Robots Meet Children, Development of Semi-Autonomous Control Systems for Children-Robot Interaction in the Wild

Abstract

In the study of social and service robots, it is necessary to obtain more valuable quantitative and qualitative data of humans interacting with these robots in natural and unstructured environments. This data can be used to better design applications and robots with higher social impact. In this paper, we show how the integration of perceptual robot control and decision-making architecture can enable the creation of semi-autonomous robots for Children-Robot Interaction (CRI) “in the wild”. Unlike most previous works in this field, our proposal mainly uses distributed, open, cross-platform and reactive software components. In order to enhance scalability and modularity, the modeling of both the autonomous and remote-controlled robot behaviors are performed using a Behavior Tree (BT) approach. We experimentally validate the proposed system with a pilot study in a public and highly noisy environment outside the laboratory with a large group of children interacting in different activities (two stories, a dance, a game, and autonomous interaction). Experimental settings and preliminary insights are also described.