A cognitive robot architecture based on tactile and visual information

In this paper, we propose an architecture for a cognitive robot based on ta ctile and visual information. Visual information contains various features such as location and area of various colored regions. Most of these feature s are irrelevant for object recognition to achieve the given task. In the a rchitecture, tactile information plays a key role in the selection of visua l features and discritization of selected features. In order to find approp riate visual features we use a correlation coefficient between the values o f the features and action series. Then the ChiMerge algorithm is employed t o discritize the value of the selected feature into a small number of inter vals. Consequently, quantization of a state space to accomplish the given t ask is achieved. An appropriate behavior to the given task is acquired by u sing this state space with reinforcement learning algorithm. We give experi mental results of computer simulation to show the validity of our method.