基于Agent技术的复杂适应系统分析与建模

该文简单介绍了霍兰教授提出的复杂适应系统（CAS）理论,概述了agent和多agent系统（MAS）的概念及其特点,重点介绍了基于agent技术的复杂适应系统的分析方法及建模方法,对其特点及相对于传统方法的优势作了比较.对于由多个Agent组成的复杂适应系统,基于Agent技术是一种十分有效的分析工具,它不但能够对复杂适应系统的体系结构进行适当的描述,而且为复杂适应系统演化过程的建模提供了有力的工具.基于Agent技术的分析与建模是探讨复杂适应系统演化的新动向.     

基于混合蚁群算法的MAS任务分配*

在MAS（多agent系统）中,由于任务的复杂性和agent求解问题能力的不同,任务和agent不再是传统的一对一的关系。为解决MAS的任务分配问题,提出了任务与agent之间多对多的任务分配模式。首先建立了任务分配的数学模型,并导出分配优化的目标函数;其次利用混合蚁群算法快速收敛和分布式求解的特点实现任务分配的组合优化。对实验仿真的结果分析表明,多对多的任务分配模式能够明显提升多agent系统的性能。     

基于Gaia与JADE框架的MAS建模

为了更加有效地开发MAS系统,简化系统建模流程,传统的建模方法已暴露出其不足之处,因此必须采用新的分析和设计建模方法.提出了Gaia建模方法和JADE框架,Gaia为复杂开放式软件系统的设计和分析提供了清晰的解决方案,JADE则是在符合FIPA规范的MAS系统下开发Agent应用的一种软件框架结构,通过结合Gaia和JADE框架对一个针对手机用户的个人信息MAS系统进行了建模,验证了结合使用Gaia建模和JADE框架能够显著的简化MAS系统开发流程,提高开发效率.     

多agent集成方法在复杂系统故障预报中的应用

基于普适性角度,从单个agent系统出发,结合计算机网络结构,研究了多agent集成方法在复杂系统故障预报中的理论规则。首先,对多agent集成系统的各子agent预测模块进行赋值,并给出相应的信任程度;然后,计算决策agent的预测值,判断系统“异常”与否。同时,给出多agent集成系统的工作机理,并设计出多agent集成系统的网络模型;最后,仿真试验验证了该方法的有效性和合理性。分析研究表明,多agent集成方法在复杂系统的故障预报中有广阔的工程应用前景。关键词复杂系统,故障预报,多agent集成方法     

列车安全距离控制形式化建模与验证

随着我国铁路的迅速发展,对列车运行安全性的要求越来越高。采用Event-B形式化建模方法研究了高速列车安全距离控制形式化验证问题,以Event-B形式化仿真工具Rodin为基础,通过结合多智能体理论,引入感知决策法则,实现了无线闭塞中心(RBC)与列车的车地通信,建立了多列车运行的安全距离控制模型。仿真研究了高速列车最小间隔追踪控制运行,对列车安全距离控车行为进行了形式化建模并进行了POs证明义务验证。仿真结果表明,对于CTCS列车控制系统的复杂逻辑关联行为,采用提出的Event-B和多智能体系统(MAS)结合的形式化验证方法,可进行系统规范的模型验证,对于复杂系统的逻辑验证有较强的实际意义。     

列车安全距离控制形式化建模与验证

随着我国铁路的迅速发展,对列车运行安全性的要求越来越高。采用Event-B形式化建模方法研究了高速列车安全距离控制形式化验证问题,以Event-B形式化仿真工具Rodin为基础,通过结合多智能体理论,引入感知决策法则,实现了无线闭塞中心(RBC)与列车的车地通信,建立了多列车运行的安全距离控制模型。仿真研究了高速列车最小间隔追踪控制运行,对列车安全距离控车行为进行了形式化建模并进行了POs证明义务验证。仿真结果表明,对于CTCS列车控制系统的复杂逻辑关联行为,采用提出的Event-B和多智能体系统(MAS)结合的形式化验证方法,可进行系统规范的模型验证,对于复杂系统的逻辑验证有较强的实际意义。     

个性化Agent在RoboCup仿真比赛中的应用

MAS(multi agent system)是解决复杂、动态、分布式智能应用问题的重要技术,RoboCup仿真比赛提供了一个测试各种MAS理论的平台.基于Agent个性设计并实现了RoboCup仿真球队.实验结果表明该方法比能力对等的合作更加默契,更能体现MAS的自适应性.     

多Agent系统的上下文感知增强

如今,多agent系统(multi-agent system,简称MAS)被广泛用于开发各种应用系统.当前,开放、动态、难控的计算平台要求软件系统能够根据系统自身及其环境状态信息及其改变,动态地调节自身的行为,即具备一定的上下文感知能力.然而,现有工作并未就如何向既有的MAS应用系统引入上下文感知能力提出有效的解决方案.依照关注分离原则,结合面向上下文程序设计技术(context-oriented programming,简称COP)、反射技术(reflection)以及代码植入技术(code instrumentation),提出了一套MAS系统上下文感知增强框架和底层支撑技术.开发人员可以在既有应用源码不可得的情况下,自动地将指定agent类型转换为(扩展为)具有上下文感知能力的agent类型.此外,利用底层运行支撑环境,系统管理员可以在系统运行时刻动态地调整指定agent的上下文感知行为.     

基于Multi-Agent系统的多飞行器协同路径规划方法的研究

无人多飞行器(UAV)协同技术是当前分布式人工智能的一个热点领域,其中一个关键技术在于如何实现多UAV集群根据复杂环境中目标、威胁、地形变化以及各UAV之间的性能约束动态进行实时性航路规划。提出一种基于Multi-agent系统的多UAV对实时动态多目标进行路径规划的方法。其核心是基于Multi-agent系统的decen-tralized控制方案。在Multi-agent平台上,实现了agent对于环境、目标、任务等路劲规划约束条件的建模,同时提出了多agent动态路径规划方法的实现方案。方案使用DisCSP模型框架,将基于真实复杂战场环境的实时路径规划问题所涉及的多复杂限制条件,抽象成Multi-agent系统中的各个约束条件,通过多agent间Dynamic Programming过程求解多UAV实时动态多目标的路径规划和协同任务分配的ABT算法,并实现在动态威胁和地形以及动态目标下具备集群协同能力的多UAV实时仿真系统。     

高层体系结构与多智能体仿真集成研究

对于复杂大系统的建模与仿真,多智能体系统可以提供并行逻辑的支持,而高层体系结构(HLA)则可以提供公共的技术框架.两者的结合能有效地提高系统仿真与建模的效果,但是HLA/RTI的规范对多智能体系统的灵活性与开放性有一定的限制,不能充分发挥智能体在推理交互和协商合作等高层次通信方面的优秀特性.在高层体系结构下利用JADE平台工具对多智能体仿真环境进行了研究,提出了集成系统的总体架构和具体方法,并建立了原型系统,通过实验验证了集成方案的可行性,为进一步的研究奠定了基础.     

基于云模型的Agent 联盟评价

针对多 Agent 系统中影响联盟功效的因素存在较强的模糊性和不确定性的问题,将定性定量转换模型—–云模型引入 Agent 联盟评价中,提出一种新的基于云模型的联盟评价方法.与已有方法相比,该方法简单有效,易于工程实现,可较好地处理评价指标的模糊性和不确定性,从而为复杂控制系统中的最佳联盟选择提供更科学、更合理的决策.     

基于组织的多Agent系统建模方法研究

广域网异质开放的环境给多Agent系统的建模带来了挑战,如何降低系统分析的复杂度、如何提高建模效率成为一个亟待解决的问题.文章从系统工程的角度出发,提出了一种基于组织的多Agent系统建模方法OrgMAS,将复杂系统划分为若干组织分别进行建模设计,从而降低系统分析复杂度.同时,该方法将本体建模贯穿于整个建模过程,使得异质Agent之间能够通过本体映射获得相同的语义基础.文章还介绍了支持该方法的建模辅助工具OrgMAS Tool,该工具在辅助建模的同时进行模型校验,从而能够提高建模效率.     

A graph grammar-based formal validation of object-process diagrams

Two basic requirements from a system’s conceptual model are correctness and comprehensibility. Most modeling methodologies satisfy only one of these apparently contradicting requirements, usually comprehensibility, leaving aside problems of correctness and ambiguousness that are associated with expressiveness. Some formal modeling languages do exist, but in these languages a complete model of a complex system is fairly complicated to understand. Object-process methodology (OPM) is a holistic systems modeling methodology that combines the two major aspects of a system—structure and behavior—in one model, providing mechanisms to manage the complexity of the model using refinement-abstraction operations, which divide a complex system into many interconnected diagrams. Although the basic syntax and semantics of an OPM model are defined, they are incomplete and leave room for incorrect or ambiguous models. This work advances the formal definition of OPM by providing a graph grammar for creating and checking OPM diagrams. The grammar provides a validation methodology of the semantic and syntactic correctness of a single object-process diagram.     

多智能体系统中并发动态知识、信念和肯定性逻辑的研究

将知识、信念和肯定性逻辑从单个智能体扩展到多智能体系统,并且实现了多智能体系统中知识、信念和肯定性逻辑与具有并发动态属性的行为之间的很好结合.以此为基础,提出了多智能体系统中并发动态知识、信念和肯定性逻辑,简称CDKBC逻辑.为了对CDKBC逻辑进行解释,也给出了CDKBC模型,并且讨论了知识、信念和肯定性之间的关系,即知识蕴涵着肯定性,肯定性蕴涵着信念.文中也给出了一个相应的证明系统(即公理系统),证明了该系统是可靠的和完备的,并且证明了系统的有效性问题是EXPTIME完全的.最后论文给出了CDKBC逻辑的实例.     

基于MAS的智能公交调度系统的建模与实现

城市公共交通具有结构复杂、并发性和分布性等特点,使用数学建模的方法很难有效解决其调度问题,而多Agent系统MAS是一种解决复杂系统问题的有效方式,因此应用MAS来解决公交调度问题。首先,建立了城市公交系统的概念模型;然后,根据MAS的建模思想对公交调度相关实体进行抽象,建立了智能公交调度系统的MAS模型,并给出了Dispatch Agent、Bus Agent和Stop Agent的建模方法以及知识库的构建、推理和学习的流程;最后,完成系统实现。     

Cooperative Multi-Agent Learning: The State of the Art

Cooperative multi-agent systems (MAS) are ones in which several agents attempt, through their interaction, to jointly solve tasks or to maximize utility. Due to the interactions among the agents, multi-agent problem complexity can rise rapidly with the number of agents or their behavioral sophistication. The challenge this presents to the task of programming solutions to MAS problems has spawned increasing interest in machine learning techniques to automate the search and optimization process. We provide a broad survey of the cooperative multi-agent learning literature. Previous surveys of this area have largely focused on issues common to specific subareas (for example, reinforcement learning, RL or robotics). In this survey we attempt to draw from multi-agent learning work in a spectrum of areas, including RL, evolutionary computation, game theory, complex systems, agent modeling, and robotics. We find that this broad view leads to a division of the work into two categories, each with its own special issues: applying a single learner to discover joint solutions to multi-agent problems (team learning), or using multiple simultaneous learners, often one per agent (concurrent learning). Additionally, we discuss direct and indirect communication in connection with learning, plus open issues in task decomposition, scalability, and adaptive dynamics. We conclude with a presentation of multi-agent learning problem domains, and a list of multi-agent learning resources.     

Towards modeling and runtime verification of self-organizing systems

According to the fact that the intrinsic dynamism of self-organizing systems challenges the existing methods of engineering for modeling reliable complex systems, in this paper, we propose a new formal-based method to model self-organizing systems. The capabilities of the proposed method which are used to address several challenges in design, development and analysis of self-organizing systems are: modularity and robustness, decentralized control and scalability, required adaptation types, flexible and adaptive control mechanism, separation of adaptation and business logic, and safe adaptation. To evaluate the proposed method, we use self-organizing traffic management system as a case study and exploit the proposed method for modeling this dynamic system. Moreover, we propose and employ a novel policy-based runtime verification mechanism to ensure that the safety properties are satisfied by the implementation at runtime. We provide our case study prototype using Java and the Ponder2 toolkit and apply our runtime verification method to show its proper reaction capabilities to the property violations. This benefit is the result of using dynamic policies in our method to control the behavior of systems.     

An event-based approach for formally verifying runtime adaptive real-time systems

Real-time and embedded systems are required to adapt their behavior and structure to runtime unpredicted changes in order to maintain their feasibility and usefulness. These systems are generally more difficult to specify and verify owning to their execution complexity. Hence, ensuring the high-level design and the early verification of system adaptation at runtime is very crucial. However, existing runtime model-based approaches for adaptive real-time and embedded systems suffer from shortcoming linked to efficiently and correctly managing the adaptive system behavior, especially that a formal verification is not allowed by modeling languages such as UML and MARTE profile. Moreover, reasoning about the correctness and the precision of high-level models is a complex task without the appropriate tool support. In this work, we propose an MDE-based framework for the specification and the verification of runtime adaptive real-time and embedded systems. Our approach stands for Event-B method to formally verify resources behavior and real-time constraints. In fact, thanks to MDE M2T transformations, our proposal translates runtime models into Event-B specifications to ensure the correctness of runtime adaptive system properties, temporal constrains and nonfunctional properties using Rodin platform. A flood prediction system case study is adopted for the validation of our proposal.