idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
The European Research Council (ERC) is providing EUR 1.65 million in funding for his research project on autonomous systems that adapt during operation to cope with complex tasks and changing environments. The Starting Grant is considered a highly prestigious mark of excellence for early-career researchers in Europe.
Prof. Manfred Bischoff, Vice Rector for Research and Sustainable Development at the University of Stuttgart, congratulates the junior professor on his success: “The European Research Council is recognizing Andrea Iannelli as an outstanding early-career researcher whose work lays essential foundations for future technologies in areas highly relevant to society such as health and mobility.”
Iannelli, who conducts research at the Institute for Systems Theory and Automatic Control (IST) of the University of Stuttgart, plans to contribute to the theoretical foundations of autonomous systems through his ERC Starting Grant project “Feedback ADaptive mechanisms for online Optimization and Control (FADOC).” The fundamental research project will receive EUR 1.65 million in funding over five years.
Autonomous systems in dynamic environments
A mobility platform has to reroute vehicles when demand suddenly shifts or a road closes; a medical device like, for example, a hearing aid has to continuously fine-tune itself to different acoustic environments or to a patient suddenly feeling tired or overwhelmed. Autonomous technologies are expected to operate in environments that are changing and whose future behaviours is difficult to predict exactly. They must cope with unforeseen events, incomplete information and the presence of humans, whose needs and behaviour are difficult to foresee.
“Today's autonomous systems are remarkably capable, but they still struggle when the environment where they operate changes unexpectedly and its future evolution is uncertain,” says Andrea Iannelli. This is because they are very often based on a simplifying separation principle: First, data is collected and based on this, models are estimated and controllers are designed. Then, these pre-computed algorithms are deployed on the system. This is exactly where FADOC comes in: “Our goal is to develop autonomous systems that simultaneously learn, take decisions, and act, using the information gathered during operation to immediately adapt their behavior whenever this is required.”
A fundamentally new framework
To this end, the project proposes a fundamentally new scientific framework that brings together control theory, optimization, and statistical learning. The goal is to develop a new class of adaptive systems consisting of three components that are in constant communication with one another: fast algorithms that continuously solve an evolving optimization problem; learning routines that convert raw sensor data into reliable estimates of the unknown quantities together with their uncertainty; and control strategies that combine both components while simultaneously providing mathematical guarantees that the overall system behaves in a safe and robust manner. The last point is particularly ambitious, because improving performance with adaptation and achieving rigorous guarantees are still a trade-off in autonomous systems today. “FADOC aims to achieve both at the same time – no existing framework can do that right now,” says Iannelli.
FADOC lays theoretical foundations that could advance many socially relevant technologies. For example, autonomous coordination of public transport vehicles, which is considered a promising lever for reducing climate-damaging greenhouse gases. Moreover, the bottlenecks in the healthcare system call for medical technologies that do not require time-consuming manual adjustments but can automatically and safely adapt to individual patients in real time. “FADOC is a basic research project, however its impact on modern compelling applications could be far-reaching”, says Iannelli.
Personal information
Jun.-Prof. Dr. Andrea Iannelli studied aerospace engineering in Pisa, Italy, and earned his doctoral degree in control and dynamic systems at the University of Bristol in the United Kingdom. Following a postdoctoral research position at ETH Zurich, he was appointed junior professor at the University of Stuttgart in November 2022. He currently leads the Trustworthy Autonomy for Smart Adaptive Systems (TASAS) group at the Institute for Systems Theory and Automatic Control (IST). In addition, Iannelli is a participating researcher in the “Data-Integrated Simulation Science” (SimTech) Cluster of Excellence, a fellow at the Stuttgart Center for Simulation Science, and a faculty member of the International Max Planck Research School for Intelligent Systems.
About the ERC Starting Grant
With its Starting Grants, the European Research Council recognizes early-career researchers for ambitious and groundbreaking projects in fundamental research. The ERC will provide EUR 705 million in funding to 421 early-career researchers. Of these, 84 grants will go to German researchers, and 89 ERC Starting Grant projects will be conducted at German universities or extra-university research institutions. The grant recipients come from more than 50 countries.
Jun.-Prof. Andrea Iannelli, University of Stuttgart, Institute for Systems Theory and Automatic Control (IST), tel. +49 711 685 67730, email: andrea.iannelli@ist.uni-stuttgart.de
https://www.uni-stuttgart.de/en/university/news/all/ERC-Starting-Grant-for-contr...
https://www.ist.uni-stuttgart.de/
https://www.uni-stuttgart.de/en/research/profile/simulation-science/
With his project FADOC, Jun.-Prof. Andrea Iannelli aims to improve future autonomous systems.
Quelle: Uli Regenscheit
Copyright: University of Stuttgart
Merkmale dieser Pressemitteilung:
Journalisten, Wirtschaftsvertreter, Wissenschaftler, jedermann
Elektrotechnik, Informationstechnik, Maschinenbau
überregional
Forschungsprojekte, Wettbewerbe / Auszeichnungen
Englisch
With his project FADOC, Jun.-Prof. Andrea Iannelli aims to improve future autonomous systems.
Quelle: Uli Regenscheit
Copyright: University of Stuttgart
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).
