assessment result.
7. The vehicle risk assessment method according to claim 6, wherein the input layer contains the same number of neurons as the target assessment index, and the hidden layer contains 6 neurons.
8. The vehicle risk assessment method according to claim 6, characterized in that the method further comprises:
acquiring sample data acquired in a real environment, wherein the sample data comprises sample indexes of multiple dimensions;
performing principal component analysis on the sample indexes of the multiple dimensions to obtain target sample indexes in the sample indexes of the multiple dimensions;
performing risk assessment on the target sample indexes by using the risk assessment model to obtain a sample assessment result of the vehicle;
constructing a loss function of the risk assessment model based on the sample assessment result and a real assessment result of the sample data;
and adjusting model parameters of the risk assessment model based on the loss function of the risk assessment model.
9. The vehicle risk assessment method according to claim 1, characterized in that acquiring risk assessment data of the vehicle includes:
carrying out attack feasibility analysis on the vehicle to obtain an attack feasibility grade of the vehicle, wherein the attack feasibility grade comprises a plurality of indexes of a first dimension;
performing influence grading on the vehicle to obtain a vulnerability influence grade of the vehicle, wherein the vulnerability influence grade comprises a plurality of indexes of a second dimension;
and obtaining the risk assessment data based on the attack feasibility level and the vulnerability influence level.
10. A vehicle, characterized by comprising:
a memory storing an executable program;
a processor for running the program, wherein the program when run performs the vehicle risk assessment method of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310015425.9A CN116026612A (en) | 2023-01-05 | 2023-01-05 | Vehicle risk assessment method and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310015425.9A CN116026612A (en) | 2023-01-05 | 2023-01-05 | Vehicle risk assessment method and vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116026612A true CN116026612A (en) | 2023-04-28 |
Family
ID=86075519
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310015425.9A Pending CN116026612A (en) | 2023-01-05 | 2023-01-05 | Vehicle risk assessment method and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116026612A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120313875A (en) * | 2025-04-15 | 2025-07-15 | 郑州大学 | Method and device for evaluating colorimetric properties of painting illumination based on light source spectrum |
-
2023
- 2023-01-05 CN CN202310015425.9A patent/CN116026612A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120313875A (en) * | 2025-04-15 | 2025-07-15 | 郑州大学 | Method and device for evaluating colorimetric properties of painting illumination based on light source spectrum |
| CN120313875B (en) * | 2025-04-15 | 2025-11-18 | 郑州大学 | Method and device for evaluating painting illumination color rendering property based on light source spectrum |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114815605B (en) | Automatic driving test case generation method and device, electronic equipment and storage medium | |
| CN107103171B (en) | Modeling method and device of machine learning model | |
| CN112308454B (en) | Method, device, equipment and storage medium for evaluating automobile steering stability | |
| CN112735535A (en) | Prediction model training method, prediction model training device, data prediction method, data prediction device and storage medium | |
| CN104869126A (en) | Network intrusion anomaly detection method | |
| CN114139601B (en) | A method and system for evaluating artificial intelligence algorithm models in power inspection scenarios | |
| CN108830407B (en) | Sensor distribution optimization method in structural health monitoring under multi-working conditions | |
| CN118014801B (en) | A method and system for evaluating the effect of building ecological seawall oyster reefs | |
| CN112580536A (en) | High-order video vehicle and license plate detection method and device | |
| EP2434411A1 (en) | Computer-implemented method for analyzing multivariate data | |
| CN112967755A (en) | Cell type identification method for single cell RNA sequencing data | |
| CN114118292B (en) | A fault classification method based on linear discriminant neighborhood preserving embedding | |
| CN113553754A (en) | Memory, fire risk prediction model construction method, system and device | |
| CN116026612A (en) | Vehicle risk assessment method and vehicle | |
| CN115296837A (en) | SSA optimization-based sustainable integrated intrusion detection method | |
| CN111401444B (en) | Prediction method, device, computer equipment and storage medium for red wine origin | |
| CN108931815B (en) | Lithology grading identification method | |
| CN117035983A (en) | Method and device for determining credit risk level, storage medium and electronic equipment | |
| CN120105311B (en) | Vehicle-mounted data processing method and system | |
| JP6950647B2 (en) | Data determination device, method, and program | |
| CN120636235A (en) | Virtual driving scene generation method for intelligent driving school mode based on AI model | |
| CN115293560A (en) | Bank branch site selection method and system based on analytic hierarchy process | |
| CN118656633B (en) | Prediction method and device for driving style in dangerous state scene | |
| CN118506922B (en) | Soil arsenic pollution risk assessment method, device and equipment based on machine learning | |
| CN113744070A (en) | New energy automobile insurance cost prediction method and device and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |