Emotionally motivated reinforcement learning based controller

There have been several attempts to model emotions in autonomous agents and robotics. The use of emotions in conjunction with reinforcement learning in particular has attracted attention since both notions are borrowed analogies from psychology. The work presented here is an approach to robot control based on modeling emotions within reinforcement learning algorithm. The main contribution of this paper is the use of fuzzy cognitive maps (FCM) to facilitate the modeling of emotions and inferencing for action selection. This approach does not use feeling estimation; instead a direct link between sensory data and emotions is used for emotional estimation. An emotion based reinforcement learning algorithm is proposed for action selection in robotic control.