Title: AI Techniques for Resilient Human-Human-Robot Interaction
Duration: 18 months
Starting Date: 1/12/2022
PROJECT DATA
TASK SYNOPSIS
The aim is to investigate how to provide an embodied artificial Intelligence with an awareness level to enable resilient interaction with humans. This requires identifying the necessary cognitive capabilities in terms of perception of social and non-verbal signals and intention recognition, and to develop machine learning techniques that rely on noisy data from real-world interactions. Indeed, while datasets used in typical HRI studies are recorded assuming a face-to-face interaction, into-the-wild ones pose significant challenges to be addressed to obtain the proper situational awareness. We aim to avoid humans misusing robots by adopting Explainable AI approaches to provide explanations about robots’ behaviors and actions coming from possible erroneous perceptions. We want to establish the level of Theory of Mind needed by robots to achieve transparent and trustworthy interactions to foster human-robot trust in long-lasting relationships.
PARTNERS
– Università degli Studi di Genova
– Università degli Studi di Napoli Federico II
– Consiglio Nazionale delle Ricerche
FUNDING
Italian Ministry for Universities and Research (MUR) under the grant PNRR FAIR (MUR: PE0000013).