Function safety controllability analysis method and related equipment
By building a CART decision tree and evaluating functional safety controllability using machine learning technology, it solves the problem of strong subjectivity caused by relying on expert experience in the existing technology, and achieves more accurate and efficient functional safety analysis.
Patent Information
- Application Number
- CN202411224136.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The controllability assessment of existing functional safety hazard analysis and risk assessment relies too much on expert experience, resulting in strong subjectivity of judgment results and lack of objectivity and accuracy.
By using machine learning technology, by building a CART decision tree, optimized evaluation based on historical evaluation data, functional safety and controllability are evaluated, manual operations are reduced, and analysis accuracy and efficiency are improved.
Through deep learning technology, the accuracy of functional safety controllability judgment is improved, errors caused by subjective factors of experts are reduced, automated controllability judgment is realized, and the accuracy and efficiency of functional safety analysis are improved.
Smart Images

Figure CN119067450A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of functional safety hazard analysis and risk assessment, and in particular to a functional safety controllability analysis method and related equipment. Background Art
[0002] With the development of science and technology, machine learning technology has been widely used in various fields, especially in data processing and classification, showing extremely high efficiency and accuracy. At the same time, functional safety hazard analysis and risk assessment, as an important means to ensure the safe operation of the system, are also constantly developing and improving. Functional safety hazard analysis and risk assessment indicate that personnel can fully control hazardous events to avoid the estimation of the probability of specific injuries. For each hazardous event, the controllability of the driver or other personnel in the operating scenario should be estimated based on a certain reason.
[0003] In existing technologies, the controllability assessment of functional safety hazard analysis and risk assessment mainly relies on the experience and knowledge of experts, and the controllability is determined manually. After the determination, the final controllability is determined after multiple rounds of review by team members. However, the above method relies too much on expert experience, resulting in a highly subjective determination result, lacking objectivity and accuracy. Summary of the invention
[0004] In view of the above problems, the present invention provides a functional safety controllability analysis method and related equipment, the main purpose of which is to solve the problem that functional safety controllability analysis is too dependent on manual experience.
[0005] In order to solve at least one of the above technical problems, in a first aspect, the present invention provides a functional safety controllability analysis method, the method comprising:
[0006] Determining optimized evaluation data, wherein the optimized evaluation data is used to establish a data set;
[0007] Construct a CART decision tree based on the data set;
[0008] Functional safety controllability is evaluated based on the CART decision tree.
[0009] Optionally, determine optimization evaluation data, including:
[0010] Obtaining historical assessment data, wherein the assessment data is used to assess the functional safety controllability of historical cases, and the historical assessment data is determined based on five dimensions: failure impact, driving scenario, driving operation, people in danger, and environmental conditions;
[0011] extracting characteristic values of the historical evaluation data;
[0012] A noise removal operation and an outlier elimination operation are performed on the characteristic values to determine optimized evaluation data, wherein the optimized evaluation data is used to construct the data set.
[0013] Optionally, the above method further includes: a Gini index calculation operation,
[0014] The specific steps of the Gini index calculation operation include:
[0015] Determine the possible values for each dimension;
[0016] The data set of the corresponding dimension is divided based on the possible values to calculate the Gini index of the data set of each dimension, wherein the Gini index is used to characterize the impurity of the data set.
[0017] Optionally, the above method further includes: a child node generation operation,
[0018] The specific steps of the child node generation operation include:
[0019] Determine the optimal features and optimal binary cut-off points corresponding to the smallest Gini index;
[0020] Generating child nodes from the node based on the optimal feature and the optimal binary split point;
[0021] The optimized evaluation data in the data set is distributed to the child nodes.
[0022] Optionally, constructing a CART decision tree based on the data set includes:
[0023] Output the sub-nodes that meet the preset conditions as part of the CART decision tree;
[0024] The sub-nodes that do not meet the preset conditions are subjected to the Gini index calculation operation and the sub-node generation operation again until the preset conditions are met.
[0025] Optionally, the above method further includes:
[0026] Based on the pruning algorithm, a pruning operation is performed starting from the bottom of the CART decision tree, wherein the pruning operation is used to reduce the overfitting phenomenon of the CART decision tree.
[0027] Optionally, the above method further includes:
[0028] Determine the internal nodes of each pruning node of the CART decision tree;
[0029] Determine a first loss function of a tree with the internal node as a single node and a second loss function of a subtree with the internal node as a root node;
[0030] Performing a pruning operation based on the first loss function and the second loss function to obtain a sub-number sequence;
[0031] The optimal subtree in the sub-number sequence is determined based on a cross-validation method, wherein the optimal subtree is the CART decision tree.
[0032] In a second aspect, an embodiment of the present invention further provides a functional safety controllability analysis device, comprising:
[0033] A determination unit, used to determine optimized evaluation data, wherein the optimized evaluation data is used to establish a data set;
[0034] A construction unit, used for constructing a CART decision tree based on the data set;
[0035] An evaluation unit is used to evaluate the functional safety controllability based on the CART decision tree.
[0036] In order to achieve the above-mentioned purpose, according to a third aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the above-mentioned program is executed by a processor, the steps of the above-mentioned functional safety controllability analysis method are implemented.
[0037] In order to achieve the above-mentioned purpose, according to the fourth aspect of the present invention, there is provided an electronic device, comprising at least one processor and at least one memory connected to the processor; wherein the above-mentioned processor is used to call the program instructions in the above-mentioned memory to execute the steps of the above-mentioned functional safety controllability analysis method.
[0038] Through the above technical scheme, the functional safety controllability analysis method and related equipment provided by the present invention, for the problem that the functional safety controllability analysis is too dependent on manual experience, the present invention determines the optimized evaluation data, wherein the optimized evaluation data is used to establish a data set; a CART decision tree is constructed based on the data set; and the functional safety controllability is evaluated based on the CART decision tree. In the above scheme, by introducing machine learning technology, a method for deep learning of a large amount of historical data is used, thereby improving the accuracy of controllability judgment and reducing errors caused by subjective factors of experts. By integrating the optimized machine learning model into the safety analysis tool, automated controllability judgment is achieved, reducing the tediousness of manual operation, not only improving the accuracy and efficiency of functional safety analysis, but also realizing model optimization and automated decision-making through machine learning technology, which has significant advantages.
[0039] Correspondingly, the functional safety controllability analysis device, equipment and computer-readable storage medium provided by the embodiments of the present invention also have the above-mentioned technical effects.
[0040] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:
[0042] Figure 1 A schematic diagram of a flow chart of a functional safety controllability analysis method provided by an embodiment of the present invention is shown;
[0043] Figure 2 A schematic block diagram of the composition of a functional safety controllability analysis device provided by an embodiment of the present invention is shown;
[0044] Figure 3 A schematic block diagram of the composition of an electronic device for functional safety controllability analysis provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION
[0045] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.
[0046] In order to solve the problem that functional safety controllability analysis is too dependent on manual experience, an embodiment of the present invention provides a functional safety controllability analysis method, such as Figure 1 As shown, the method includes:
[0047] S101, determining optimization evaluation data, wherein the optimization evaluation data is used to establish a data set;
[0048] The above step S101 further includes S1011:
[0049] S1011. Acquire historical evaluation data, wherein the evaluation data is used to evaluate the functional safety controllability of historical cases, and the historical evaluation data is determined based on five dimensions: failure impact, driving scenario, driving operation, people in danger, and environmental conditions; extract characteristic values of the historical evaluation data; perform noise removal and outlier removal operations on the characteristic values to determine optimized evaluation data, wherein the optimized evaluation data is used to construct the data set.
[0050] Specifically, this application collects and organizes historical data related to functional safety hazard analysis and risk assessment, including controllability analysis cases mentioned in ISO26262 and GB34590 standards and scenario analysis and controllability determination of previous projects; at the same time, based on existing accident data and papers and other information, the driver's controllability of the vehicle is determined under different environments based on the driver's reaction time to different accident scenarios and ISO26262 and GB34590 standards.
[0051] It should be noted that the collected raw data include the impact of failure (such as: unexpected lateral control of the vehicle, too much or too little lateral control of the vehicle, unexpected longitudinal control of the vehicle, too much or too little longitudinal control of the vehicle, etc.), driving scenarios (such as: highways, urban roads, etc.), driving operations (such as: straight driving, parking, etc.), environmental conditions (such as: sunny, rainy and snowy, etc.), and people in danger (such as: people near the vehicle, other vehicles near the vehicle, etc.). The driver's controllability is scored based on these five dimensions.
[0052] Furthermore, the present application extracts eigenvalues from the five dimensions of impact of failure, driving scenarios, driving operations, environmental conditions, and people in danger using continuous natural numbers to represent eigenvalues to construct a data set, and performs data cleaning on the obtained results, including removing noise and outliers to improve data quality.
[0053] In summary, this application takes into account that the evaluation methods in the prior art have limited ability to process large amounts of data, and that a large amount of manpower and time are required to complete the data evaluation. As vehicle functions increase, controllability in a large number of scenarios needs to be analyzed, which cannot meet the needs of functional safety hazard analysis and risk assessment analysis in a big data environment. Therefore, by introducing machine learning technology, deep learning is performed on a large amount of historical data. By inputting data in five dimensions, including the impact of failure, driving scenarios, driving operations, environmental conditions, and people in danger, the trained model is used to obtain results, which can improve efficiency and complete the processing of large amounts of data faster and more accurately.
[0054] S102, constructing a CART decision tree based on the data set;
[0055] It can be understood that a decision tree is a model for decision-making based on a tree structure. It classifies data sets through multiple conditional discrimination processes and finally obtains the required results. Structurally, the starting point of the decision tree is the root node, the intermediate decision process is the internal node, and the classification result is the leaf node. The decision tree algorithm is a supervised learning algorithm that uses the ideas of classification and regression to build a mathematical model based on the characteristics of the data, so as to achieve data screening and decision-making goals. The node division basis of the decision tree is divided into the following three types: ID3 decision tree, C4.5 decision tree, and CART decision tree. (ID3 is based on information entropy and information gain for classification; C4.5 is based on information gain rate for classification, which is an improved version of ID3; CART is based on the Gini coefficient and can be used for both classification and regression).
[0056] This application takes into account that CART is a binary tree, which has a faster computing speed than C4.5. In addition, CART uses the Gini coefficient as the impurity measure of the variable, which reduces a large number of logarithmic operations. Therefore, the method of constructing a CART decision tree is adopted to conduct deep learning on a large amount of historical data for functional safety hazard analysis and risk assessment analysis.
[0057] The above step S102 also includes S1021, S1022, S1023, S1024 and S1025:
[0058] S1021, Gini index calculation operation, the specific steps of the Gini index calculation operation include:
[0059] Determine the possible values corresponding to each dimension; divide the data set of the corresponding dimension based on the possible values to calculate the Gini index of the data set of each dimension, wherein the Gini index is used to characterize the impurity of the data set.
[0060] Specifically, in the classification problem, assuming there are K categories, the probability that a sample point belongs to the Kth category is P K , then the Gini index of probability distribution is defined as:
[0061]
[0062] Therefore, for a given dataset D, its Gini impurity is:
[0063]
[0064] Among them, C k is the sample subset belonging to the kth class in D, namely C0, C1, C2, C3, four subclasses, |C k | represents the number of samples in controllability category k, and K is the number of controllability categories (C0, C1, C2, C3).
[0065] From the definition of Gini impurity, we can see that if the number of sample categories in the data set D is greater, the corresponding "impurity" will be greater, which means that the data set is "impure", which is also similar to the nature of information entropy. On the contrary, if there is only one category in the data set, the corresponding Gini impurity will be 0.
[0066] This application takes the above factors into consideration. In the process of generating the decision tree, classification is performed based on five features (effect of failure, driving scenario, driving operation, environmental conditions, and people in danger). For each feature A, a possible value a is taken to divide the data set D into two parts, D1 and D2, namely:
[0067] D1={(x,y)∈D|A(x)=α}
[0068] D2=D-D1
[0069] Then under the condition of feature A, the Gini impurity of data set D is defined as:
[0070]
[0071] Where Gini(D,A) represents the uncertainty of the data set D after segmentation A=α.
[0072] The above possible value refers to the value of dividing the data set D into two parts, D1 and D2. For example, among the influences of feature failure: lateral loss of control, longitudinal loss of control, lateral excessive displacement, longitudinal excessive displacement, lateral too small displacement, longitudinal too small displacement, etc., then this possible value can be that excessive displacement is divided into one category and excessive displacement is divided into one category, or it can be divided into one category horizontally and one category vertically. The Gini coefficients of different data sets D are calculated and compared respectively.
[0073] Similar to conditional entropy, the smaller Gini(D,A), the more feature A can reduce the uncertainty of data set D. Therefore, the Gini index is calculated and determined. When constructing a decision tree through the CART algorithm, the present application selects the feature value that minimizes the Gini impurity for sample division.
[0074] S1022, child node generation operation, the specific steps of the child node generation operation include: determining the optimal features and optimal binary cut points corresponding to the smallest Gini index; generating child nodes from the nodes based on the optimal features and optimal binary cut points; and allocating the optimized evaluation data in the data set to the child nodes.
[0075] The CART decision tree uses the Gini index to select the optimal feature and determine the optimal binary cut point of the feature. Two child nodes are generated from the node, and the optimized evaluation data of the data set are allocated to the child nodes.
[0076] S1023, outputting the child nodes that meet the preset conditions as part of the CART decision tree; and re-performing the Gini index calculation operation and the child node generation operation on the child nodes that do not meet the preset conditions until the preset conditions are met.
[0077] The above process of recursively constructing a binary decision tree generates a decision tree based on the training data set. The generated decision tree should be as large as possible, and nodes are established from the root from top to bottom. At each node, a feature value should be selected to split.
[0078] S1024. Perform a pruning operation starting from the bottom of the CART decision tree based on a pruning algorithm, wherein the pruning operation is used to reduce overfitting of the CART decision tree.
[0079] This application considers that after complex decision trees are generated, they are prone to overfitting, so pruning is required. The generated tree is pruned with a validation data set and the optimal subtree is selected. At this time, the minimum loss function is used as the pruning standard.
[0080] Specifically, we first start to prune the generated CART decision tree T0 from its bottom until T n The root node forms a subsequence {T0,T1,...T n}, where Ti+1 is generated from Ti and Tn is the root node.
[0081] S1025. Determine the internal nodes of each pruned node of the CART decision tree; determine a first loss function with the internal node as a single-node tree and a second loss function with the internal node as a subtree with the internal node as a root node; perform a pruning operation based on the first loss function and the second loss function to obtain a sub-number sequence; determine the optimal subtree in the sub-number sequence based on a cross-validation method, wherein the optimal subtree is the CART decision tree.
[0082] Specifically, the present application uses a recursive method to prune the CART decision tree. Increase α from small to large, 0 = α0 < α1 < ... < αn < ∞, and generate a series of intervals [α i ,α i+1 ),i=0,1,..,n; the subtree sequence obtained by pruning corresponds to the interval α∈[α i ,α i+1 ),i=0,1,...,n,the optimal subsequence {T0,T1,...,T n}, the subtrees in the sequence are nested.
[0083] Furthermore, pruning starts from the overall tree T0.
[0084] For any internal node t of T0, the first loss function of t as a single node tree is:
[0085] C α (t) = C (t) + α
[0086] Subtree T with t as root node t The second loss function is:
[0087] C α (T) = C(T) + α|T|
[0088] When α=0 or α is very small, C α (T t ) <C α (t), when α increases to a certain extent, there is C α (T t )=C α (t), when α continues to increase, the inequality is reversed, that is, if the following equation is satisfied:
[0089]
[0090] T t It has the same loss function as t, but t has fewer nodes, so t is better than T t It is better to use T t To do this, for each internal node t in T0, the calculation is:
[0091]
[0092] g(t) indicates the degree of reduction of the overall loss function after pruning. Prune T1 with the smallest g(t) in T0, and set the smallest g(t) to α1. T1 is the optimal subtree in the interval [α1, α2). Prune in this way until the root node is reached. In this process, the value of α is continuously increased to generate new intervals.
[0093] In the subtree sequence T0,T1,...T obtained by pruning n In the above example, the subtree sequence is tested on an independent validation data set through cross-validation, and the optimal subtree T is selected from them. α Finally, the optimal subtree can be used as a decision tree model to perform controllability determination for functional safety hazard analysis and risk assessment.
[0094] S103. Evaluate functional safety controllability based on the CART decision tree.
[0095] Through the above technical solution, the functional safety controllability analysis method provided by the present invention solves the problem that the functional safety controllability analysis is too dependent on manual experience. By introducing machine learning technology, a method for deep learning of a large amount of historical data is used to improve the accuracy of controllability judgment and reduce the errors caused by subjective factors of experts. By using the data of five dimensions, namely the impact of input failure, driving scene, driving operation, environmental conditions, and people in danger, the trained model is used to obtain the results, which can improve efficiency and complete the processing of large amounts of data faster and more accurately. By integrating the optimized machine learning model into the safety analysis tool, the automatic controllability judgment is realized, the tedious manual operation is reduced, and the work efficiency is improved. Compared with the existing controllability judgment method, the present invention not only improves the accuracy and efficiency of functional safety analysis, but also realizes the optimization of the model and automated decision-making through machine learning technology, which has significant advantages.
[0096] Furthermore, as a response to the above Figure 1 In order to realize the method shown in the figure, the embodiment of the present invention also provides a functional safety controllability analysis device for analyzing the above Figure 1 The device embodiment corresponds to the aforementioned method embodiment. For ease of reading, this device embodiment will not repeat the details of the aforementioned method embodiment one by one, but it should be clear that the device in this embodiment can correspond to all the contents of the aforementioned method embodiment. Figure 2 As shown, the device comprises: a determination unit 21, a construction unit 22 and an evaluation unit 23, wherein
[0097] A determination unit 21, configured to determine optimization evaluation data, wherein the optimization evaluation data is used to establish a data set;
[0098] A construction unit 22 is used to construct a CART decision tree based on the data set;
[0099] The evaluation unit 23 is used to evaluate the functional safety controllability based on the CART decision tree.
[0100] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and a functional safety controllability analysis method is implemented by adjusting kernel parameters, which can solve the problem that functional safety controllability analysis is too dependent on manual experience.
[0101] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, and when the program is executed by a processor, the functional safety controllability analysis method is implemented.
[0102] An embodiment of the present invention provides a processor, which is used to run a program, wherein the functional safety controllability analysis method is executed when the program is running.
[0103] An embodiment of the present invention provides an electronic device, the electronic device comprising at least one processor and at least one memory connected to the processor; wherein the processor is used to call program instructions in the memory to execute the functional safety controllability analysis method as described above
[0104] An embodiment of the present invention provides an electronic device 30, such as Figure 3 As shown, the electronic device includes at least one processor 301, and at least one memory 302 and a bus 303 connected to the processor; wherein the processor 301 and the memory 302 communicate with each other through the bus 303; the processor 301 is used to call the program instructions in the memory to execute the above-mentioned functional safety controllability analysis method.
[0105] The intelligent electronic devices in this article can be PCs, PADs, mobile phones, etc.
[0106] The present application also provides a computer program product, which, when executed on a process management electronic device, is suitable for executing a program that initializes the above-mentioned functional safety controllability analysis method steps.
[0107] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0108] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0109] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0110] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0111] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0112] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device is caused to execute the following Figure 1 This corresponds to the flow of control of the memory in the embodiment.
[0113] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integration. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)), etc.
[0114] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0115] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.
[0116] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0117] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0118] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.
[0119] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A functional safety controllability analysis method, characterized in that: include: Determining optimized evaluation data, wherein the optimized evaluation data is used to establish a data set; Construct a CART decision tree based on the data set; Functional safety controllability is evaluated based on the CART decision tree.
2. The method according to claim 1, characterized in that Determine optimization assessment data, including: Obtaining historical assessment data, wherein the assessment data is used to assess the functional safety controllability of historical cases, and the historical assessment data is determined based on five dimensions: failure impact, driving scenario, driving operation, people in danger, and environmental conditions; extracting characteristic values of the historical evaluation data; A noise removal operation and an outlier elimination operation are performed on the characteristic values to determine optimized evaluation data, wherein the optimized evaluation data is used to construct the data set.
3. The method according to claim 1, characterized in that The method further comprises: a Gini index calculation operation, The specific steps of the Gini index calculation operation include: Determine the possible values for each dimension; The data set of the corresponding dimension is divided based on the possible values to calculate the Gini index of the data set of each dimension, wherein the Gini index is used to characterize the impurity of the data set.
4. The method according to claim 3, characterized in that The method further comprises: a child node generation operation, The specific steps of the child node generation operation include: Determine the optimal features and optimal binary cut-off points corresponding to the smallest Gini index; Generating child nodes from the node based on the optimal feature and the optimal binary split point; The optimized evaluation data in the data set is distributed to the child nodes.
5. The method according to claim 4, characterized in that The step of constructing a CART decision tree based on the data set comprises: Output the sub-nodes that meet the preset conditions as part of the CART decision tree; The sub-nodes that do not meet the preset conditions are subjected to the Gini index calculation operation and the sub-node generation operation again until the preset conditions are met.
6. The method according to claim 1, characterized in that Also includes: Based on the pruning algorithm, a pruning operation is performed starting from the bottom of the CART decision tree, wherein the pruning operation is used to reduce the overfitting phenomenon of the CART decision tree.
7. The method according to claim 6, characterized in that Also includes: Determine the internal nodes of each pruning node of the CART decision tree; Determine a first loss function of a tree with the internal node as a single node and a second loss function of a subtree with the internal node as a root node; Performing a pruning operation based on the first loss function and the second loss function to obtain a sub-number sequence; The optimal subtree in the sub-number sequence is determined based on a cross-validation method, wherein the optimal subtree is the CART decision tree.
8. A functional safety controllability analysis device, characterized in that: Also includes: A determination unit, used to determine optimized evaluation data, wherein the optimized evaluation data is used to establish a data set; A construction unit, used for constructing a CART decision tree based on the data set; An evaluation unit is used to evaluate the functional safety controllability based on the CART decision tree.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed by a processor, the steps of the functional safety controllability analysis method according to any one of claims 1 to 7 are implemented.
10. An electronic device, characterized in that: The electronic device includes at least one processor and at least one memory connected to the processor; wherein the processor is used to call program instructions in the memory to execute the steps of the functional safety controllability analysis method as described in any one of claims 1 to claim 7.
Citation Information
Patent Citations
Vehicle feature evaluation method based on hierarchical clustering and decision tree
CN110570655A
Driving risk assessment method, apparatus and device, and storage medium
CN110968839A
Automobile driving controllability assessment method and vehicle function safety risk assessment method
CN115923816A
Vehicle function safety controllability grade evaluation method, device, equipment and medium
CN117422344A
