TY - JOUR
T1 - Tool-body assimilation of humanoid robot using a neurodynamical system
AU - Nishide, Shun
AU - Tani, Jun
AU - Takahashi, Toru
AU - Okuno, Hiroshi G.
AU - Ogata, Tetsuya
N1 - Funding Information:
Manuscript received October 12, 2010; revised February 25, 2011, June 05, 2011, and September 16, 2011; accepted October 09, 2011. Date of publication December 06, 2011; date of current version June 08, 2012. This work was supported by JST PRESTO (Information Environment and Humans), Grant-in-Aid for Creative Scientific Research (19GS0208), and Grant-in-Aid for Scientific Research (B) (21300076).
PY - 2012
Y1 - 2012
N2 - Researches in the brain science field have uncovered the human capability to use tools as if they are part of the human bodies (known as tool-body assimilation) through trial and experience. This paper presents a method to apply a robot's active sensing experience to create the tool-body assimilation model. The model is composed of a feature extraction module, dynamics learning module, and a tool-body assimilation module. Self-organizing map (SOM) is used for the feature extraction module to extract object features from raw images. Multiple time-scales recurrent neural network (MTRNN) is used as the dynamics learning module. Parametric bias (PB) nodes are attached to the weights of MTRNN as second-order network to modulate the behavior of MTRNN based on the properties of the tool. The generalization capability of neural networks provide the model the ability to deal with unknown tools. Experiments were conducted with the humanoid robot HRP-2 using no tool, I-shaped, T-shaped, and L-shaped tools. The distribution of PB values have shown that the model has learned that the robot's dynamic properties change when holding a tool. Motion generation experiments show that the tool-body assimilation model is capable of applying to unknown tools to generate goal-oriented motions.
AB - Researches in the brain science field have uncovered the human capability to use tools as if they are part of the human bodies (known as tool-body assimilation) through trial and experience. This paper presents a method to apply a robot's active sensing experience to create the tool-body assimilation model. The model is composed of a feature extraction module, dynamics learning module, and a tool-body assimilation module. Self-organizing map (SOM) is used for the feature extraction module to extract object features from raw images. Multiple time-scales recurrent neural network (MTRNN) is used as the dynamics learning module. Parametric bias (PB) nodes are attached to the weights of MTRNN as second-order network to modulate the behavior of MTRNN based on the properties of the tool. The generalization capability of neural networks provide the model the ability to deal with unknown tools. Experiments were conducted with the humanoid robot HRP-2 using no tool, I-shaped, T-shaped, and L-shaped tools. The distribution of PB values have shown that the model has learned that the robot's dynamic properties change when holding a tool. Motion generation experiments show that the tool-body assimilation model is capable of applying to unknown tools to generate goal-oriented motions.
KW - Active sensing
KW - humanoid robots
KW - recurrent neural network
KW - tool-body assimilation
UR - http://www.scopus.com/inward/record.url?scp=84862334332&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84862334332&partnerID=8YFLogxK
U2 - 10.1109/TAMD.2011.2177660
DO - 10.1109/TAMD.2011.2177660
M3 - Article
AN - SCOPUS:84862334332
SN - 1943-0604
VL - 4
SP - 139
EP - 149
JO - IEEE Transactions on Autonomous Mental Development
JF - IEEE Transactions on Autonomous Mental Development
IS - 2
M1 - 6095595
ER -