2A1-L06 Oculomotor Control using Images based on Compound Control
Y. Ichihara, Yusei TSUBOI, Shingo Shimoda, Takefumi Asahi, Hidenori Kimura, H. Fujimoto · The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec) · 2007
Compound control is the bio-mimetic control and learning method. In order to learn based on compound control, feedback ioop in environment is required. Image is the most important environment sensor for living beings. Image processing and behavior control based on compound control is the important to generate various kinds of behaviors. This paper presents an initial network, which is one of the most important element compound control, for image processing based on compound control. This initial network is composed of three layers. Calculation elements corresponding to a retina are distributed to 2 dimensional array. OKR, one of the ocular motions, is realized by using the proposal initial network.