ロボットサッカーにおける生体配列解析手法を強化するセルオートマタ型フィードバック機構
Cellular Automata based Feedback Mechanism in Strengthening biological Sequence Analysis Approach to Robotic Soccer
ロボットサッカーのエージェントに配列解析アルゴリズムを適用し、セルオートマタ型分類器をバケットブリゲード法で強化することで、試合中にリアルタイムで戦略を学習・適応させる手法を提案した。
著者: P. Kiran Sree, G. V. S. Raju, S. Viswandha Raju, N. S. S. S. N Usha Devi
分類: cs.MA, cs.RO
原文アブストラクト
This paper reports on the application of sequence analysis algorithms for agents in robotic soccer and a suitable representation is proposed to achieve this mapping. The objective of this research is to generate novel better in-game strategies with the aim of faster adaptation to the changing environment. A homogeneous non-communicating multi-agent architecture using the representation is presented. To achieve real-time learning during a game, a bucket brigade algorithm is used to reinforce Cellular Automata Based Classifier. A technique for selecting strategies based on sequence analysis is adopted.
