Analysis of robotic arm´s behavior using Self Organizing Map combined with consciousness-based architecture module

Our research has been focused on developing human-robot interactions. We attempted to give a robotic arm has a consciousness resembling that of a human or an animal, so we have improved a hierarchical structure model, which defines the relationship between the consciousness field and the behavior module in the robotic arm. This model is called Consciousness-Based Architecture (CBA) and can be used to select the action of a robot. Indeed, the robot is programmed to select its own action by evaluating the changing situation based on its motivation value and image recognition. We have investigated the application of brain-inspired technology such as a Self Organizing Map (SOM) into the robotic arm. In this paper, we introduce a prototype function for autonomous action selection and show a classification of the robotic arm´s behavior using SOM combined with a CBA module.