ロボット研究地図　複数自律ロボット系における漸進的戦略形成に関する研究

• 太田 順

  東京大学工学系研究科

• 横川 洋一

  豊田自動織機製作所

• 新井 民夫

  東京大学工学系研究科

書誌事項

タイトル別名

• Research Map of Robotics. Study on Asymptotic Strategy-Making in Multiple Autonomous Robot System.

抄録

A method of distributed strategymaking in multiple autonomous system is presented in this paper. Every robot in the system has several tactics to be selected, and set of probabilities to select tactics are called “strategy”. The algorithm to make strategy in this paper is iteration of four parts as follows: (1) select and try a certain tactic by means of the strategy, (2) get payoff value for applied tactic from environment, (3) revise the estimated payoff value of tactics, (4) revise the strategy by using (3) . Effectiveness of the proposed method is verified by path selecting simulation of multiple mobile robots. Convergence to optimal solution in the proposed method is discussed by using pay-off matrices in game theory.

収録刊行物

• 日本ロボット学会誌

  日本ロボット学会誌 14 (3), 379-385, 1996

  一般社団法人 日本ロボット学会

キーワード

• Multiple Mobile Robot System
• Strategy
• Reinforcement Leaning
• Theory of Game

詳細情報

• CRID: 1390282679701703040
• NII論文ID: 10007547322
• NII書誌ID: AN00141189
• DOI: 10.7210/jrsj.14.379
• ISSN: 18847145
• ISSN: 02891824
• 本文言語コード: ja
• データソース種別:
  • JaLC
  • Crossref
  • CiNii Articles
• 抄録ライセンスフラグ: 使用不可