Title :
Direct perception and action system for unknown object grasping
Author :
Hiroyuki Masuta;Hun-ok Lim;Tatsuo Motoyoshi;Ken´ichi Koyanagi;Toru Oshima
Author_Institution :
Toyama Prefectural University, Dept. of Intelligent Systems Design Engineering, Imizu, Toyama, Japan
Abstract :
This paper discusses the direct perception for grasping and the action for perception using depth sensor. The previous method should recognize the accurate physical parameters to grasp an unknown object. Hence, we propose a sensation based perception which perceives exclusively the relevant information to the behavior of the robot. To perceive an unknown object, we have proposed the plane detection based approach of the SPD-SE. The SPD-SE may have applicability to robot perception at real time, and has advantages that the point group and properties of an unknown object can be extracted at the same time. The sensation of grasping is explained by inertia tensor and fuzzy inference. The sensation of grasping affords the possibility of action to a robot directly without inference from physical information such as size, posture and shape. As experimental results, we show that the robot can detect a relevant information of a grasping behavior directly. And, the sensation of grasping presents own state and environmental state together by one parameter.
Keywords :
"Robot sensing systems","Three-dimensional displays","Grasping","Robot kinematics","Intelligent robots","Service robots"
Conference_Titel :
Robot and Human Interactive Communication (RO-MAN), 2015 24th IEEE International Symposium on
DOI :
10.1109/ROMAN.2015.7333637