TY - GEN
T1 - Tool-body assimilation model using a neuro-dynamical system for acquiring representation of tool function and motion
AU - Takahshi, Kuniyuki
AU - Ogata, Tetsuya
AU - Tjandra, Hadi
AU - Yamaguchi, Yuki
AU - Suga, Yuki
AU - Sugano, Shigeki
PY - 2014
Y1 - 2014
N2 - In this paper, we propose a tool-body assimilation model that implements a multiple time-scales recurrent neural network (MTRNN). Our model allows a robot to acquire the representation of a tool function and the required motion without having any prior knowledge of the tool. It is composed of five modules: image feature extraction, body model, tool dynamics feature, tool recognition, and motion recognition. Self-organizing maps (SOM) are used for image feature extraction from raw images. The MTRNN is used for body model learning. Parametric bias (PB) nodes are used to learn tool dynamic features. The PB nodes are attached to the neurons of the MTRNN to modulate the body model. A hierarchical neural network (HNN) is implemented for tool and motion recognition. Experiments were conducted using OpenHRP3, a robotics simulator, with multiple tools. The results show that the tool-body assimilation model is capable of recognizing tools, including those having an unlearned shape, and acquires the required motions accordingly.
AB - In this paper, we propose a tool-body assimilation model that implements a multiple time-scales recurrent neural network (MTRNN). Our model allows a robot to acquire the representation of a tool function and the required motion without having any prior knowledge of the tool. It is composed of five modules: image feature extraction, body model, tool dynamics feature, tool recognition, and motion recognition. Self-organizing maps (SOM) are used for image feature extraction from raw images. The MTRNN is used for body model learning. Parametric bias (PB) nodes are used to learn tool dynamic features. The PB nodes are attached to the neurons of the MTRNN to modulate the body model. A hierarchical neural network (HNN) is implemented for tool and motion recognition. Experiments were conducted using OpenHRP3, a robotics simulator, with multiple tools. The results show that the tool-body assimilation model is capable of recognizing tools, including those having an unlearned shape, and acquires the required motions accordingly.
UR - http://www.scopus.com/inward/record.url?scp=84906663404&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84906663404&partnerID=8YFLogxK
U2 - 10.1109/AIM.2014.6878254
DO - 10.1109/AIM.2014.6878254
M3 - Conference contribution
AN - SCOPUS:84906663404
SN - 9781479957361
T3 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM
SP - 1255
EP - 1260
BT - AIM 2014 - IEEE/ASME International Conference on Advanced Intelligent Mechatronics
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2014 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM 2014
Y2 - 8 July 2014 through 11 July 2014
ER -