Developing ePartners for human-robot teams in space based on ontologies and formal abstraction hierarchies Online publication date: Mon, 30-Oct-2017
by Tibor Bosse; Leo Breebaart; Jurriaan Van Diggelen; Mark A. Neerincx; Joaquim Rosa; Nanja J.J.M. Smets
International Journal of Agent-Oriented Software Engineering (IJAOSE), Vol. 5, No. 4, 2017
Abstract: Manned space missions are typically performed by teams composed of humans as well as technical systems and are situated in complex, dynamic and safety-critical domains. Intelligent electronic partners (ePartners) can play an important role here to support human-robot teams in their collaborative problem solving process when things do not go as planned. To engage in effective team collaboration, ePartners, humans and robots must align their communication at the right level of abstraction. In this paper, an approach is put forward to represent the functionality of human-robot teams in a formal manner using abstraction hierarchies. In this way, formal relations between domain knowledge at the system, functional and mission-oriented levels are established and reasoning rules are used to navigate through these relations. As a consequence, ePartners are equipped with the ability to reason about the status of a mission, propose solutions in non-nominal situations and provide explanations for the proposed solutions. The approach has been implemented within a mobile application on a tablet that can be used to support astronaut-robot teams during space missions. The application has been evaluated during an experiment at the European Space Research and Technology Centre (ESTEC) in the context of a Mars mission.
Online publication date: Mon, 30-Oct-2017
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Agent-Oriented Software Engineering (IJAOSE):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email email@example.com