Picking up One of the Folded and Stacked Towels by a Single Arm Robot
Satonori Demura, Kazuki Sano, Wataru Nakajima, Kotaro Nagahama, Keisuke Takeshita, Kimitoshi Yamazaki
- Year
- 2018
- Citations
- 16
Abstract
This paper describes methods for doing a manipulation of picking up the top one of towels folded and stacked on a table. In this task, It is necessary to guarantee the reproducibility of shape of the picked towel when placing it again. Also, it is necessary to carry out lifting motion without moving other remaining towels. To overcome these issues, we propose work procedures and also propose methods for each procedure. The topics of this paper include the grasping position detection. We introduce YOLO to correctly detect the portion where should be grasped for stable manipulation. Also, reinforcement learning is applied for a grasping motion acquisition. We introduce Q-learning to obtain a simple and successful manipulation. The effectiveness of the proposed procedure and the methods were proven through experiments using an actual robot.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002