Jonas G. Matt

I am a second-year PhD candidate in Florian Dörfler’s group at the Automatic Control Laboratory (IfA) of ETH Zürich, with Saverio Bolognani and Giuseppe Belgioioso as additional mentors.

In my research, I aim to appy tools from control theory, game theory, and optimization to understand and shape the behavior of strategic agents in complex systems such as power grids. I am currently working on two main threads: developing “functional” incentive mechanisms for procuring control services (see here), and inverse optimization methods for learning agent behavior from observed decisions.

My previous work has ranged from controlling the continuum physics of soft robots to real-time voltage control in power grids. Outside academia, I have gained professional experience in data analytics, renewable energy systems, and IoT technologies.

If you would like to learn more about my work or are looking for open projects, have a look at the MAESTRO project.

news

Apr 11, 2026	Our paper on A Welfarist Approach to Fair Generation Curtailment (arXiv) has been accepted to the 2026 PowerUp Conference in Boulder, CO. I am looking forward to meeting the global power systems community there in September.
Sep 11, 2025	I presented my work done with ABB on the Lossless Compression of Time Series Data at the IEEE ETFA 2025 conference in Porto. If you are curious about how to best compress your time series, check out the paper.
Sep 04, 2025	My poster on Incentive Design for Procuring Control Services received the Community Best Paper Award at the Swiss CLOCK Summit 2025! Check it out here: pdf

selected publications

preprint

Optimal Functional Incentives for Control: The Linear-Quadratic Case with Bilinear Incentives

Jonas G. Matt, Saverio Bolognani, and Florian Dörfler

Apr 2026

arXiv
preprint

A Welfarist Perspective on Fair Generation Curtailment

Jonas G. Matt, Ilia Shilov, and Saverio Bolognani

May 2026

arXiv
preprint

Data-Driven Soft Robot Control via Adiabatic Spectral Submanifolds

Roshan S. Kaundinya, John Irvin Alora, Jonas G. Matt, and 3 more authors

Mar 2025

Abs arXiv

The mechanical complexity of soft robots creates significant challenges for their model-based control. Specifically, linear data-driven models have struggled to control soft robots on complex, spatially extended paths that explore regions with significant nonlinear behavior. To account for these nonlinearities, we develop here a model-predictive control strategy based on the recent theory of adiabatic spectral submanifolds (aSSMs). This theory is applicable because the internal vibrations of heavily overdamped robots decay at a speed that is much faster than the desired speed of the robot along its intended path. In that case, low-dimensional attracting invariant manifolds (aSSMs) emanate from the path and carry the dominant dynamics of the robot. Aided by this recent theory, we devise an aSSM-based model-predictive control scheme purely from data. We demonstrate the effectiveness of this data-driven model on various dynamic trajectory tracking tasks on a high-fidelity and high-dimensional finite-element model of a soft trunk robot. Notably, we find that four- or five-dimensional aSSM-reduced models outperform the tracking performance of other data-driven modeling methods by a factor up to 10 across all closed-loop control tasks.