Implementación de un juguete inteligente para la enseñanza de niños de 2 a 7 años, usando una técnica de Soft Computing.

Abstract

The present titling work had as objective to implement an intelligent toy for the playful and cognitive interaction for children in pre-operational stage, creating favorable environments for the perception of colors and geometric figures. The playhead device provides users with a graphical interface with several options including stories, songs, games, image gallery and facial recognition (Face ID) to activate the teaching program. The face recognition system is based on the use of cascade classifiers and AdaBoost algorithm to train a neuronal network, implemented as an adaptation of the model proposed by Paul Viola and Michael Jones. Artificial Vision techniques were used to determine the contour and color of 2D objects in controlled spaces and a real-time tone filter to find the values of the spectral bands in RGB format. All this information was processed on a Raspberry Pi minicomputer. Also, a differential locomotion system was implemented that allows the robot to move on flat surfaces together with sensors that emit signals to an Arduino processor to coordinate movements and to plan trajectories. The results obtained from the interaction with children were satisfactory, out of a sample of 192 children, 98.44% liked it. On the other hand, specialists in psychology and education qualified the robot as an innovative didactic resource and that contributes to special education helping children with intellectual deficiency. With the implementation of the intelligent toy it is recommended to deepen the research that allows to contribute to the raising of the cognitive bases.