Page 251 - Mechatronics for Safety, Security and Dependability in a New Era

Ch48-I044963.fm Page 235 Tuesday, August 1, 2006 4:04 PM
Page 235
Tuesday, August
1, 2006
Ch48-I044963.fm
4:04 PM
235
235
GENERATED IMAGE FEATURE BASED SELECTIVE
ATTENTION MECHANISM BY VISUO-MOTOR LEARNING

1
Department of Adaptive Machine Systems, Graduate School of Engineering, Osaka University,
2-1 Yamada-oka, Suita, Osaka 565-0871 Japan

ABSTRACT
Visual attention is an essential mechanism of an intelligent robot. Existing research typically specifies
in advance the attention control scheme required for a given robot to perform a specific task. However,
a robot should be able to adapt its own attention control to varied tasks. In our previous work, we
proposed a method of generating a filter to extract an image feature by visuo-motor learning. The
generated image feature extractor is considered to be generalized knowledge to accomplish a task of a
certain class. We propose an attention mechanism, by which the robot selects the generated feature
extractors based on its task-oriented criterion.

KEYWORDS

Mobile Robot, Selective attention, Image feature generation, Image feature selection, Task-oriented

INTRODUCTION

Attention control is an essential mechanism for an intelligent robot to avoid processing enormous
amounts of data. It is a data reduction process to facilitate decision making. With regard to visual
attention control, it involves selection of focus, image features, and so on. Existing research typically
specifies in advance the attention control scheme required for a given robot to perform a specific task.
However, a robot should be able to adapt its own attention control to varied tasks and environments.

We have focused on visual attention control related to a robot's actions to accomplish a given task and
proposed a method in which a robot generates an image feature extractor (i.e., image filter) which is
necessary for the selection of actions through visuo-motor map learning (Minato & Asada, 2003). The
robot's learning depends on the experience gathered while performing a task. In this method, the robot
uses only one feature extractor for a given task. For more complex tasks, however, multiple feature
extractors are necessary to accomplish the tasks and a method of selecting them should be addressed.

Some research has focused on a method of feature selection based on task-relevant criteria. McCallum
(1996) proposed a method in which a robot learns not only its action but feature selection using