Publikation

Using Doctrines for Human-Robot Collaboration to Guide Ethical Behavior

Geert-Jan Kruijff, Miroslav Janicek

In: Geert-Jan Kruijff , Fiora Pirri , Panagiotis Papadakis (Hrsg.). AAAI 2011 Fall Symposium Robot-human team-work in dynamic adverse environments. AAAI Fall Symposium November 4-6 Arlington Virginia United States AAAI 11/2011.

Abstrakt

We consider human-robot teams as a socio- technological complex. Humans, robots, multi-modal interfaces are all tightly interwoven into a complex system. It is this system as a whole that acts, that brings about situation awareness, that achieves goals in reality. It is this system that ultimately needs to behave according to a set of rules that "we" consider to be "good." Our goal in this paper is to see how we could build up a formal model of human-robot teaming, which would make it possible for a robot 1. to use that model to track the social dynamics of the team; 2. within that context, to contribute to a team behavior that is "good,"; (and not, through action or inaction, move towards "bad"; behavior); and, 3. to make its actions fully accountable.

Projekte

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence