A gen t和多A gen t系统已经成为计算机科学和分布式人工智能研究的一个重要方向。该文重点介绍了基于多A gen t技术的操作票生成系统的模型的建立,研究了通过A gen t选择意图模型来实现操作任务的规划,以及对单个A gen t的推理机制和...A gen t和多A gen t系统已经成为计算机科学和分布式人工智能研究的一个重要方向。该文重点介绍了基于多A gen t技术的操作票生成系统的模型的建立,研究了通过A gen t选择意图模型来实现操作任务的规划,以及对单个A gen t的推理机制和知识库中的符号化描述,并讨论了A gen t自学习及多A gen t之间的交互通信等问题。开发实现基于多A gen t变电站倒闸操作票生成系统,在某变电站得到良好的应用。实践表明,基于多A gen t系统能大大提高工作效率和准确度,特别是基于A gen t选择意图模型来实现操作任务的规划更能适应外界环境的变化。展开更多
提出了一种利用多智能体(M u lti-A gent System s,MA S)方法学指导复杂机电设备嵌入式控制系统开发的方法.该方法利用智能体(A gent)概念处理复合型控制系统的多个子控制器;用MA S方法学指导构造A gent之间的通信及协调机制;利用多任...提出了一种利用多智能体(M u lti-A gent System s,MA S)方法学指导复杂机电设备嵌入式控制系统开发的方法.该方法利用智能体(A gent)概念处理复合型控制系统的多个子控制器;用MA S方法学指导构造A gent之间的通信及协调机制;利用多任务实时操作系统(R eal-T im e O per-ating System,RTOS)的多任务并行处理能力和系统服务在嵌入式系统中实现MA S系统.本方法已经应用于遥操作移动机器人嵌入式控制系统的开发.展开更多
A group of cooperative agents can finish complicated missions that are difficult for a large machine.In the past two decades,spacecraft attitude coordination has attracted significant research attention owing to its w...A group of cooperative agents can finish complicated missions that are difficult for a large machine.In the past two decades,spacecraft attitude coordination has attracted significant research attention owing to its wide potential applications.This paper presents a survey of recent research progress on the spacecraft attitude consensus problem,paying particular attention to the papers published in major aerospace,dynamics,automation,and robotics journals since 2015.Attitude consensus concepts for centralized,decentralized,and distributed cases are reviewed.This overview summarizes results on system dynamics and consensus algorithms based on frequently used attitude representations,such as Euler angles,modified Rodrigues parameters,unit quaternions,and rotation matrices.Studies conducted under complicated operating conditions are also covered.Experimental results on attitude consensus are discussed.In the final section,the main conclusions are drawn and several potential research directions are provided.展开更多
文摘A gen t和多A gen t系统已经成为计算机科学和分布式人工智能研究的一个重要方向。该文重点介绍了基于多A gen t技术的操作票生成系统的模型的建立,研究了通过A gen t选择意图模型来实现操作任务的规划,以及对单个A gen t的推理机制和知识库中的符号化描述,并讨论了A gen t自学习及多A gen t之间的交互通信等问题。开发实现基于多A gen t变电站倒闸操作票生成系统,在某变电站得到良好的应用。实践表明,基于多A gen t系统能大大提高工作效率和准确度,特别是基于A gen t选择意图模型来实现操作任务的规划更能适应外界环境的变化。
文摘针对实际多智能体系统对交互经验的庞大需求,在单智能体领域分布式架构的基础上,提出概率经验优先回放机制与分布式架构并行的多智能体软行动-评论者算法(multi-agent soft Actor-Critic with probabilistic prioritized experience replay based on a distributed paradigm, DPER-MASAC).该算法中的行动者以并行与环境交互的方式收集经验数据,为突破单纯最近经验在多智能体高吞吐量情况下被高概率抽取的局限性,提出更为普适的改进的基于优先级的概率方式对经验数据进行抽样利用的模式,并对智能体的网络参数进行更新.为验证算法的效率,设计了难度递增的2类合作和竞争关系共存的捕食者-猎物任务场景,将DPER-MASAC与多智能体软行动-评论者算法(multi-agent soft Actor-Critic, MASAC)和带有优先经验回放机制的多智能体软行动-评论者算法(multi-agent soft Actor-Critic with prioritized experience replay, PER-MASAC)2种基线算法进行对比实验.结果表明,采用DPER-MASAC训练的捕食者团队其决策水平在最终性能和任务成功率2个维度上均有明显提升.
基金Manuscript received March 5, 2010 accepted March 2, 2011 Supported by National Natural Science Foundation of China （61004103）, National Research Foundation for the Doctoral Program of Higher Education of China （20100111110005）, China Postdoctoral Science Foundation （20090460742）, and Natural Science Foundation of Anhui Province of China （090412058, 11040606Q44）
文摘提出了一种利用多智能体(M u lti-A gent System s,MA S)方法学指导复杂机电设备嵌入式控制系统开发的方法.该方法利用智能体(A gent)概念处理复合型控制系统的多个子控制器;用MA S方法学指导构造A gent之间的通信及协调机制;利用多任务实时操作系统(R eal-T im e O per-ating System,RTOS)的多任务并行处理能力和系统服务在嵌入式系统中实现MA S系统.本方法已经应用于遥操作移动机器人嵌入式控制系统的开发.
基金This work was supported by the National Natural Science Foundation of China under Grant No.12102174the Science and Technology on Space Intelligent Control Laboratory(Grant No.2021-JCJQ-LB-010-17)+1 种基金the Fundamental Research Funds for the Central Universities(No.NP2022301)the Research Fund of State Key Laboratory of Mechanics and Control of Mechanical Structures(Nanjing University of Aeronautics and astronautics)(Grant No.MCMS-I-0122K01).
文摘A group of cooperative agents can finish complicated missions that are difficult for a large machine.In the past two decades,spacecraft attitude coordination has attracted significant research attention owing to its wide potential applications.This paper presents a survey of recent research progress on the spacecraft attitude consensus problem,paying particular attention to the papers published in major aerospace,dynamics,automation,and robotics journals since 2015.Attitude consensus concepts for centralized,decentralized,and distributed cases are reviewed.This overview summarizes results on system dynamics and consensus algorithms based on frequently used attitude representations,such as Euler angles,modified Rodrigues parameters,unit quaternions,and rotation matrices.Studies conducted under complicated operating conditions are also covered.Experimental results on attitude consensus are discussed.In the final section,the main conclusions are drawn and several potential research directions are provided.
