面向运行时协作的异构Agent能力选择与补偿方法研究  

Runtime-Oriented Collaborative Heterogeneous Agent's Capability Choosing and Compensation

作　　者：李爽[1] 刘玮[1] 吴坤[1] 王晶[1] LI Shuang LIU Wei WU Kun WANG Jing(Hubei Province Key Laboratory of Intelligent Robot, Wuhan Institute of Technology,Wuhan 430205 ,Chin)

机构地区：[1]武汉工程大学智能机器人湖北省重点实验室,湖北武汉430205

出　　处：《烟台大学学报（自然科学与工程版）》2017年第2期167-172,共6页Journal of Yantai University(Natural Science and Engineering Edition)

基　　金：国家自然科学基金资助项目(61502355);武汉工程大学科学研究基金(K201475);武汉工程大学第七届研究生教育创新基金(CX2015060)

摘　　要：介绍了自适应Agent的能力模型,并在此基础上提出一种基于规划的能力选择和补偿方法.首先,定义了能力元模型,能力元模型表示了业务和信息系统的设计,包括目标、协作、能力、上下文;然后,提出一种来进行Agent能力协作的能力选择与补偿方法,该方法解决了多Agent能力与任务之间协作问题;最后,通过用AGVs模拟演示医疗垃圾运输系统验证了能力选择与补偿方法.Dynamic decision and planning are needed at runtime in self-adaptation and multi-agent cooperative systems. In the description for Agent＇s execution and performance in different environments and state,the agent of adaptive collaboration system is required to support run-time dynamic decision-making and planning. The capability model of adaptive Agent is introduced,and a planning-based capability choosing and compensation method is presented. First of all,we define a meta-model to describe the design of business and information systems including objectives,coordination,ability and context. Then we propose a capability choosing and compensation to establish a collaboration in Agent＇s capability,settling the coordination problems between the agent capability and the task.Finally,the capability-collaborative modeling is well tested by using AGVs to simulate medical waste transporting system.

关 键 词：能力选择与补偿 能力驱动开发 医疗垃圾运输系统 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]