报告详情
  1. 活动首页
  2. 报告清单
  3. 报告详情
Adaptive Agent Community Framework for Real-Time Autonomous Decision Ecosystems
编号：150 访问权限：仅限参会人 更新：2026-07-27 12:58:35 浏览：13次 Online

报告开始：2026年07月31日 14:55（Asia/Kolkata）

报告时间：15min

所在会场：[S7] Disruptive Technologies for Manufacturing [S7-2] Disruptive Technologies for Manufacturing

演示文件

提示：该报告下的文件权限为仅限参会人，您尚未登录，暂时无法查看。

摘要
Abstract—In autonomously operating decision ecosystems, more and more intelligent agents will be deployed in a distributed way, and the agents have to cooperate and operate under conditions where limited bandwidth and changing environmental conditions occur. Centralized controllers, federated learning pipelines and swarm heuristics solve some aspects of the problem, but are all difficult to simultaneously optimize for accuracy, trust, latency & resilience at scale. This paper introduces the Adaptive Agent Community Framework (AACF), which is a self-organising architecture in which autonomous agents participate in communities that are dynamically formed according to the agents' trust estimates, share knowledge among themselves, and achieve adaptive consensus on whatever they have to do using a learning decision engine that constantly receives feedback from the execution of the decision. The conceptual structure combines the process of trust assessment with the creation of the community and also links it with adaptive learning in a closed operational concept and not a static pipeline. Mathematical models are introduced to determine trust scoring, community affinity, knowledge utility, decision confidence, weighted consensus, community stability and adaptation efficiency and a twelve-step operational algorithm is presented. AACF achieves an accuracy in decision making of 95.1%, decisiveness in governance of 92.6%, average response latency after adoption of about 96ms, outperforming the Centralized, Federated Multi-agent and Swarm Intelligence baselines, highlighting the level of accuracy, robustness, scalability and adaptivity, in addition to reducing the communication overhead by up to 33% while applying Simulation experiments to different numbers of agents from 10 to 100. From these results, it can be concluded that trust-aware adaptation with community participation can be a sound direction towards achieving scalable approaches to Real-Time Autonomous Decision Ecosystems.
 
关键词
adaptive agent community; multi-agent systems; trust assessment; consensus generation; autonomous decision-making; real-time systems; distributed intelligence.
报告人
Elamathi E
ASSISTANT PROFESSOR K.Ramakrishnan College of Engineering

稿件作者
Kanchana K Meenakshi College of Arts and Science, Meenakshi Academy of Higher Education and Research
Narendra G CMR College of Engineering & Technology, Hyderabad
Saurabh Mr School of Engineering and Technology (SET), CGC University, Mohali -140307
NIRAKULANATHAN P Shri Venkateshwara Padmavathy Engineering College
BHUVANESWARI M Karpagam Academy of Higher Education Coimbatore
Elamathi E K.Ramakrishnan College of Engineering
发表评论
验证码 看不清楚，更换一张
全部评论
重要日期

  • 会议日期

    07月30日

    2026

    至

    08月01日

    2026

  • 07月28日 2026

    初稿截稿日期

  • 08月03日 2026

    注册截止日期

主办单位
The United Societies of Science
承办单位
Kongunadu College of Engineering and Technology
协办单位
IEEE Section
IEEE Madras Section
联系方式
登录查看完整联系方式
历届会议
移动端
在手机上打开
小程序
打开微信小程序
客服
扫码或点此咨询