Exploring Negative Emotions to Preserve Social Distance in a Pandemic Emergency

Abstract

In this work, we present a multi-agent robotic system which explores the use of unpleasant emotions triggered by visual, sound and behavioural affordances of autonomous agents to interact with humans for preserving social distance in public spaces in a context of a pandemic emergency. The idea was born in the context of the Covid-19 pandemic, where discomfort and fear have been widely used by governments to preserve social distancing. This work does not implicitly endorse the use of fear to keep order but explores controlled and moderate automated exploitations. On the contrary, it deeply analyses the pros and cons of the ethical use of robots with emotion recognition and triggering capabilities. The system employs a swarm of all-terrain hexapods patrolling a public open space and generally having a discrete and seamless presence. The goal is to preserve the social distance among the public with effective but minimal intervention, limited to anomaly detection. The single agents implement critical tasks: context detection strategies, triggering negative emotions at different degrees of arousal using affordances ranging from appearance and simple proximity or movements to disturbing sounds or explicit voice messages. The whole system exhibits an emerging swarm behaviour where the agents cooperate and coordinate in a distributed way, adapting and reacting to the context. An innovative contribution of this work, more than the application, is the use of unpleasant emotions affordances in an ethical way, to attract user attention and induce the desired behaviour in the emergency. This work also introduces a method for assessment of the emotional level of individuals and groups of people in the context of swarm agents. The system extends the experience of the gAItano hexapod project, an autonomous agent with image detection and planned object relocation capabilities.