Enhancing position-based visual servoing performance through transformer-based acceleration-level reinforcement learning

Abstract Visual servoing is a fundamental approach for robotic manipulation that relies on visual feedback to precisely control robot motion. Most methods are capable of generating velocity control signals to guide the camera to the desired position and orientation, which often exhibit limitations i...

Ausführliche Beschreibung

Bibliografische Detailangaben

Hauptverfasser:	Zongdao Li, Wenkai Chen, Siqin Wang, Pengfei Wang, Ye Yuan, Tao Wu, Jianwei Zhang, Qingdu Li
Format:	Artigo
Sprache:	Inglês
Veröffentlicht:	Springer 2025-08-01
Schriftenreihe:	Complex & Intelligent Systems
Schlagworte:	Position-based visual servoing Acceleration-level controller Deep reinforcement learning Robot manipulator
Online-Zugang:	https://doi.org/10.1007/s40747-025-02056-8
Tags:	Tag hinzufügen Keine Tags, Fügen Sie das erste Tag hinzu!