基于命令的多智能体组织模型研究

针对经典的BDI模型并不能很好地描述Agent的思维状态的问题.提出了一种基于命令的BGC模型,能够对Agent的行为意图和社会属性如命令、义务等进行描述.对群体思维状态进行了分析,构造了一种面向命令和角色的MAS组织结构COS,对角色任务关系和组织形成过程等进行了描述.实际仿真应用表明,BGC模型及COS组织结构能够较好地满足CMAS对灵活性和可靠性的要求.     

基于MAS的多机械手协调工作控制

为了解决全自动化晶体切片角分类机三个机械手协调工作的控制问题,设计了以MAS理论为基础的分布式网络控制器.系统利用黑板结构来实现全局动态目标下的状态信息的共享,采用多Agent理论对三个机械手进行协调工作控制,解决了机械手之间的运动耦合问题,使之能够在同一个时间和不同的空间内协调与配合工作,完成整体工作任务.基于此设计的设备可改善工人的劳动环境、提高生产效率、确保晶体切片切角的测量与分类准确.     

Component allocation and job sequencing for two series-connected SMD placement machines

The problem of component allocation and sequencing PCB assembly jobs for two series-connected SMD placement machines is addressed. With the allocation of components, a balancing of the workload per PCB for the two machines should be achieved. Furthermore, when switching from one type of PCB to another, setup times must be considered. Two heuristic procedures are developed and tested in extensive numeric analyses of a number of realistic case models. The problem examined and the database used represent a concrete case of an industrial application in which very diversified small jobs are to be processed.     

基于免疫Agent的风力发电机组故障自愈系统的研究

以风力发电机组为研究对象,将Multi-Agent技术与故障自愈的原理结合,并引入人工免疫的思想,以故障的消除和抑制为目的,探讨一种新型的具有自身免疫能力的智能自愈理论和方法研究。采用Multi-Agent技术、人工免疫原理和专家系统等混合智能技术构建故障自愈免疫系统的结构模型,阐述了基于免疫agent的风力发电机组故障自愈系统的组织结构和诊断流程。该系统具有可扩充性、诊断资源的可重用性、诊断过程的自主性等,为机电装备故障诊断和自愈提供了一种新思路。     

An optimal cooperation in a team of agents subject to partial information

The objective of this work is to design a controller for a team of agents to accomplish cohesive motion with consensus on an agreed upon trajectory in both leaderless and modified leader–follower structures. The agent's dynamical model is considered to be governed by a general linear representation instead of a point-mass model which is commonly used in the literature. A semi-decentralised optimal control strategy is designed based on minimisation of individual cost functions over a finite horizon using local information. Decentralisation is achieved through incorporating interaction terms in the team members model. Minimisation of our proposed cost function results in a modified version of the consensus algorithm. Corresponding to the leader–follower structure, the utilisation of a corrective feedback from followers to the leader provides potential advantages in terms of improving robustness of the team to faults in individual agents. Simulation results are presented to demonstrate the effectiveness and capabilities of our proposed method in achieving user predefined requirements and specifications.     

Data-enabled extremum seeking: A cooperative concurrent learning-based approach

This paper introduces a new class of feedback-based data-driven extremum seeking algorithms for the solution of model-free optimization problems in smooth continuous-time dynamical systems. The novelty of the algorithms lies on the incorporation of memory to store recorded data that enables the use of information-rich datasets during the optimization process, and allows to dispense with the time-varying dither excitation signal needed by standard extremum seeking algorithms that rely on a persistence of excitation (PE) condition. The model-free optimization dynamics are developed for single-agent systems, as well as for multi-agent systems with communication graphs that allow agents to share their state information while preserving the privacy of their individual data. In both cases, sufficient richness conditions on the recorded data, as well as suitable optimization dynamics modeled by ordinary differential equations are characterized in order to guarantee convergence to a neighborhood of the solution of the extremum seeking problems. The performance of the algorithms is illustrated via different numerical examples in the context of source-seeking problems in multivehicle systems.     

多Agent诊断系统的设计

面向代理的技术对于解决复杂动态系统的问题求解，实现集成化、智能化、自动化和网络化具有重要作用。分别介绍了Agnet和Agent结构，为提高诊断系统的自学习能力、并行处理能力和应急反应能力，提出了Agent诊断系统的设计方案。     

流程企业供应链管理的建模问题研究

分析了流程企业的集成体系结构特点,提出了适合流程企业的基于多Agent的供应链管理的框架模型,并给出了agent的内部结构,最后给出其在流程企业中的应用。     

基于多Agent技术的分布式变电站远程视频监控系统研究

针对传统变电站多媒体监控系统在灵活性、可扩展性等方面的不足,引入智能Agent技术．建立了基于Multi-Agent的分布式多媒体监控系统模型。该模型将传统监控系统的功能划分入不同的Agent工作组,通过各Agent之间基于消息机制的通信与协作实现恶劣条件下视频流的鲁棒传输,并使系统具有感知性、主动性和自适应性。分析了系统结构与工作原理,给出了Agent的功能模型．并对模型中数据压缩组和错误控制组中所分别采用的自适应快速运动估计算法及精细运动矢量场差错掩盖算法进行了介绍。分析表明．该模型的视频压缩和传输策略可随系统结点数目和通信模式的变化而动态调整。     

基于图神经网络的最大化代数连通度算法

随着智能体数量的增加, 多智能体系统中潜在的通信链路数量呈指数级增长. 过多冗余链路的存在给系统带来了大量的能源浪费和维护成本, 而盲目地去除链路又会降低系统的稳定性和安全性. 代数连通度是衡量图连通性的重要指标之一. 然而, 传统的半正定规划(SDP)方法和启发式算法在求解大规模场景下的最大化代数连通度问题时非常耗时. 在本文中, 我们提出了一种监督式的图神经网络模型来优化多智能体系统的代数连通度. 我们将传统的SDP方法应用于小规模任务场景中, 得到足够丰富的训练样本和标签. 在此基础上, 我们训练了一个图神经网络模型, 该模型可用于更大规模的任务场景中. 实验结果表明, 当需要去除15条边时, 我们的模型的平均性能达到了传统SDP方法的98.39%. 此外, 我们的模型计算时间极其有限, 可以推广到实时场景中去.