Publikation

Experiences with USAR mobile interfaces: The need for persistent geo-localized information

Benoit Larochelle, Geert-Jan Kruijff, NJJM Smets, J van Diggelen

In: 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (Hrsg.). IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS-13) November 3-8 Tokyo Japan Seiten 5333-5338 IEEE/RSJ 2013.

Abstrakt

Urban search and rescue (USAR) missions are unique and unpredictable. Communication and coordination is difficult, with high-level actors (e.g. mission commander) lacking local terrain knowledge while low-level actors (e.g. robot operators, in-field rescuers) lack global situation awareness (SA). In-field actors have high mobility and a direct view of the field, but so far they could communicate this SA almost exclusively through radio. Unfortunately, words are imprecise and unorganized, and thus not easily analyzable and retrievable. As part of the NIFTi1 project, we developed mobile applications to help in-field actors share their SA with the rescue team. We also performed high-fidelity USAR simulations and missions at fire fighting training sites and disaster areas. We discovered a need for persistent geo-localized information and propose a novel system architecture that integrates pictures taken from robots and from in-field rescuers into the existing systems at the command post.

Projekte

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence