TEACHBOT: Intelligent Child-Robot Interaction System for designing and implementing edutainment scenarios with emphasis on visual information. In this project, a robust intelligent robotic system will be developed for the design and execution of edutainment scenarios, focusing on child action and emotion recognition. The three main research directions can be summarized as followed:

  1. development of novel, real-time methods for recognizing children’s actions while they are interacting with a robot, by employing computer vision and machine learning algorithms with small datasets
  2. development of an emotional state monitoring unit, which will be able to continuously monitor the affective state of the child through its facial expressions and body posture. This unit will allow the system to adapt its behavior, so that it is congruent with the child’s emotions.
  3. integration of the above technologies, along with existing technologies of speech recognition and synthesis, into a robust intelligent robotic system that will give the user-teacher the ability to create edutainment scenarios.
Selection 002
SoftGrip project