idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
The end of the year saw the launch of the TEMA (Trusted Extremely Precise Mapping and Prediction for Emergency Management) consortium project, which is funded with more than 11 million euros as part of the EU's Horizon Europe initiative. Within the TEMA consortium, 20 European partner institutions from industry, research and politics are striving to improve the management of natural disasters in real time. Using heterogeneous data to create meaningful area maps, researchers hope to predict the course of extreme natural phenomena more accurately in the future and visualize them using augmented reality (AR). TEMA will run until November 2026 and is led by Aristotle University in Thessaloniki.
In light of current climate developments, transnational extreme natural events such as floods and forest fires are going to increase significantly in the future. Precise Natural Disaster Management (NMD) is crucial in this context to prevent them from becoming life-threatening. Key interpretation and communication processes, however, still hold considerable potential for improvement.
This is where the partners of the TEMA consortium come in by developing technologies for the automated generation of meaningful maps of endangered or affected areas. To this end, researchers from Fraunhofer HHI are serving as task lead of the "Explainable and robust analytics" group, combining data from a wide variety of sources. These sources include autonomous devices and smart sensors, satellites and official weather services as well as social media. The advantage of this heterogeneous data is the ability to create the most comprehensive, up-to-date and multi-faceted picture of the current hazard situation. The broad research and data collection area of TEMA (Germany, Finland, Italy and Greece) allows for a powerful investigation of a variety of scenarios over a longer period of time. The Fraunhofer HHI research team analyzes the collected data and uses concept-based XAI methods to produce accurate and explainable replication and modeling of real NDM events, taking into account quality standards and evaluation criteria for XAI methods.
Based on the area maps and the resulting analyses, the researchers aim to develop an AI-supported NDM platform that will be validated and demonstrated specifically in dealing with forest fires and floods. Ultimately, human decision-makers in hazardous situations will be able to access the TEMA platform via an intuitive AR interface, which in turn will provide predictions on the progression of extreme weather phenomena and up-to-date response recommendations. This interface will enable users to better understand complex NDM phenomena and easily evaluate alternative response strategies through intelligent selection of platform parameters combined with novel combinations of heterogeneous (visual and non-visual) data and AR.
Dr. Sebastian Lapuschkin, leader of the research group “Explainable Artificial Intelligence” and project manager of TEMA at Fraunhofer HHI states: "Forest fires and floods do not stop at national borders. Natural disaster management poses a key European challenge for our future that needs to be solved on a European level. We are pleased to be able to contribute our expertise in the field of Explainable and Robust AI, especially in the context of NDM." Fraunhofer HHI already conducts research in a leading position of the ITU/WHO focus group "AI for Natural Disaster Management" and the BMWK project DAKI-FWS on the use of AI for the prevention and management of natural disasters.
In addition to a broad database, the cross-country collaboration will ensure significantly facilitated interoperability of the NDM platform between EU countries and regions. Furthermore, it provides the basis for offering scalable and efficient NDM Analytics-as-a-Service (NDM-AaaS). Potential end users of the TEMA platform are mainly civil protection authorities but also first responders.
In addition to Aristotle University and Fraunhofer HHI, the collaborative partners in TEMA are the German Aerospace Center, Engineering Ingegneria Informatica S.P.A. (Italy), Atos IT Solutions and Services Iberia SL (Spain), Atos Spain S. A. (Spain), Universidad de Sevilla (Spain), Tecnosylva SL (Spain), Northdocks GmbH (Paris-Lodron), University of Salzburg (Austria), The Lisbon council for economic competitiveness and social renewal ASBL (Belgium), Latitudo 40 S. R.L. (Italy), Nelen & Schuurmans Technology (Netherlands), University of Messina (Italy), Kajaanin Ammattikorkeakoulu Oy (Finland), Kajaanin Kaupunki (Finland), Center for security studies / Kentro Meleton (Greece), Municipality of Mantoudi-Limni-Agia Anna (Greece), Autonomous Region of Sardinia (Italy) and the Bavarian Red Cross.
Dr. rer. nat. Sebastian Lapuschkin
Head of Explainable Artificial Intelligence Group
tel. +49 30 31002-371
e-Mail: sebastian.lapuschkin@hhi.fraunhofer.de
Martina Müller
Head of Corporate Communications
tel. +49 30 31002-242
e-Mail: martina.mueller@hhi.fraunhofer.de
TEMA research team at project kick-off in Thessaloniki
© Ioanna Koroni
Merkmale dieser Pressemitteilung:
Journalisten, Studierende, Wirtschaftsvertreter, Wissenschaftler
Elektrotechnik, Geowissenschaften, Informationstechnik, Meer / Klima, Umwelt / Ökologie
überregional
Forschungsprojekte, Kooperationen
Englisch
TEMA research team at project kick-off in Thessaloniki
© Ioanna Koroni
Sie können Suchbegriffe mit und, oder und / oder nicht verknüpfen, z. B. Philo nicht logie.
Verknüpfungen können Sie mit Klammern voneinander trennen, z. B. (Philo nicht logie) oder (Psycho und logie).
Zusammenhängende Worte werden als Wortgruppe gesucht, wenn Sie sie in Anführungsstriche setzen, z. B. „Bundesrepublik Deutschland“.
Die Erweiterte Suche können Sie auch nutzen, ohne Suchbegriffe einzugeben. Sie orientiert sich dann an den Kriterien, die Sie ausgewählt haben (z. B. nach dem Land oder dem Sachgebiet).
Haben Sie in einer Kategorie kein Kriterium ausgewählt, wird die gesamte Kategorie durchsucht (z.B. alle Sachgebiete oder alle Länder).
