Hideaki Takeda's Publication
- K. Terada, T. Nakamura, H. Takeda and T. Nishida:
A cognitive robot architecture based on tactile and visual information,
Advanced Robotics, Vol. 13, No. 8, pp. 767–778 (2000).
In this paper, we propose an architecture for a cognitive robot
based on tactile and visual information. Visual information contains various
features such as location and area of each colored region. Most of these
features are irrelevant for object recognition to achieve the given task. In
the architecture, tactile information plays a key role in selection of visual
features and discritization of selected features. In order to find
appropriate visual features we use correlation coefficient between values of
features and action series. Then ChiMerge algorithm is employed to discritize
the value of the selected feature into a small number of intervals.
Consequently, quantization of a state space for accomplishing the given task
is achieved. By using this state space to reinforcement learning algorithm,
an appropriate behavior to the given task is acquired. To show validity of
our method, we show an experimental result of computer simulation. Key words:
Cognitive Robot, Vision, Tactile Sensor, Feature Selection, RoboCup
Hideaki Takeda (National Institute of Informatics)