• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2023 Fiscal Year Final Research Report

An Experimental Study for Highly-Reliable Cooperative Behavior of Swarm-Robots using Blockchain and Distributed Deep Learning

Research Project

  • PDF
Project/Area Number 21K03979
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 20020:Robotics and intelligent system-related
Research InstitutionTomakomai National College of Technology (2022-2023)
Ehime University (2021)

Principal Investigator

Sugimoto Masashi  苫小牧工業高等専門学校, 創造工学科, 准教授 (40780424)

Co-Investigator(Kenkyū-buntansha) 曽利 仁  津山工業高等専門学校, 総合理工学科, 教授 (10353327)
都築 伸二  愛媛大学, 理工学研究科(工学系), 教授 (60236924)
漆原 史朗  香川高等専門学校, 電気情報工学科, 教授 (90311092)
Project Period (FY) 2021-04-01 – 2024-03-31
Keywords深層学習 / LPWA / 群ロボット / ブロックチェーン
Outline of Final Research Achievements

In this study, we aim to propose a system combining Low Power, Wide Area (LPWA) systems, Blockchain, and Multi-Agent Autonomous Decentralized Deep Learning for controlling swarm robots. We have simulated a cooperative operation system using Dueling DQN and modeled LoRa wireless communication with Blockchain. The results of our validation experiments confirmed that the system achieved performance comparable to conventional multi-agent reinforcement learning systems. Moving forward, we intend to implement the system in actual hardware to investigate its specific behavior in real-world environments.

Free Research Field

ソフトコンピューティング

Academic Significance and Societal Importance of the Research Achievements

本研究では,群ロボットの制御に,Low Power, Wide Area(LPWA)システム,Blockchain,およびMulti-Agent自律分散型深層学習を組み合わせたシステムを提案した.これにより,エネルギ効率の高い広域通信技術とセキュアなデータ管理,および高度な自律学習を融合させた新たな協調動作システムを構築した.シミュレーションを通じて,Dueling DQNとLoRa無線,Blockchainを用いた本システムが,従来のマルチエージェントシステムと遜色ないパフォーマンスを発揮することを確認した.この成果は,スマートシティや農業,災害対応など広範な社会的課題への応用が期待される.

URL: 

Published: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi