Ron Petrick's Webpages

Spotlight talk

Planning for Social Interaction in a Robot Bartender Domain, R. Petrick, slides from the ICAPS Best Paper Spotlight talk presented at the AAAI Conference on Artifical Intelligence (AAAI-13), Bellevue, Washington, USA, 2013-07-18.

[ slides ]

Abstract

A robot coexisting with humans must not only be able to perform physical tasks, but must also be able to interact with humans in a socially appropriate manner. In many social settings, this involves the use of social signals like gaze, facial expression, and language. We describe an application of planning to task-based social interaction using a robot that must interact with multiple human agents in a simple bartending domain. Social states are inferred from low-level sensors, using vision and speech as input modalities. The knowledge-level PKS planner is used to construct plans with task, dialogue, and social actions, as an alternative to current mainstream approaches of interaction management. The resulting system has been tested with human users in a series of drink ordering scenarios.

This talk is based on the paper Planning for Social Interaction in a Robot Bartender Domain from ICAPS 2013, which was joint work with M. E. Foster.