日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

力制御arXiv:1503.00603

剛環境下でのハイブリッド位置・力軌道追従のためのスイッチング制御

Switching control for tracking of a hybrid position-force trajectory

シェア:XThreadsFacebookLINEはてブBluesky

剛い環境と接触するマニピュレータで位置と力の時間変化軌道を追従するスイッチング制御を提案し、安定性条件を解析した上で、過大な減衰を避けるためにコンプライアント手首の再設計指針を示した論文。

著者: D. J. F. Heck, A. Saccon, N. van de Wouw, H. Nijmeijer

分類: cs.RO

原文アブストラクト

This work proposes a control law for a manipulator with the aim of realizing desired time-varying motion-force profiles in the presence of a stiff environment. In many cases, the interaction with the environment affects only one degree of freedom of the end-effector of the manipulator. Therefore, the focus is on this contact degree of freedom, and a switching position-force controller is proposed to perform the hybrid position-force tracking task. Sufficient conditions are presented to guarantee input-to-state stability of the switching closed-loop system with respect to perturbations related to the time-varying desired motion-force profile. The switching occurs when the manipulator makes or breaks contact with the environment. The analysis shows that to guarantee closed-loop stability while tracking arbitrary time-varying motion-force profiles, the controller should implement a considerable (and often unrealistic) amount of damping, resulting in inferior tracking performance. Therefore, we propose to redesign the manipulator with a compliant wrist. Guidelines are provided for the design of the compliant wrist while employing the designed switching control strategy, such that stable tracking of a motion-force reference trajectory can be achieved and bouncing of the manipulator while making contact with the stiff environment can be avoided. Finally, numerical simulations are presented to illustrate the effectiveness of the approach.

関連論文

  1. 1Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  2. 2newmoとKDDIが資本業務提携 2028年に堺市で自動運転タクシー商用化目指す提携・買収
  3. 3GXOとApptronik、Apolloの物流現場試験を拡大 最大1,000台の展開を構想量産・展開
  4. 4NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  5. 5UC BerkeleyとNVIDIAの研究者、AIエージェントでロボット制御コードを仮想環境で改善研究・技術

PR・アフィリエイト広告

OZ GAMING(GPU搭載BTOゲーミングPC)

OZ GAMING(GPU搭載BTOゲーミングPC)の公式広告バナーOZ GAMINGで完成品PCを見る ↗
PR本紙発行元 EmplifAI