Auf dem Laufenden bleiben

Keep up to date with the NomadZ robotic football team here, where team member Nicole Damblon tells us how their 2023 has gone.

Congratulations to two NCCR Automation researchers who have won SNSF Swiss National Science Foundation grants this week. Niao He who won a starting grant for her project “Optimization for Modern Reinforcement Learning: from Principles to Scalability” and Mengshuo Jia who won for a project on rethinking power system computation from a fundamental and non-traditional perspective.

As a journal editor and award-winning researcher, Francis Bach has his finger on the pulse of AI. He talked to us about exciting new findings, how to spread ideas, and why convexity isn’t always the right goal.

New episode of the inControl podcast! At our recent NCCR Symposium host Alberto Padoan grabbed Tryphon Georgiou (UC Irvine) for a two part chat on modeling, robustness and fundamental limitations of feedback systems. This episode covers marathons, interpolation problems, metrics and the interplay between feedback and causality.

Congratulations to Wouter Jongeneel who has won the 2023 INFORMS Optimization Society Award for his paper "Small Errors in Random Zeroth-Order Optimization are Imaginary" (co-authored with Man-Chung Yue and Daniel Kuhn)

Congratulations to Paolo Gherardo Carlet, Andrea Favato, Saverio Bolognani and Florian Dorfler for winning the Transactions on Power Electronics (TPEL) Prize Paper Award for their paper "Data-Driven Continuous-Set Predictive Current Control for Synchronous Motor Drives"

Herkömmliche Lösungen zur Verkehrsentlastung haben ein Problem: Sie sind von Natur aus ungerecht. Durch die Erhebung von Gebühren für den Zugang zu Strassen werden einkommensschwache Bevölkerungsschichten stark benachteiligt. Unsere Forscher haben vielleicht einen Weg gefunden, die Welt ein bisschen gerechter zu machen.

Am 24. Oktober 2023 hat die Gründungsversammlung vom Reallabor Sisslerfeld für Energie- und Mobilitätsthemen in Münchwilen stattgefunden. Unter den Anwesenden waren neben Dieter Egli, Regierungsrat des Kantons Aargau und dem Präsidenten der ETH Zürich, Joël Mesot, weitere hochrangige Vertreter aus Politik, Verwaltung, Forschung und Industrie.

We're back with Season 2! 🎉 In our first episode, Davide Scaramuzza from University of Zurich takes us on a journey through magic, autonomous vision-based navigation, agile drone racing, and event-based cameras. 🤖 Ready for takeoff?

The Third Champéry Power Conference on 4-9 February 2024 is organized by the Institute of Sustainable Energy at the School of Engineering of Valais and the Automatic Control Laboratory of the Swiss Institute of Technology (ETH) in Zürich. NCCR Automation are proud to sponsor this event.

In the last week of September we welcomed 80 people to the NCCR Symposium on Complex Interconnected Systems and decision making in measure spaces. This is how it went...

The distinction between “convex” or “non-convex” optimization isn’t just a technicality. Convexity makes for calculations that can guarantee the best result in an automated process – here’s how.

Thank you to Control Systems Magazine who featured NCCR Automation as their Institute in Control.

Each year, NCCR Automation works with a number of our sister NCCRs to put on the bench2biz event, a two week workshop for pre-seed research ideas. Each year, carefully curated experts are invited to join to give advice and feedback, and as this initiative is run through the NCCRs, the organisers can give personalized, tailored feedback and coaching.

For 2022’s edition, NCCR Automation member Manos Barmpounakis took part and was coached by KTT officer Ben Sawicki. We asked Manos for his thoughts on the experience.

Apply by the end of September to spend a period of up to six months in an NCCR Automation lab.

Der Mathematiker Yurii Nesterov ist einer der Hauptakteure auf dem Gebiet der konvexen Optimierung, die für viele Anwendungen in der Automatisierungstechnik entscheidend ist. Unter anderem dank seiner Durchbrüche dehnt sich dieses Gebiet auf viele andere Disziplinen aus. Damit ist er nicht so glücklich, wie man meinen könnte.

Humans have been trying to do more with less for a very long time. We’d like cars and planes that go further but use less fuel and have a smaller carbon footprint. We’d like food that is optimally nutritious but also affordable, and we’d like energy solutions for our house, cities and countries that are effective but also green and affordable.

The jury gave the following motivation: "The paper is being recognized for making a timely contribution to address noise-corrupted datasets in data-driven modeling and control through a principled statistical approach. The paper stands out for its completeness and high-impact potential in applications including predictive control and system identification".

Der Energiesektor verändert sich und Energieflexibilität entwickelt sich zu einem neuen Markt. Aber wie kann diese Flexibilität bei so viel Unsicherheit effektiv gehandelt werden? Julie Rousseau entwickelt ein Modell, das die Komplexität der realen Lebenswelt aufschlüsselt.

Congratulations to Varsha Behrunani for winning best poster award at the DTU PES summer school: Advances in OR and AI

Die Vorbereitungen für den diesjährigen RoboCup, der in Bordeaux, Frankreich, stattfinden wird, sind in vollem Gange. Dieses internationale Roboterfußballturnier vereint Studententeams aus verschiedenen Ländern, die in 7-gegen-7-Spielen gegeneinander antreten. Neben dem Wettbewerb ist der RoboCup auch eine Robotik-Konferenz.

Many congratulations to Baiwei Guo for winning Best PhD Paper Award at ECC23.

Auch wenn sich die Möglichkeiten von Internet-of-Things-Anwendungen vervielfachen, wird das Wachstum durch Energieprobleme gebremst. Eine Strategie ist das Energy Harvesting, das jedoch mit grundlegenden Einschränkungen verbunden ist. Naomi Stricker arbeitet an Lösungen.

We hosted 100 people at ETH's conference centre in Monte Verita, Ticino, and what a week we had!

In einer Zeit, in der Länder auf der ganzen Welt von Energieknappheit bedroht sind, haben ETH Zürich / NFS Automation -Forscher eine wirkungsvolle und kostengünstige Softwarelösung entwickelt, um die Nutzung dezentraler Energieressourcen zu verbessern. Dieser innovative Ansatz ermöglicht es Netzbetreibern, beispielsweise Photovoltaikanlagen durch einen flexiblen, reaktionsschnellen und datengesteuerten Betrieb optimal zu nutzen.

Obwohl neuronale Netze oft beeindruckende Leistungen erbringen, gelingt es ihnen manchmal nicht, einfache Konzepte zu erfassen, die der Mensch intuitiv versteht - ein Effekt, der sogar Leben gefährden kann. Aber können wir unser menschliches Wissen auf neuronale Netze übertragen, damit sie sinnvollere Lösungen lernen? Loris Di Natale, Forscher am NFS Automatisierung, hat einen Plan um sicherzustellen dass automatisierte Maschinen uns nicht ausschliessen.

Julie Rousseau is a PhD student at ETH Zurich and Empa looking at stabilising the energy grid.

Anna Scampicchio is a postdoctoral researcher at ETH Zurich who does theoretical research on why some systems behave the way they do.

Rahel Rickenbach is a PhD student at the Institute of Dynamic Systems and Control, at ETH Zurich.

Jelena Trisovic is a Ph.D. student at the ETH AI Center, she obtained an M.Sc. in Electrical Engineering and Information Technology from ETH Zürich and a B.Sc. in Electrical Engineering and Computing from the University of Belgrade.

Alisa Rupenyan is a senior scientist at the Automatic Control Lab at ETH Zurich and Head of the Advanced Control and IoT group at Inspire, the technology transfer organization at ETH Zurich. She leads research projects in the intersection between process optimization, industrial control, and machine learning for applications strongly connected to manufacturing.

Clara Galimberti is a PhD student at EPFL. She received her degree in Electronic Engineering from Universidad Nacional de Rosario (UNR), Argentina, in 2018. Her research lies at the intersection of machine learning and control theory.