Feeling Artificial Intelligence. Cognitive Decision-Making Model Borrowed from Living Beings

Feeling AI (FAI) as a new kind of Hybrid AI aimed to improve the autonomy of service provided by robots is discussed. Using as an embedded system, the Hybrid AI based on symbolic and machine learning techniques combined with an “Emotional Shell” and a pre-trained set of different skills is problematic due to the required computer facilities and limited robot autonomy in the physical world. This article proposes, in contrast to the symbolic approach, to set up FAI on the cognitive response-making model borrowed from the simplest living beings that demonstrate the capability of performing all its functions under all conditions in autonomous mode. Based on findings of neurobiology and cognitive science about hydra behavior, the implementation of four cognitive functions, such as perception, drives, emotions, and attention, in the response-making mechanism, is shown. A borrowed model formalized on fuzzy Certainty Factor (CF) and presented as a homogeneous granular knowledge structure of independent response prototypes that are processed by a cognitive response-making engine is discussed. The proposed model implemented in the blueprint of FAI's basic layer is given.