Silvia Rossi

Silvia Rossi is a Full Professor of Computer Science at the Department of Electrical Engineering and Information Technologies, University of Naples Federico II, where she serves as the Scientific Director of the PRISCA Lab (Projects of Intelligent Robotics and Advanced Cognitive Systems). She holds an M.Sc. in Physics (2001) from the University of Naples Federico II and a Ph.D. in Information and Communication Technologies (2006) from the University of Trento. Prof. Rossi has led and contributed to numerous EU and international research projects. She is currently the Principal Investigator and Coordinator of major initiatives including HORIZON-MSCA-2023-DN SWEET (Social aWareness for sErvicE roboTs), and the national PRIN project ADVISOR (ADaptiVe legIble robotS for trustwORthy health coaching). She has chaired the IEEE RO-MAN Conferences and actively serves on program committees for leading conferences in Human-Robot Interaction and Artificial Intelligence. Her research focuses on Socially Assistive Robotics, Human-Robot Interaction, Cognitive Architectures, and User Profiling, with particular emphasis on computational methods enabling autonomous agents to adapt their behavior for effective and trustworthy interaction with humans. Prof. Rossi has authored over 200 publications in international journals, books, and conference proceedings.

Talk Title: Understanding and Being Understood: Cognitive Robotics Challenges from an HRI Point of View

Abstract:

Meaningful human–robot interaction depends on a two-way process of understanding. Robots must perceive and interpret human goals, emotions, and intentions in ways that are sensitive to individual differences and situational context – an ability that parallels aspects of the Theory of Mind in humans. Such context-aware perception depends on the integration of multimodal sensing, adaptive inference, and representations of social and environmental context – capabilities that remain at the frontier of cognitive robotics. By tailoring their internal models and responses to who people are, what they are doing, and how they are feeling, robots can enable more fluid, trustworthy, and personalized interactions. However, understanding must be reciprocal. Humans also need to make sense of robot actions, anticipate their behavior, and feel confident in the robot’s competence and intentions. Designing robot behaviors that are legible, predictable, and contextually appropriate is therefore equally essential. This demands an understanding of how humans form mental models of artificial agents, attribute agency and emotion, and anticipate behavior – core topics that bridge robotics and cognitive science. This talk will explore these two complementary dimensions – personalized, context-aware human modeling and socially interpretable robot behavior – as foundations for cognitive systems that can both understand and be understood. The discussion will highlight current research challenges, emerging methodologies, and the implications of mutual understanding for the design of safe, adaptive, and socially integrated robotic agents.