CN112434440A - Intelligent parking method and system based on multi-Agent self-adaptive online verification - Google Patents
Intelligent parking method and system based on multi-Agent self-adaptive online verification Download PDFInfo
- Publication number
- CN112434440A CN112434440A CN202011403489.9A CN202011403489A CN112434440A CN 112434440 A CN112434440 A CN 112434440A CN 202011403489 A CN202011403489 A CN 202011403489A CN 112434440 A CN112434440 A CN 112434440A
- Authority
- CN
- China
- Prior art keywords
- intelligent parking
- parking system
- probability
- state transition
- vehicle
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Artificial Intelligence (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent parking method based on multi-Agent self-adaptive online verification, which comprises the following steps of: extracting key states of the intelligent parking system and events causing state transition; expressing the behavior of the intelligent parking system by using key states and events, and constructing a state transition diagram without transition probability of the intelligent parking system; converting the main environmental influence factors into probability values by a Bayesian network method, and taking the probability values as transition probabilities among states; adding the transition probability among the states into the state transition diagram to obtain a complete state transition diagram; extracting the requirements and targets of the intelligent parking system according to the service logic of the intelligent parking system; the method comprises the following steps of performing formal description on requirements and targets of the intelligent parking system by using a probability calculation tree logic formula; constructing a discrete time Markov model, and loading a probability calculation tree logic formula; and analyzing by using an inspection tool, and using the obtained verification result for the decision of the intelligent parking system in an uncertain environment.
Description
Technical Field
The invention relates to a multi-Agent self-adaptive online verification method, in particular to an intelligent parking method and system based on multi-Agent self-adaptive online verification.
Background
Adaptive systems can adjust their behavior and structure through closed-loop control to respond to these changes, thereby achieving compliance in changing demands and system failures. On-line Verification (Runtime Verification) is a mathematic-based method and can help an adaptive system to realize closed-loop control. The online verification can perform formal analysis on the system, help the adaptive system make adaptive decisions in a control loop and update a system model. On-line validation can identify conditions that violate requirements and, under certain conditions, can make predictions, and can then drive the system to perform adaptations in a manner that restores compliance with those requirements. The multi-Agent system consists of a plurality of agents, the agents cooperate with each other and make decisions to achieve a unified goal, and the agents have a common knowledge base. The multi-Agent system is divided into an isomorphic Agent and an isomerous Agent, wherein the isomerous Agent is composed of a plurality of agents with the same capability; the latter consists of multiple agents with similar or different capabilities. Different agents have different structures and limited information, and in the running process of the system, the agents need to complete tasks according to self capacity or cooperate with other agents to complete the tasks. Therefore, when the system is in an open and changing dynamic environment, the Agent body or cooperation can generate a fault condition in the process of completing the task. Therefore, the problem that how to reduce the occurrence of faults or recover the tire installation in time after the faults occur so as to ensure the reliability of the system becomes urgent to solve. The traditional self-adaptive online verification method comprises the following steps: early, the method is realized through formalization methods such as system test, theorem proving or equivalence checking, but the method is generally applied to offline stages such as system design or maintenance and cannot be used in the dynamic operation process of the system so as to generate self-adaptive behavior to the environment. In the recently-appearing methods such as dynamic reconfiguration based on a telemedicine service system and dynamic resource management of a cloud computing infrastructure, the method cannot be applied to a heterogeneous system because the method is an adaptive method used for a single system or a homogeneous Agent system. The self-adaptive online verification method is applied to the heterogeneous Agent system, verification can be performed in the system operation process, the result is used for supporting system reconstruction, and the method is suitable for the heterogeneous Agent system.
The intelligent parking system consists of a large parking lot and parking robots, wherein the parking robots can be classified into a plurality of types according to functions, and the parking robots complete a parking operation process through cooperation. The system is a heterogeneous Agent system in a dynamic environment, so that the system is influenced by environmental factors in the operation process, so that a fault condition occurs in the parking process, and a system target requires that the system needs to process and recover the fault condition in time.
Disclosure of Invention
The invention aims to provide a multi-Agent-based self-adaptive online verification method, and the intelligent parking system is constructed into a probability model and is subjected to online verification, so that the system can continuously keep stability in the operation process.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the intelligent parking method based on multi-Agent self-adaptive online verification comprises the following steps:
s1, extracting key states of the intelligent parking system and events causing state transition;
s2, representing the behavior of the intelligent parking system by using the key states and events, and constructing a state transition diagram without transition probability of the intelligent parking system;
s3, selecting main factors which obviously influence the intelligent parking system;
s4, converting the main factors into probability values through a Bayesian network method, and taking the probability values as transition probabilities among states;
s5, adding the transition probability among the states into the state transition diagram to obtain a complete state transition diagram;
s6, extracting the requirements and the targets of the intelligent parking system according to the service logic of the intelligent parking system;
s7, performing formal description on the requirements and the targets of the intelligent parking system by using a probability calculation tree logic formula;
s8, constructing a discrete time Markov model, and loading a probability calculation tree logic formula;
and S9, analyzing by using an inspection tool, and using the obtained verification result for the decision of the intelligent parking system in the uncertain environment.
According to the technical scheme, the key states comprise that the lifting robot is idle, scheduled, arrives at a vehicle floor, loads the vehicle and arrives at a parking space floor to unload the vehicle, and the parking robot is idle, scheduled, arrives at the lifting robot, loads the vehicle and arrives at the parking space to unload the vehicle.
In the above technical solution, step S4 specifically includes:
taking the state of faults occurring in the running process of the lifting robot and the parking robot as a fault set W ═ W1,W2,…,WnAnd classifying fault symptom data as a symptom set E ═ E1,E2,…,EmAnd (5) taking the fault set W as a first layer and the symptom set E as a second layer to construct a double-layer Bayesian network, and connecting nodes of the two layers in a matching manner according to the fault form to obtain the Bayesian network, wherein n and m represent the number of faults and symptoms.
The invention also provides an intelligent parking system based on multi-Agent self-adaptive online verification, which comprises the following components:
the state and event extraction module is used for extracting key states of the intelligent parking system and events causing state conversion;
the state transition diagram building module is used for representing the intelligent parking system behaviors by using key states and events and building a state transition diagram without transition probability of the intelligent parking system;
the influence factor selection module is used for selecting main factors which obviously influence the intelligent parking system;
the probability value conversion module is used for converting the main factors into probability values by a Bayesian network method and using the probability values as transition probabilities among the states;
the state transition diagram supplementing module is used for adding the transition probability among the states into the state transition diagram to obtain a complete state transition diagram;
the demand and target extraction module is used for extracting demands and targets of the intelligent parking system according to the service logic of the intelligent parking system;
the demand and target description module is used for formally describing demands and targets of the intelligent parking system by using a probability calculation tree logic formula;
the discrete time Markov model building module is used for building a discrete time Markov model and loading a probability calculation tree logic formula;
and the verification module is used for analyzing by using an inspection tool and using the obtained verification result for the decision of the intelligent parking system in the uncertain environment.
According to the technical scheme, the key states comprise the idle state, the reservation state, the arrival vehicle floor, the loading vehicle state and the arrival parking space floor unloading vehicle state of the lifting robot, and the idle state, the reservation state, the arrival lifting robot position, the loading vehicle state and the arrival parking space floor unloading vehicle state of the parking robot.
In connection with the above technical solution, the probability value conversion module is specifically configured to use a state of a fault occurring in an operation process of the lifting robot and the parking robot as a fault set W ═ W1,W2,…,WnAnd classifying fault symptom data as a symptom set E ═ E1,E2,…,EmAnd (5) taking the fault set W as a first layer and the symptom set E as a second layer to construct a double-layer Bayesian network, and connecting nodes of the two layers in a matching manner according to the fault form to obtain the Bayesian network, wherein n and m represent the number of faults and symptoms.
The invention also provides a computer readable storage medium which can be executed by the processor, wherein a computer program is stored in the storage medium, and the computer program executes the intelligent parking method based on the multi-Agent adaptive online verification in the technical scheme.
The invention has the following beneficial effects: the invention relates to a method for modeling behavior abstraction of an intelligent parking system into a Markov model, embodying influence factors in a system environment into transition probability in the Markov model, formalizing a system demand target into a probability calculation tree logic formula, verifying in a probability model checker and using the result in decision making. An online verification technology is fused in the multi-heterogeneous-Agent self-adaptive system, so that the fault probability of the system behavior can be uninterruptedly obtained in the running process of the system, and further, a behavior strategy under a change environment can be generated online to drive the system to evolve in a self-adaptive manner.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a schematic overall flow chart of an intelligent parking method based on multi-Agent adaptive online verification according to an embodiment of the present invention;
FIG. 2(a) is a state transition diagram of Agent-A without transition probability in an embodiment of the present invention;
FIG. 2(B) is a state transition diagram of Agent-B without transition probability in the embodiment of the present invention;
FIG. 3 is a Bayesian network in an embodiment of the present invention;
FIG. 4(a) is a complete state transition diagram of Agent-A in an embodiment of the present invention;
FIG. 4(b) is the complete state transition diagram of Agent-A in the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In order to further understand and understand the intelligent parking implementation method based on multi-Agent adaptive online verification, the following detailed description is provided by using embodiments and accompanying drawings. The implementation case is as follows:
the whole intelligent parking lot is a cross-layer large scene, wherein the parking robots comprise a lifting robot A and a parking robot B, and A, B can be single or multiple. The robot A is responsible for cross-layer transportation of the vehicles, and the robot B is responsible for same-layer transportation of the vehicles and parking and vehicle taking operations. After the vehicle owner parks the vehicle to a special parking lot and confirms the parking lot, the robot A transports the vehicle to the floor where the parking lot is located from the ground in a cross-layer mode, the robot B in an available state is informed before the vehicle arrives at the floor, the robot B hands over the vehicle after arriving at the robot A, and after the hand-over is completed, the empty available parking lot is found through a routing algorithm and the parking process is completed. When the vehicle owner needs to take the vehicle, the robot B informs the robot A in the process of taking back the vehicle and going to the robot A, the robot A arrives at the floor where the robot B is located and completes vehicle transfer with the robot B, and then the vehicle is conveyed to the ground and returned to the vehicle owner.
The robot in the scene is regarded as Agent, so that multiple heterogeneous agents exist in the whole field, cooperation and interaction exist among the heterogeneous agents, and the environment where the heterogeneous agents are located is a dynamic open environment, so that the system is subjected to system faults caused by multiple uncertainties in operation, and the system stability is enhanced by using a multi-Agent-based self-adaptive online verification method.
Since the car taking and parking processes are similar, only the parking process is shown here.
As shown in fig. 1, the intelligent parking method based on multi-Agent adaptive online verification in the embodiment of the present invention specifically includes the following steps:
step 1: the behavior logic of the intelligent parking system is analyzed, wherein the main behaviors are provided by the robot A, B, and the key states of the behaviors and the conditions for state transition are mentioned, and the key states are shown in the A-Agent key state table.
TABLE A-Agent Key status Table
Step 2: with NiAnd FiRespectively representing the normal running state and the fault condition of the system, and constructing a system state transition diagram according to an Agent key state table to obtainTo the embodiment of Agent without transition probability state transition diagram as in FIG. 2, wherein (a) represents the state transition diagram of Agent-A and (B) represents the state transition diagram of Agent-B. The initial state of the Agent is N0If the system runs in a normal state, sends a signal to the Agent and is normally received and processed, the state of the system is N0Jump to N1If the Agent-A can not receive the signal due to any fault condition in the process, the state of the Agent-A is N0Jump to F1And is always F before the failure is resolved1Self-circulation is carried out, and the fault is removed and then the system jumps back to N1. By the same token, can be obtained from NiJump to Ni+1Or Fi+1And FiJump to NiThe process of (2);
and step 3: analyzing influence factors in a system environment, wherein the influence factors comprise mechanical factors, electrical factors, software factors or abnormal interference in a system operation environment and the like, and the analysis method mainly exists in an Agent interaction process and an Agent execution respective task process;
and 4, step 4: the influencing factors in the processes are converted into probability values through a Bayesian network method, and the specific process is described as follows. Taking the state of the fault occurring in the Agent operation process as a fault set W ═ W1,W2,…,WnAnd classifying fault symptom data as a symptom set E ═ E1,E2,…,EmN and m represent the number of faults and symptoms. The fault set W is used as a first layer, the symptom set E is used as a second layer to construct a double-layer Bayesian network, and nodes of the two layers are connected in a matching mode according to a fault mode, so that the Bayesian network in the form of the implementation case shown in FIG. 3 can be obtained. Therefore, the probability of a certain fault can be estimated under the condition that the Agent shows certain symptoms, namely the posterior probability of various faults is calculated through a Bayes formula. These posterior probabilities are taken as transition probabilities between states;
and 5: substituting the calculated probability value into fig. 2 to obtain a complete state transition diagram of the system, such as the complete state transition diagram of the embodiment of fig. 4;
step 6: analyzing the service logic of the system, such as the object and behavior, the operation rule, the work flow, the data flow direction and the like of the system, and extracting the system requirement and the target, wherein the system requirement is the stable operation time of the system;
and 7: formalizing the above requirement, and describing it as a logical formula of a probability computation tree in the form of formula (1), which means the probability that the system loses more than 10 messages within time T, and the result is a probability value, where P represents the probability? Representing that the value obtained by the formula is in the form of the place, F represents a Finally keyword, T represents time and is a non-negative number, and the messages _ lost represents the number of lost messages and can be defined when analyzing the requirement;
P＝？[F<＝T messages_lost>10] (1)
and 8: in a probability model checking tool Prism, constructing a discrete time Markov model of a system according to a state transition diagram, and loading a probability calculation tree logic formula into a model checker;
and step 9: the probability of system failure, the stage of system operation when failure occurs, and the like can be obtained by analyzing with an inspection tool Prism, and the verification results are used as the data support of system self-adaptation, so that further planning is carried out.
The intelligent parking system based on multi-Agent self-adaptive online verification comprises the following components:
the state and event extraction module is used for extracting key states of the intelligent parking system and events causing state conversion;
the state transition diagram building module is used for representing the intelligent parking system behaviors by using key states and events and building a state transition diagram without transition probability of the intelligent parking system;
the influence factor selection module is used for selecting main factors which obviously influence the intelligent parking system;
the probability value conversion module is used for converting the main factors into probability values by a Bayesian network method and using the probability values as transition probabilities among the states;
the state transition diagram supplementing module is used for adding the transition probability among the states into the state transition diagram to obtain a complete state transition diagram;
the demand and target extraction module is used for extracting demands and targets of the intelligent parking system according to the service logic of the intelligent parking system;
the demand and target description module is used for formally describing demands and targets of the intelligent parking system by using a probability calculation tree logic formula;
the discrete time Markov model building module is used for building a discrete time Markov model and loading a probability calculation tree logic formula;
and the verification module is used for analyzing by using an inspection tool Prism, and using the obtained verification result for the decision of the intelligent parking system in the uncertain environment.
Further, the critical states include idle, scheduled, arrival at vehicle floor, loading vehicle, arrival at parking space floor, unloading vehicle of the lifting robot, and idle, scheduled, arrival at lifting robot, loading vehicle, arrival at parking space, unloading vehicle of the parking robot.
Further, the probability value conversion module is specifically configured to use a state of a fault occurring in the operation process of the lifting robot and the parking robot as a fault set W ═ W1,W2,…,WnAnd classifying fault symptom data as a symptom set E ═ E1,E2,…,EmAnd (5) taking the fault set W as a first layer and the symptom set E as a second layer to construct a double-layer Bayesian network, and connecting nodes of the two layers in a matching manner according to the fault form to obtain the Bayesian network, wherein n and m represent the number of faults and symptoms.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
Claims (7)
1. An intelligent parking method based on multi-Agent self-adaptive online verification is characterized by comprising the following steps:
s1, extracting key states of the intelligent parking system and events causing state transition;
s2, representing the behavior of the intelligent parking system by using the key states and events, and constructing a state transition diagram without transition probability of the intelligent parking system;
s3, selecting main factors which obviously influence the intelligent parking system;
s4, converting the main factors into probability values through a Bayesian network method, and taking the probability values as transition probabilities among states;
s5, adding the transition probability among the states into the state transition diagram to obtain a complete state transition diagram;
s6, extracting the requirements and the targets of the intelligent parking system according to the service logic of the intelligent parking system;
s7, performing formal description on the requirements and the targets of the intelligent parking system by using a probability calculation tree logic formula;
s8, constructing a discrete time Markov model, and loading a probability calculation tree logic formula;
and S9, analyzing by using an inspection tool, and using the obtained verification result for the decision of the intelligent parking system in the uncertain environment.
2. The intelligent parking method based on multi-Agent adaptive online verification according to claim 1, wherein the key states comprise idle, scheduled, arrival at a vehicle floor, loading of a vehicle, arrival at a parking space floor and unloading of a vehicle by a lifting robot, and idle, scheduled, arrival at a lifting robot, loading of a vehicle, arrival at a parking space unloading of a vehicle by a parking robot.
3. The intelligent parking method based on multi-Agent adaptive online verification according to claim 1, wherein step S4 specifically comprises:
taking the state of faults occurring in the running process of the lifting robot and the parking robot as a fault set W ═ W1,W2,…,Wn}; taking fault symptom data classification as symptom set E ═ { E ═ E1,E2,…,EmN and m represent the number of faults and symptoms, and the faults are collectedAnd W is used as a first layer, the symptom set E is used as a second layer to construct a double-layer Bayesian network, and nodes of the two layers are connected in a matching mode according to a fault mode to obtain the Bayesian network.
4. An intelligent parking system based on multi-Agent self-adaptive online verification is characterized by comprising:
the state and event extraction module is used for extracting key states of the intelligent parking system and events causing state conversion;
the state transition diagram building module is used for representing the intelligent parking system behaviors by using key states and events and building a state transition diagram without transition probability of the intelligent parking system;
the influence factor selection module is used for selecting main factors which obviously influence the intelligent parking system;
the probability value conversion module is used for converting the main factors into probability values by a Bayesian network method and using the probability values as transition probabilities among the states;
the state transition diagram supplementing module is used for adding the transition probability among the states into the state transition diagram to obtain a complete state transition diagram;
the demand and target extraction module is used for extracting demands and targets of the intelligent parking system according to the service logic of the intelligent parking system;
the demand and target description module is used for formally describing demands and targets of the intelligent parking system by using a probability calculation tree logic formula;
the discrete time Markov model building module is used for building a discrete time Markov model and loading a probability calculation tree logic formula;
and the verification module is used for analyzing by using an inspection tool and using the obtained verification result for the decision of the intelligent parking system in the uncertain environment.
5. The intelligent parking system based on multi-Agent adaptive online verification according to claim 4, wherein the key states comprise idle, scheduled, arrival at vehicle floor, loading vehicle, arrival at parking space floor, unloading vehicle of lifting robot, and idle, scheduled, arrival at lifting robot, loading vehicle, arrival at parking space, unloading vehicle of parking robot.
6. The intelligent parking method based on multi-Agent adaptive online verification according to claim 4, wherein the probability value conversion module is specifically configured to use a state of a fault occurring in the operation process of the lifting robot and the parking robot as a fault set W ═ W [ (-W) of the fault set1,W2,…,WnAnd classifying fault symptom data as a symptom set E ═ E1,E2,…,EmAnd (5) taking the fault set W as a first layer and the symptom set E as a second layer to construct a double-layer Bayesian network, and connecting nodes of the two layers in a matching manner according to the fault form to obtain the Bayesian network, wherein n and m represent the number of faults and symptoms.
7. A computer-readable storage medium, which is executable by a processor, and in which a computer program is stored, the computer program executing the intelligent parking method based on multi-Agent adaptive online verification according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011403489.9A CN112434440A (en) | 2020-12-02 | 2020-12-02 | Intelligent parking method and system based on multi-Agent self-adaptive online verification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011403489.9A CN112434440A (en) | 2020-12-02 | 2020-12-02 | Intelligent parking method and system based on multi-Agent self-adaptive online verification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112434440A true CN112434440A (en) | 2021-03-02 |
Family
ID=74692286
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011403489.9A Pending CN112434440A (en) | 2020-12-02 | 2020-12-02 | Intelligent parking method and system based on multi-Agent self-adaptive online verification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112434440A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130024417A1 (en) * | 2011-07-18 | 2013-01-24 | Andreas Joanni | Method, system and computer program product for automatic generation of bayesian networks from system reliability models |
| CN106527373A (en) * | 2016-12-05 | 2017-03-22 | 中国科学院自动化研究所 | Workshop automatic scheduling system and method based on mutli-intelligent agent |
| CN107645412A (en) * | 2017-09-11 | 2018-01-30 | 南京航空航天大学 | A kind of Web service combination multiple target verification method under open environment |
| CN111098852A (en) * | 2019-12-02 | 2020-05-05 | 北京交通大学 | Parking path planning method based on reinforcement learning |
-
2020
- 2020-12-02 CN CN202011403489.9A patent/CN112434440A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130024417A1 (en) * | 2011-07-18 | 2013-01-24 | Andreas Joanni | Method, system and computer program product for automatic generation of bayesian networks from system reliability models |
| CN106527373A (en) * | 2016-12-05 | 2017-03-22 | 中国科学院自动化研究所 | Workshop automatic scheduling system and method based on mutli-intelligent agent |
| CN107645412A (en) * | 2017-09-11 | 2018-01-30 | 南京航空航天大学 | A kind of Web service combination multiple target verification method under open environment |
| CN111098852A (en) * | 2019-12-02 | 2020-05-05 | 北京交通大学 | Parking path planning method based on reinforcement learning |
Non-Patent Citations (2)
| Title |
|---|
| 叶幸瑜等: "基于马尔可夫模型的多agent自适应在线验证", 《计算机应用研究》 * |
| 田琼等: "基于马尔科夫链的停车寻位模型与仿真", 《交通运输系统工程与信息》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Nafz et al. | Constraining self-organisation through corridors of correct behaviour: The restore invariant approach | |
| EP3588290A1 (en) | Resources management in internet of robotic things (iort) environments | |
| Komma et al. | An approach for agent modeling in manufacturing on JADE™ reactive architecture | |
| US9553929B2 (en) | Episodic coordination model for distributed applications | |
| CN102916830B (en) | Implement system for resource service optimization allocation fault-tolerant management | |
| CN116860463A (en) | Distributed self-adaptive spaceborne middleware system | |
| Pankratova | Creation of Physical Models for Cyber-Physical Systems | |
| CN115934344A (en) | Heterogeneous distributed reinforcement learning calculation method, system and storage medium | |
| CN112084004A (en) | Container detection and maintenance method and system for container application | |
| CN102930752A (en) | Training platform for virtual prototype disassembling sequence model based on finite-state machine | |
| Barabash et al. | The assessment of the quality of functional stability of the automated control system with hierarchic structure | |
| CN112434440A (en) | Intelligent parking method and system based on multi-Agent self-adaptive online verification | |
| CN112686605A (en) | Logistics vehicle fault emergency matching processing system | |
| Klöpper et al. | Planning with utility and state trajectory constraints in self-healing automotive systems | |
| CN115185812A (en) | Formal verification method for design layer of embedded operating system | |
| CN109189031B (en) | Distributed control system with layered framework, method and application | |
| Preisler et al. | Scalability and robustness analysis of a multi-agent based self-healing resource-flow system | |
| Ye et al. | Runtime Verification of Multi-Agent Self-Adaptive System | |
| Li et al. | Distributed sensor analysis for fault detection in tightly-coupled multi-robot team tasks | |
| Vallée et al. | Detection of anomalies in a transport system using automation agents with a reflective world model | |
| Berruet et al. | A component based approach for the design of FMS control and supervision | |
| Wan et al. | A self-adaptation framework for dealing with the complexities of software changes | |
| Levitin et al. | Optimal distribution of nonperiodic full and incremental backups | |
| Renz et al. | Mesoscopic stochastic models for validating self-organizing multi-agent systems | |
| CN117539642B (en) | Credit card distributed scheduling platform and scheduling method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210302 |
|
| WD01 | Invention patent application deemed withdrawn after publication |