Self-supervised 6-DoF Robot Grasping by Demonstration via Augmented Reality Teleoperation System

arXiv:2404.03067v1 Announce Type: cross
Abstract: Most existing 6-DoF robot grasping solutions depend on strong supervision on grasp pose to ensure satisfactory performance, which could be laborious and impractical when the robot works in some restricted area. To this end, we propose a self-supervised 6-DoF grasp pose detection framework via an Augmented Reality (AR) teleoperation system that can efficiently learn human demonstrations and provide 6-DoF grasp poses without grasp pose annotations. Specifically, the system collects the human demonstration from the AR …

abstract arxiv augmented reality cs.cv cs.ro detection framework grasping performance reality robot solutions supervision teleoperation type via