Towards Verbal Explanations by Collaborating Robot TeamsShow others and affiliations
2019 (English)In: International Conference on Social Robotics (ICSR’19), Workshop Quality of Interaction in Socially Assistive Robots, Madrid, Spain, November 26-29, 2019, 2019Conference paper, Oral presentation only (Refereed)
Abstract [en]
In this paper, we describe ongoing work on understandable teams of robots collaborating to solve a common task while communicating their current, suggested or planned actions in natural language to human bystanders. We propose an algorithm for generating optimal sequences of partitions to be used by the robots to generate natural language utterances. These optimal sequences of partitions maintain the conversational principle of informativeness proposed by Grice.We further propose a method to incorporate policies, such as a uniform distribution of workload.
Place, publisher, year, edition, pages
2019.
Keywords [en]
natural language generation, human-robot interaction, conversational principles, Gricean maxims, explainable AI, understandable robot teams
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:umu:diva-166090OAI: oai:DiVA.org:umu-166090DiVA, id: diva2:1377072
Conference
International Conference on Social Robotics (ICSR’19), Workshop Quality of Interaction in Socially Assistive Robots, Madrid, Spain, November 26-29, 2019.
2019-12-102019-12-102019-12-12Bibliographically approved