Robot feeling formation based on image features

Feeling and emotion are important to human being during his/her learning process, also valuable to be adopted into intelligent machines. This research presents a system which forms and expresses feelings of a robot. The vision information of robot is used and the environment features are categorized by a hierarchical SOM (Self-Organization Map). The proposed SOM here combines a feature map, an action map and a feeling map together, and the action policy of robot is modified by reinforcement learning rule, feeling degree is complicated by a dequantization algorithm. The experiments through computer simulation and an autonomous robot in the real environment showed robot´s feelings formation process and its emotional actions.