Route optimization for autonomous bulldozer by distributed deep reinforcement learning

Yasuhiro Osaka, Naoya Odajima, Yutaka Uchimura

研究成果: Conference contribution

抄録

Since the publication showed DQN based reinforcement learning methods exceeds human's score in Atari 2600 video games, various deep reinforcement learning have bee researched. This paper proposes a method to control bulldozer autonomously by learning the sediment leveling route using PPO that enables distributed deep reinforcement learning. The simulator was originally developed that enables to reproduce the behavior of small and uniform sediment. By incorporating an LSTM that processes the input state as time-series data into the agent network, more than 95% of the sediment in the target area on average was achieved. In addition, the generalization performance for unknown condition was evaluated, by giving unlearned conditions were given as initial setups.

本文言語English
ホスト出版物のタイトル2021 IEEE International Conference on Mechatronics, ICM 2021
出版社Institute of Electrical and Electronics Engineers Inc.
ISBN(電子版)9781728144429
DOI
出版ステータスPublished - 2021 3 7
イベント2021 IEEE International Conference on Mechatronics, ICM 2021 - Kashiwa, Japan
継続期間: 2021 3 72021 3 9

出版物シリーズ

名前2021 IEEE International Conference on Mechatronics, ICM 2021

Conference

Conference2021 IEEE International Conference on Mechatronics, ICM 2021
国/地域Japan
CityKashiwa
Period21/3/721/3/9

ASJC Scopus subject areas

  • 人工知能
  • 機械工学
  • 制御と最適化

フィンガープリント

「Route optimization for autonomous bulldozer by distributed deep reinforcement learning」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル