高速なガウス過程回帰を用いた予測モデルに基づくサンプリングベース動作計画

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2015/05/01

タイトル(英語): Sampling-based Motion Planning with a Prediction Model using Fast Gaussian Process Regression

著者名: 岡留　有哉（大阪大学 基礎工学研究科 システム創成専攻），中村　泰（大阪大学 基礎工学研究科 システム創成専攻），石黒　浩（大阪大学 基礎工学研究科 システム創成専攻）

著者名(英語): Yuya Okadome (Osaka University, Graduate of Engineering Science, Department of System Innovation), Yutaka Nakamura (Osaka University, Graduate of Engineering Science, Department of System Innovation), Hiroshi Ishiguro (Osaka University, Graduate of Engineering Science, Department of System Innovation)

キーワード: ガウス過程，動作計画，局所性鋭敏型ハッシュ　　Gaussian process regression，motion planning，locality-sensitive hashing

要約(英語): Recently, motion/path planning methods become popular and achieve practical applications such as autonomous car. Most of studies focuses on the planning based on a given mathematical model, but the development of data-driven system identification method is also crucial for practical applications, since the precise model of a control target is not always available in advance. In this research, we propose a motion planning method where a fast Gaussian process regression is used as a model of the control target, since Gaussian process regression is a powerful non-parametric method which is widely used in various practical and complicate applications. Thanks to Bayesian property of Gaussian process regression, our method can deal with uncertainty of the prediction. We apply our method to the control problem of simple-pendulum with non-linearity and achieve the swinging up task.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.135 No.5 （2015） 特集：看護現場の安全性確保のための支援技術

本誌掲載ページ: 526-533 p

原稿種別: 論文／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/135/5/135_526/_article/-char/ja/