Learning to localize itself-vision-motor association for legged robot
Takahiro Miyashita, Shneider Doi, Minoru Asada, Koh Hosoda · 2003
Presents a method that acquires a vision-motor association for a legged robot as a bottom-up technique for self-localization by action based perception categorization. By the method, the legged robot can move to the position where it can capture the desired view. The validity of the method is shown by a preliminary experiment, and future work is described.