since 2002-01-05
Research Statement: ( 2000-03-25)
(1)多智能体控制系统结构对控制系统采用具有开放性的分布式体系结构,研究如何赋予系统各组成实体或子系统以较大的自主权,使其形成智能体并构成网络,人也可作为该网络中的一个成员,在此网络中,各智能体在逻辑上是平等的,在物理上是分散的,在功能上是独立的。它们之间是一种松散耦合的关系,并可根据需要建立和变更关系的层次,各智能体间通过消息传递相互联系,协调合作,共同完成控制任务,这种控制体系结构允许系统中的任何智能体可以随时加入或退出系统,也可以根据当时的任务和总体目标,通过公用消息机制进行联合,即具备以全局或局部动态组态的方式来解决问题的能力。这里的分布系统具有开放性,具体有以下三层含义:①结构的开放性:能容纳来自系统内部的变化(如故障),能适应外界对系统的干扰和改变(如改变物理设备的配置);②任务的开放性:任务是随时输入、随时处理的;③问题求解的开放性:问题求解过程能接受信息和知识的变化,从而改进控制策略,并按照目标作出决策。
(2)基于现场总线的智能体构造智能体可以是抽象实体也可以是物理实体,现场总线设备这种物理实体可构造成智能体,使其具有感知、通信,行动和控制及推理的基本功能,它不仅能作用于自身而且可以施动于环境,并能接受环境的反馈信息,重新评估、控制自己的行为。主动去适应环境的改变,不仅如此,智能体还可以根据需要自由地与其它智能体结成伙伴或层次关系来共同完成同一目标下的某种任务。即一方面要解决智能体自身的学习和问题求解能力,另一方面要增强其与外界环境的协调能力,具体来说就是强化局部数据,削弱全局数据,变紧密耦合方式为松散耦合方式,从而提高系统的适应性和容错性,为保证全局,还要构造协调智能体,完成协同工作任务。
(3) 多智能体的协同机制协同工作是多智能体分布式控制研究的重要问题之一,如怎样规范智能体的思考和行为,怎样将个体与群体联系起来产生局部的或全局的联合行动去完成某一控制任务,以及如何对联合行动的初始化和整个过程进行指导和监控,还要考虑冲突的解决办法、个别智能体故障后的互相替代、问题的分解与协同等问题。Haken的协同学理论指出,子系统之间的合作和竞争在一定条件下,可以自发导致宏观有序。控制系统的智能化,不是通过预先准备大量的行为规则,而是通过智能体化的控制设备的自组织来实现的。这种自组织的关键就是在协同原则(如有序化、层次化、渐进机构化、渐进中心化等)的指导下,制定智能体行为的法律体系,使智能体可以通过对局部不完全信息的掌握来实现全局的合作。从系统的协同基本原理出发,可细化面向分布式控制领域的智能体行为规范和通信语言。
智能体的构造分为设备智能体(含传感智能体、执行智能体等)和协调智能体。设备智能体的构造将在分布式控制系统基础上,增强设备的智能程度(如对环境的适应能力等)和通信能力来实现。协调智能体将完成设备智能体间、设备智能体与协调智能体间及协调智能体自身间的通信、协作和控制,由软件实现。多智能体的协同机制将通过制定各种控制智能体间的通信规范实现。目前已有智能体通信语言出现,但对控制领域的具体情况而言并不能满足要求,在系统协同发展的一般原则指导下,在现有智能体通信语言的基础上,可吸收现场总线网络协议中的有效成份,确定控制智能体间通信的词汇表、内部语言和外部语言;在外部语言层,详细规范通信体系结构、通信模式、语法和语义及语言的层次结构。
(2001-04-01)
The history of distributed control system has experienced stages of DCS, Field Bus System, and Intelligent Control Maintenance and technical Management System (ICMMS). Two important trends in the development of distributed control system are the intelligentization of control matters and increasing synergism of subsystems. Its further development has close relationship with agent technology and CSCW. The intelligentization of control system is not implemented by preparing a lot of behavior rules in advance, but by the self-organization of control matters from the view of agent. The key issue in the self-organization is how to enact law system that enables global cooperation between agents with incomplete local information under the guidance of synergetic principles, such as order increasing, hierarchy structure, progressive centralization and progressive mechanization.
Agent in this system can be classified into Device Agent and Management Agent. The Device Agent is implemented on the basis of distributed intelligent control maintenance and technical management system with enhancement of the communicative ability between devices. The Management Agent undertakes such tasks as the communication, cooperation and control between Device Agent, between Device Agent and Management, and between Management Agent themselves. The Management Agent is implemented by software. Synergism between agents is implemented by defining communicative protocols between various control agents. Although some agent language have already been developed, them are not suitable entirely for the special requirement of control system. Based on the LonTalks language of LonWorks Fieldbus, the vocabulary, internal language and external language of communication between control agents can be defined. At the external language layer, the structure of communication system, communication mode, grammar, semantics and the hierarchical structure of language need to be speicified.
[Return to Jie Bao's Homepage]