Two LMU research projects awarded with ERC Proof of Concept Grants

Computer science professor Majid Zamani and biophysicist professor Benedikt Sabaß have been awarded a Proof of Concept Grant (PoC) by the European Research Council (ERC). The grant is aimed at scientists who have already been awarded an ERC grant and serves to transfer ideas from ERC-funded research projects to specific practical applications.

“I congratulate Benedikt Sabaß and Majid Zamani on this success,” says Prof. Dr. Matthias Tschöp, President of LMU. “Their projects from two highly relevant fields demonstrate how LMU research can have an impact on society and be translated into practical applications.”

Reducing the methane emissions of cattle

Enlarge

Professor Benedikt Sabaß is head of the Cell Biophysics and Statistical Mechanics research group at LMU (Faculty of Veterinary Medicine and Faculty of Physics). In May 2025, he was appointed Professor of Experimental Physics and Medical Physics at TU Dortmund. The project BacForClimate will be carried out at both LMU and TU Dortmund and is funded with approximately 150,000 euros.

As the effects of climate change increasingly take hold, new solutions that significantly reduce greenhouse gas emissions are urgently needed. One of the most effective strategies is to reduce methane emissions. A significant proportion of this methane is produced by microbes in the rumen of cattle. Feed additives are capable of reducing this production. However, the compounds currently available achieve reductions of only 10-30 percent, which is insufficient.

In an ERC Starting Grant funded project, researchers in the group led by Benedikt Sabaß discovered new compounds that have the potential to be a game changer. “Supplements based on our compounds are expected to reduce the methane emissions from cattle by more than 70 percent,” he notes. “Potent anti-methanogenic compounds can also confer significant metabolic benefits, saving up to 12 percent energy.” Accordingly, the new compounds have the potential to additionally improve feed conversion, which will reduce the environmental footprint of feed production and lower the cost of cattle farming. This would make the use of climate-friendly supplements economically viable for producers.

In the Proof-of-Concept project BacForClimate (A potent, new anti-methanogenic compound for climate-friendly livestock farming), Sabaß will clarify the mode of action of the new compounds, assess their metabolic benefits, and establish a roadmap toward commercialization. Subsequently, he will develop products based on the compounds in cooperation with partners from veterinary medicine and industry. “Overall, the project will lay the foundations for an economic success of our new technology which, we hope, can eventually make a contribution to solving an urgent problem”, concludes Sabaß.

New approach for safe autonomous driving

Enlarge

Majid Zamani is professor in the Computer Science Department at the University of Colorado Boulder (USA) and a visiting professor at LMU. He will carry out his project CertiLane in collaboration with LMU as the European host institution. The grant is for approximately 150,000 euros.

Autonomous driving promises greater safety and comfort, but the technology still comes up against its limits, especially when handling complex maneuvers like lane changes. Although test vehicles collect important data over millions of kilometers, edge cases often remain undiscovered. As a result, the systems remain prone to errors – which erodes public trust. The LMU-affiliated research team led by Majid Zamani plans to remedy this situation through CertiLane (Synthesis of Certifiable Control Software for Lane Change in Autonomous Vehicles), which is based on a fundamentally different approach: “We want to move away from guessing or painstakingly ‘experiencing’ safety in favor of clearly demonstrating it – with solid theoretical proofs instead of millions of test kilometers,” says Zamani.

Building on the AutoCPS project, which received an ERC Starting Grant, the team intends to develop a system for automatic lane changing whose software is designed according to formal rules from the beginning. It will no longer vaguely describe requirements, but define them precisely in mathematical terms and make them machine-checkable. Based on this, it will automatically generate the behavior of vehicles in real operating conditions, such as changing lanes – along with verifiable proof that it responds safely and punctually under clearly defined conditions.

Validation proceeds in stages: first in a driving simulator, then on a test track, and then during real operation. To this end, the researchers will reuse proven components from their prior work—such as software tools for rapid scenario generation, methods for assessing the safety of other road users, and techniques for handling uncertainty. The new system is designed to support planning, safety, and verification teams at automakers and suppliers, who will have access to a licensable toolkit that includes formal safety certificates.

The approach promises not only less time and money spent on testing, but also greater transparency. Instead of relying on statistical experience, it will be easier to establish why a system is safe – and where its limits are. “If this succeeds, automatic lane change could become a model for further certifiable driving functions,” concludes Zamani.

Majid Zamani obtained a doctorate in electrical engineering in 2012 and a master’s degree in mathematics from the University of California, Los Angeles. Subsequently, he worked at Delft University of Technology and the Technical University of Munich. In 2019, he took up a visiting professorship at LMU and a professorship at the University of Colorado Boulder.