紀要論文 視野制限環境下での追跡問題における強化学習の学習効率向上法

山森, 一人  ,  吉田, 雅也  ,  相川, 勝

46pp.313 - 316 , 2017-07-31 , 宮崎大学工学部
ISSN:05404924
内容記述
Reinforcement learning is one of the unsupervised machine learning methods. In reinforcement learning, multiagent learning is usually used. In recent years, some researchers try to utilize drones in search of missing persons, but efficiency is not enough for the real world. In this research, we propose a multiple single role agent system which gives some roles for layered agents. Experimental results showed that our method was able to capture the target with 30% faster steps than the conventional method.
本文を読む

http://opac2.lib.miyazaki-u.ac.jp/webopac/bdyview.do?bodyid=TC10156638&elmid=Body&fname=p313_cover.pdf

このアイテムのアクセス数:  回

その他の情報