Talking Robotics

Logo


Organizers: Patrícia Alves-Oliveira,
Silvia Tulli, Miguel Vasco
contact us: talkingrobotics at gmail dot com — support us: buymeacoffe

twitter youtube

Speaker

Ilaria is a postdoc at KTH (Royal Institute of Technology), in the division of Robotics, Perception and Learning. She is currently working with Iolanda Leite on socially acceptable robot navigation, and how to communicate robot’s actions and intentions via intuitive communication (e.g. sound and movement).

In her research, Ilaria has been studying how different characteristics of an artificial agent (e.g. voice, emotional expression, human-likeness, behaviour) affect human behaviour, specifically in terms of trust. To this end, she has conducted several experiments using paradigms derived from Psychology and Behavioural Economics. For example, she applied the game-theoretic ‘game of chicken’ to model socially acceptable robot trajectories. She is also interested in preventing potential mismatches between first impressions and actual experience with robots, and she is particularly interested in creating “appropriate” voices for robots.

Ilaria has a background in Speech, Psychology, and Human-Agent Interaction. She obtained a PhD from the University of Plymouth, in the EU-funded CogNovo doctoral training centre, in 2017, with a project investigating vocal predictors of behavioural trust in Human-Agent Interaction. She then spent 2.5 years as a Marie Skłodowska-Curie postdoctoral researcher at Trinity College Dublin, School of Electronic and Electrical Engineering, where she studied how mismatching vocal and facial emotional expressions influence cooperation in Human-Agent Interaction.

Speaker Links: Website - Twitter - Google Scholar - Linkedin


Abstract

Two people walking towards each other in a colliding course is an everyday problem of human-human interaction. In spite of the different environmental and individual factors that might jeopardise successful human trajectories, people are generally skilled at avoiding crashing into each other, even when their goals are apparently at odds (e.g. two people who are both trying to catch a bus at the opposite end of the street). For robots to be able to integrate in these busy social spaces, they will need to understand the goals and motivations of their surrounding pedestrians, and at the same time ensure that their own goals can be accomplished safely.

In this talk, I will first present a series of experiments where people played the “game of chicken” with virtual robots walking towards them. We manipulated the motivation of both participants and robots, to study how people navigate around robots when both parties are trying to accomplish a certain goal. People’s trajectories in this game can be studied from the point of view of proxemics theory, and can be applied to robot’s trajectories to ensure that they are socially acceptable in different contexts. I will also talk about our speculations on how robots could communicate their goals and intentions. Since we are interested in unintentional Human-Robot Interaction, we want this communication to be as intuitive and unobtrusive as possible. I will introduce some ongoing work on robot sound design that could achieve this goal.


Papers covered during the talk