Research Article
Tool-Body Assimilation Model Based on Body Babbling and
Neurodynamical System
Kuniyuki Takahashi,
1
Tetsuya Ogata,
2
Hadi Tjandra,
1
Yuki Yamaguchi,
3
and Shigeki Sugano
1
1
Graduate School of Creative Science and Engineering, Waseda University, Tokyo 1698555, Japan
2
Graduate School of Fundamental Science and Engineering, Waseda University, Tokyo 1698555, Japan
3
Graduate School of Informatics, Kyoto University, Kyoto 6068501, Japan
Correspondence should be addressed to Kuniyuki Takahashi; takahashi@sugano.mech.waseda.ac.jp
Received 7 March 2014; Revised 15 June 2014; Accepted 15 June 2014
Academic Editor: Yi Chen
Copyright © 2015 Kuniyuki Takahashi et al. Tis is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
We propose the new method of tool use with a tool-body assimilation model based on body babbling and a neurodynamical system
for robots to use tools. Almost all existing studies for robots to use tools require predetermined motions and tool features; the motion
patterns are limited and the robots cannot use novel tools. Other studies fully search for all available parameters for novel tools, but
this leads to massive amounts of calculations. To solve these problems, we took the following approach: we used a humanoid robot
model to generate random motions based on human body babbling. Tese rich motion experiences were used to train recurrent and
deep neural networks for modeling a body image. Tool features were self-organized in parametric bias, modulating the body image
according to the tool in use. Finally, we designed a neural network for the robot to generate motion only from the target image.
Experiments were conducted with multiple tools for manipulating a cylindrical target object. Te results show that the tool-body
assimilation model is capable of motion generation.
1. Introduction
Humans are capable of expanding their ability by using
tools. Robots are expected to become more useful to society
through the use of tools. With the development of robotics
technology, robots have become very complex, with increas-
ing numbers of applicable sensors and degrees of freedom.
Terefore, complicated calculations are required to build
conventional robot tool use models. Modeling robot tool use
based on human cognitive development has been proposed
as an approach to mitigate this problem [1]. Among the
many factors of human cognitive development, tool-body
assimilation, studied in the feld of neuropsychology, has
begun to gather attention [2]. Tool-body assimilation occurs
when humans use a tool and treat it as an expansion of their
own bodies. Iriki et al. recorded neurons called “bimodal
neurons” before and afer monkeys were trained to use a
tool. Bimodal neurons respond both to tactile stimulation on
the hand and to visual stimulation. Trough tool use training,
the visual receptive feld of bimodal neurons expands from
the monkey’s hand to the surroundings of the grasping tool.
Tis result shows that tool-body assimilation occurs at the
neuron level. We aim to achieve robot tool use by this
approach; by modeling tool-body assimilation, it is possible
to alter the behavior of a posttrained body model by adding
new neurons and expressing a “body model that is using a
tool.”
Tool use with tool-body assimilation is also gathering
attention in the feld of robotics. Nabeshima et al. [3] used
visual and touch stimuli to connect bodily and sensory
information. Afer training the relationships between visual
and touch stimuli, dynamic touch was performed to predict
the inertial parameters of the tool. Te resulting simulation
model allowed the robot to perform a pulling task with a
target object located in an invisible area. However, the inertial
parameters used as tool features were determined in advance,
and the model was incapable of adapting to nonrigid bodies.
Hikita et al. [4] treated tools as an expansion of the robot’s
Hindawi Publishing Corporation
Mathematical Problems in Engineering
Volume 2015, Article ID 837540, 15 pages
http://dx.doi.org/10.1155/2015/837540