Convolutional Neural Network Applied to the Gesticulation Control of an Interactive Social Robot with Humanoid Aspect

This document presents the enforcement of a facial gesture recognition system through applying a convolutional neural network algorithm for gesticulation of an interactive social robot with humanoid appearance, which was designed in order to accomplish the thematic proposed. Furthermore, it is incorporated into it a hearing communication system for Human-Robot interaction throughout the use of visemes, by coordinating the robot's mouth movement with the processed audio of the text converted to the robot's voice (text to speech). The precision achieved by the convolutional neural network incorporated in the social-interactive robot is 61%, while the synchronization system between the robot's mouth and the robot's audio-voice differs from 0.1 s. In this manner, it is pretended to endow mechanisms social robots for a more naturally interaction with people, thus facilitating the appliance of them in the fields of children's teaching-learning, medical therapies and as entertainment means.