KR20120086844A - Hybrid Model Simulation Method for Cyber-Physical System Environments - Google Patents
Hybrid Model Simulation Method for Cyber-Physical System Environments Download PDFInfo
- Publication number
- KR20120086844A KR20120086844A KR1020110008133A KR20110008133A KR20120086844A KR 20120086844 A KR20120086844 A KR 20120086844A KR 1020110008133 A KR1020110008133 A KR 1020110008133A KR 20110008133 A KR20110008133 A KR 20110008133A KR 20120086844 A KR20120086844 A KR 20120086844A
- Authority
- KR
- South Korea
- Prior art keywords
- model
- fuzzy
- hybrid
- atomic
- cyber
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N10/00—Quantum computing, i.e. information processing based on quantum-mechanical phenomena
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/048—Fuzzy inferencing
Abstract
The present invention relates to a method of modeling and simulating a hybrid system for a cyber-physical system environment, wherein each of the cyber-physical systems has characteristics of computer-environment interaction, real-world data, high reliability, uncertainty, hybrid, etc. By incorporating inference modules using the inference method of artificial intelligence in the detailed model of, we can model complex real-world systems including human intelligence. In addition to an inference module that can predict its next state, it provides a simulation framework based on discrete event system specification formalism to simulate these models, enabling model-level simulation in a more real-world-like environment, Can raise the level.
Description
The present invention relates to a method for modeling and simulating a hybrid system for a cyber-physical system environment, and more specifically, to modeling some real-world systems with human intelligence among various detailed models of the cyber-physical system using inference modules, A simulation framework based on discrete event system specification formalism capable of simulating the model.
Cyber-physical system refers to a system that guarantees software reliability, real-time and intelligence in order to prevent unexpected occurrences and errors due to the complexity of real-world systems and computing systems. Based on the network, the cyber-physical system can be configured in a strong combination of physical systems or physical processes and the computing systems that control them. Cyber-physical systems are characterized by computer interaction, real-world large data, high reliability, uncertainty, and hybrids. The state of the system consists of variable values that affect the output of the system. The variable values are classified into discrete event systems or continuous systems according to whether the variable values are discrete or continuous, but a system in which the two are mixed is a hybrid system.
In order to ensure the stability and reliability of the system, a technique for precisely modeling the system is required from the design stage, and performance evaluation and verification through simulation may be essential. Various frameworks have been studied to model and simulate precise hybrid systems, and the simulation framework based on discrete event system specification formalism provides an environment for simulating the hybrid system model.
However, the discrete event system specification-based simulation framework does not support a method for simulating a continuous state change of the hybrid model if it is not given as a differential equation. In addition, in the case of the cyber-physical system, there is a problem that a large number of unpredictable variables exist so that differential equations may not be defined or may be represented by a plurality of differential equations near infinity.
Therefore, there is a need for a method for modeling and simulating a hybrid system by solving the above problems.
In the case of a model in which a continuous state change of a cyber-physical system is not represented by a differential equation, the present invention includes a fuzzy rule-based reasoning module in the model, which is based on a more real-world-like cyber-physical system model and discrete event system specification formalism. It aims to provide a simulation framework.
The characteristics of the continuous state change of a typical system can be represented by a differential equation, and the differential equation shows the relationship between the time and the rate of change of the state variable. However, under the cyber-physical system environment, an unknown variable or an unpredictable variable may have an unexpected effect on a specific state of the system. Therefore, there is a limit in expressing one differential equation. Therefore, in the present invention, the system model, which is difficult to be expressed by the one differential equation, is defined and classified as an unknown model.
The Unknown model is included in one category of continuous systems of the cyber-physical system. In the case of the differential equation model, which is a classification of another continuous system, it is simulated through a discretization method such as time slicing or state quantization, whereas for the unknown model, the dynamic characteristics of the system are described by fuzzy rules. In the simulation step, the next state of the model is calculated by inference based on the fuzzy rule.
In the case of the differential equation model, which is a general continuous system model that can be expressed as a differential equation, the next state of the system can be tracked by an equation, while for the unknown model, it is impossible to completely predict how the next state will change at a given point in time. Because it can not be characterized by performing the simulation in a manner to predict the state of the model by the above-mentioned inference.
An atomic-fuzzy model is defined in which the fuzzy inference module is inserted into a discrete event system specification formalism based atomic model to represent the unknown model. The atomic-fuzzy model may be input to an external state transition function of the model when an input is generated from the outside and update the state of the model therefrom. When a predetermined amount of time passes in the current state of the model, the output of the model is generated and at the same time, the internal state transition function is called to update the ecology of the model. The inference module can infer the time of occurrence of the next event from the input of the model and the current state, and the inference uses a predefined fuzzy rule, which is defined differently according to the model and used to represent the intelligence of the system. It features.
According to the simulation method for the cyber-physical system environment according to the present invention as described above, the system at the model level by representing the model using a fuzzy rule for a continuous model that is not represented by a differential equation for the simulation of the cyber-physical system Extend expressiveness, which allows for more complex and precise representation of the system.
In addition, it is possible to predict the next state of the model through fuzzy rule-based inference during the simulation process, by providing a framework that can simulate a simple continuous model represented by a differential equation and a model described by fuzzy rules together. This has the advantage of increasing the model level of simulation.
1 illustrates the classification of a hybrid cyber-physical system model of the present invention.
2 is a view showing the difference in the dynamic characteristics of the ODE model and Unknown model of the present invention.
3 illustrates the concept of an atomic-fuzzy model of the present invention.
4 is a diagram showing the components and the input-output relationship of the atomic-fuzzy model of the present invention.
5 is a diagram illustrating a membership function for fuzzy rule-based system input and output variables of the UACV model.
Hereinafter, a hybrid simulation method for a cyber-physical system environment according to the present invention will be described in detail with reference to the accompanying drawings.
1 is a diagram illustrating a classification of a hybrid cyber-physical system model of the present invention.
As shown in FIG. 1, the
Therefore, in the present invention, a system capable of defining the
In the case of the
In case of the unknown model (16), it is a model of continuous system defined in consideration of the uncertainty of the cyber-physical system environment, and means a system in which the characteristics of the system are not known and an unpredictable variable exists. The dynamic characteristics of the system can be described and the simulation can be performed through inference-based simulation.
2 is a view showing the difference in the dynamic characteristics of the ODE model and Unknown model of the present invention.
In the case of the ODE
The present invention provides an atomic-fuzzy model in which a fuzzy inference module is inserted into an atomic model based on a discrete event system specification formalism in order to represent an unknown model in order to enable such inference simulation.
3 is a diagram illustrating the concept of an atomic-fuzzy model of the present invention.
The atomic-
As shown in FIG. 3, in the atomic-
4 is a diagram showing the components and the input-output relationship of the atomic-fuzzy model of the present invention.
As shown in FIG. 4, the left side of the internal components of the atomic-fuzzy model 40 corresponds to the components of the existing atomic model based on the discrete event system specification formalism. When an input from outside of the atom-fuge model 40 occurs, the input is passed to an external
The result value through the fuzzy inference of the
Through this atomic-fuzzy model 40, it is difficult to completely predict the next state corresponding to the unknown model among the continuous systems of the cyber-physical system, or may be represented as a model for model simulation of an intelligent system.
As an example for representing the
FIG. 5 illustrates a membership function for fuzzy rule-based system input / output variables of a UAV model. Among these, FIGS. 5A, 5B, and 5C correspond to membership functions for fuzzy input variables, and FIG. 5D shows membership functions for fuzzy output variables.
These membership functions represent values received from the internal and external environment of the UAV model. Details of each fuzzy input variable and output variable can be expressed as shown in the following table.
Some of the rules of the fuzzy rule system that can be obtained by using the membership function according to the above table and FIG. 5 are as follows.
R01: If the weather is B, the durability is W, and the altitude is G, then the time of occurrence of the next state event is L.
R03: If the weather is B, the durability is W, and the altitude is H, the time of occurrence of the next state event is VS.
R08: If the weather is B, the durability is S and the altitude is M, then the time of occurrence of the next state event is S.
R23: If the weather is G, the durability is N, and the altitude is M, then the time of occurrence of the next state event is S.
R25: When the weather is G, the durability is S, and the altitude is G, the time of the next state event occurs is VS.
R27: If the weather is G, the durability is S, and the altitude is H, then the time of occurrence of the next state event is L.
According to the above rules, if the weather is bad or if the durability is low, the next event will be expected sooner or later if the current altitude is not above ground, and if the weather is good and the durability is good, it will take off. In the case of the altitude above, it can be seen that the UAV system having the characteristic of predicting that much time is left until the next event occurrence time is represented through the fuzzy rule.
Although the present invention has been described in detail with reference to exemplary embodiments above, those skilled in the art to which the present invention pertains can make various modifications without departing from the scope of the present invention. Will understand. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the claims below and equivalents thereof.
<Description of Signs for Main Parts of Drawings>
11: Hybrid Cyber-Physical System 12: Continuous Cyber-Physical System
13: ODE model 14: time based discretization
15: State Based Discretization 16: Unknown Model
17 Inference Based
40: atom-fuzzy model 41: external state transition function
42: State 43: Time Forward Function
44: fuzzy system 45: output function
46: internal state transition function
Claims (10)
Pass the input and current state of the model to the fuzzy inference module to schedule the next event occurrence according to the value returned by the fuzzy inference module,
When the next event occurs, the output function of the model is generated.
Hybrid system model, wherein the state of the model is updated by an internal state transition function
Hybrid system modeling method characterized by defining and classifying the system model, which is difficult to be represented by one differential equation, as Unknown model
A continuous system model comprising an atomic-fuzzy model with a fuzzy inference module embedded in a discrete event system specification formalism based atomic model.
While scheduling the next event occurrence by changing the state in quantum size, the fuzzy system relies on the model to infer the time of the next event occurrence of the model from the input and current state of the atomic model, using the fuzzy rules. Continuous system model, characterized in that to output the corresponding atomic model
Discrete event system specification formalism-based hybrid model simulation framework that can simulate hybrid system models, including components of the atomic model and fuzzy inference modules
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020110008133A KR20120086844A (en) | 2011-01-27 | 2011-01-27 | Hybrid Model Simulation Method for Cyber-Physical System Environments |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020110008133A KR20120086844A (en) | 2011-01-27 | 2011-01-27 | Hybrid Model Simulation Method for Cyber-Physical System Environments |
Publications (1)
Publication Number | Publication Date |
---|---|
KR20120086844A true KR20120086844A (en) | 2012-08-06 |
Family
ID=46872527
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
KR1020110008133A KR20120086844A (en) | 2011-01-27 | 2011-01-27 | Hybrid Model Simulation Method for Cyber-Physical System Environments |
Country Status (1)
Country | Link |
---|---|
KR (1) | KR20120086844A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101275172B1 (en) * | 2012-12-17 | 2013-06-18 | 국방과학연구소 | Method of simulating discrete event |
KR101299192B1 (en) * | 2012-09-14 | 2013-08-21 | 아주대학교산학협력단 | Apparatus for simulating based on modeling and simulation framework in distributed cognitive radio networks |
CN113672207A (en) * | 2021-09-02 | 2021-11-19 | 北京航空航天大学 | X language hybrid model modeling system, method and storage medium |
-
2011
- 2011-01-27 KR KR1020110008133A patent/KR20120086844A/en not_active Application Discontinuation
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101299192B1 (en) * | 2012-09-14 | 2013-08-21 | 아주대학교산학협력단 | Apparatus for simulating based on modeling and simulation framework in distributed cognitive radio networks |
KR101275172B1 (en) * | 2012-12-17 | 2013-06-18 | 국방과학연구소 | Method of simulating discrete event |
CN113672207A (en) * | 2021-09-02 | 2021-11-19 | 北京航空航天大学 | X language hybrid model modeling system, method and storage medium |
CN113672207B (en) * | 2021-09-02 | 2024-03-26 | 北京航空航天大学 | X language hybrid model modeling system, method and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Katz et al. | Towards proving the adversarial robustness of deep neural networks | |
Hussain et al. | Sensor failure detection, identification, and accommodation using fully connected cascade neural network | |
Shi et al. | An adaptive decision-making method with fuzzy Bayesian reinforcement learning for robot soccer | |
DE112017002604T5 (en) | Systems and methods for machine learning using a trusted model | |
Xiaohong et al. | Robustness evaluation method for unmanned aerial vehicle swarms based on complex network theory | |
Gharehchopogh | Neural networks application in software cost estimation: A case study | |
US8903750B1 (en) | Adaptive learning network for fusion of remaining useful life with machinery condition | |
Tyagi et al. | A rule-based approach for estimating the reliability of component-based systems | |
US10921755B2 (en) | Method and system for competence monitoring and contiguous learning for control | |
US20200202723A1 (en) | Machine learning in avionics | |
US20230004846A1 (en) | Behavioral prediction and boundary settings, control and safety assurance of ml & ai systems | |
Levitin et al. | Optimal inspections and mission abort policies for multistate systems | |
Gansch et al. | System theoretic view on uncertainties | |
WO2019160138A1 (en) | Causality estimation device, causality estimation method, and program | |
JP6902487B2 (en) | Machine learning system | |
US20200349416A1 (en) | Determining computer-executed ensemble model | |
Xu et al. | AdaFT: A framework for adaptive fault tolerance for cyber-physical systems | |
KR20120086844A (en) | Hybrid Model Simulation Method for Cyber-Physical System Environments | |
Vassev | Safe Artificial Intelligence and Formal Methods: (Position Paper) | |
CN111144572A (en) | Power distribution network disaster situation inference method and system based on tree-shaped Bayesian network | |
CN116245009A (en) | Man-machine strategy generation method | |
Schirmer et al. | A hierarchy of monitoring properties for autonomous systems | |
Lesire et al. | Estimation and conflict detection in human controlled systems | |
Hwang et al. | Induced states in a decision tree constructed by Q-learning | |
Gros | Tracking the race: Analyzing racetrack agents trained with imitation learning and deep reinforcement learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A201 | Request for examination | ||
E902 | Notification of reason for refusal | ||
E601 | Decision to refuse application |