École Nationale Supérieure de
Techniques Avancées

Abstract

The goal of the project was to create a coat-hanger robot able to interact with people around it and particularly to follow them around while asking for their belongings. The person detection is made possible by an Artificial Intelligence algorithm which analyses the images recorded by the robot’s camera. According to what is perceived of the environment, the software choses the action that is the most adequate in each case. An important part of the work was to choose these actions and to implement them in order to give the robot the most natural behavior possible. For example, it is able to mimic feelings such as anger, excitement or sadness through sounds and movements. After the interaction, the robot returns to its initial position on its own. Keywords: Interactive objects, Person detection, C++, Machine Learn-ing, Neural Networks