[June, 2026]
Our US patent on safe control inputs for vehicle actuators, a safety filter built on a maximal controlled invariant set, has been granted (US 12,662,145 B2). It is the second granted family after the German patent on stability-constrained safety filters (DE 10 2023 211 873 B3, January 2025); see the patents section below.
[May, 2026]
“L4acados: Learning-based models for acados, applied to Gaussian process-based predictive control” is now published in the IEEE Transactions on Control Systems Technology (vol. 34, no. 3). Besides the open-source software, the paper reports GP-based MPC on a full-scale research vehicle in an ISO 3888-1 double lane change, solving in about 8 ms within a 20 ms control cycle.
[April, 2026]
New preprint: Efficient uniform feasible set sampling for approximate linear MPC, joint work with Elias Milios, Felix Berkel, Felix Gruber and Melanie Zeilinger.
[November, 2025]
“Statistically consistent approximate model predictive control” has been published in the IEEE Control Systems Letters (vol. 9). This is joint work with Elias Milios, whose doctoral research at Bosch Research I supervise together with Prof. Melanie Zeilinger (ETH Zurich).
[April & July, 2025]
Two new preprints:
- Tunable real-time safety filters via set-based control barrier functions, with an experimental validation on an electric drive at a sampling time of 150 µs (revised June 2026, currently under review)
- Contract-based hierarchical control using predictive feasibility value functions
[August & December, 2024]
Joint work supported by ELO-X:
Two conference papers:
- Stability Mechanisms for Predictive Safety Filters, Conference on Decision and Control, Milan, Italy
- Predictive Stability Filters for Nonlinear Dynamical Systems Affected by Disturbances, Nonlinear Model Predictive Control Conference, Kyoto, Japan
[June, 2024]
We contributed to the Annual Learning for Dynamics & Control Conference at Oxford:
- Paper: Learning Soft Constrained MPC Value Functions: Efficient MPC Design and Implementation providing Stability and Safety Guarantees
- Tutorial: Safety Filters for Control: Concepts, Theory and Practice
[October, 2023]
I am happy to announce that our article “Data-Driven Safety Filters” has been published in the IEEE Control Systems Magazine as part of a special issue on data-driven control.
Our article includes tutorials on three classes of safety filters: Hamilton-Jacobi Reachability, Control Barrier Functions, and Predictive Control. The article covers applications to first-principle and data-driven models, a selection of model-based and learning-based controllers, and several illustrative case studies from the robotics domain.
By collaborating with experts in the areas of Control Barrier Functions (Andrew Taylor, Prof. Aaron Ames, Caltech) and Safety Filters using Hamilton-Jacobi Reachability Analysis (Jason J. Choi, Prof. Koushil Sreenath, Prof. Claire Tomlin, UC Berkeley), the article aims to provide a comprehensive overview of the state of the art, including work on predictive safety filters done in my former group ICS – Intelligent Control Systems at ETH Zurich (Prof. Melanie Zeilinger).
[May, 2023]
Our recent post on the Bosch Research Blog showcases some exciting future applications for safety-critical control methodologies and their interplay with humans and ai.
[December, 2022]
Two new pre-prints are available, which will be presented at the European Control Conference 2023:
- Approximate Predictive Control Barrier Functions using Neural Networks: A Computationally Cheap and Permissive Safety Filter
- LQG for Constrained Linear Systems: Indirect Feedback Stochastic MPC with Kalman Filtering
[October, 2022]
Our paper “Cautious Bayesian MPC: Regret Analysis and Bounds on the Number of Unsafe Learning Episodes” was accepted and will be published in the IEEE Transactions on Automatic Control in August 2023.
[June, 2022]
We have contributed to the session “A Tutorial on Nonlinear Model Predictive Control: What Advances Are On the Horizon?” at the American Control Conference, Atlanta, 2022. The corresponding paper will be published in the conference proceedings and can be found, e.g. here.
[May, 2022]
I am happy to share that our work on connecting predictive safety filters and control barrier functions, “Predictive control barrier functions: Enhanced safety mechanisms for learning-based control” will be published in the IEEE Transactions on Automatic Control in May 2023.
[March, 2022]
Our paper Learning-based Moving Horizon Estimation through Differentiable Convex Optimization Layers has been accepted for the 4th Annual Learning for Dynamics & Control Conference at Stanford and was selected for oral presentation.
[December, 2021]
Our paper A predictive safety filter for learning-based control of constrained nonlinear dynamical systems was selected as Editor’s Choice Automatica in July 2021
Videos
More videos are on my YouTube channel.
[15.5.2023]
Showcasing some of our safety-critical control methods at Bosch Research
[27.09.2021]
Spotlight talk at IROS 2021 Workshop on Safe Real-World Robot Autonomy: Summarizes some of the more recent work on predictive safety filters.
[25.02.2021]
A predictive safety filter for a miniature racing application and its combination with imitation learning to safely learn an expert policy: Preprint
[19.01.2021]
I presented the concept of predictive safety filters as part of the autonomy talks at ETH. Due to the current pandemic, the talks are available online.
Patents
I am inventor on eleven published patent families (Robert Bosch GmbH, dates 2023 to 2025) on safety filters for actuators and drives, stability-constrained safety filters, learned MPC value functions and optimization-based controller design. Two have been granted so far:
- DE 10 2023 211 873 B3 (granted January 2025): stability-constrained safety filter with reference trajectory
- US 12,662,145 B2 (granted June 2026): safe control inputs for vehicle actuators based on a maximal controlled invariant set
The full list is on Justia Patents.
About me
I am a researcher at Bosch Corporate Research in Renningen, where I lead work on safety filters and learning-based control for vehicles and electric drives, from method to full-scale test-track demonstrations, and co-lead a cross-site research programme on scalable optimal control (model predictive control and reinforcement learning). Before that, I was a doctoral and postdoctoral researcher in the Intelligent Control Systems group of Prof. Melanie Zeilinger at IDSC ETH Zurich, where I developed predictive safety filters. My research interests are safe learning-based control, model predictive control and model-based reinforcement learning. What interests me most at the moment is safe agentic control engineering: letting AI agents design control policies while safety stays strictly guaranteed. I review for IEEE TAC, Automatica, IEEE RA-L and ICRA; also, see my Google Scholar profile.
For a detailed CV, see my LinkedIn profile.