Projects |

AutoLern

Sun, 01 Jun 2025 00:00:00 +0000

I joined AutoLern toward the later phase of the project to support the application of machine learning in production environments. My contribution focused on helping translate research results into deployable solutions that could be integrated into industrial workflows rather than remaining isolated prototype models.

In particular, I worked on questions around deploying machine learning models on real systems, where robustness, maintainability, and interaction with existing automation infrastructure matter just as much as model quality. This work complemented the broader project goal of enabling self-learning machine tools that can assess model validity during operation and adapt to changing boundary conditions.

MARIA Lab

Fri, 25 Oct 2024 00:00:00 +0000

I devised the MARIA Lab as a rapid experimentation environment for robotic automation. The idea was to create a playground where new concepts in robot intelligence, automation logic, and system integration could be developed and tested quickly before being transferred to larger and more complex robotic platforms.

Our large industrial setups are highly capable, but they also make integration effort expensive and slow. I designed MARIA together with my students to lower that barrier: a lab architecture that allows researchers to prototype capabilities, evaluate system behavior, and validate automation ideas in a fast and accessible way. The lab is equipped with multiple robotic arms, vacuum grippers, a hot end for additive manufacturing, a milling system, and a variety of sensors.

Circular Factory for the Perpetual Innovative Product

Mon, 01 Apr 2024 00:00:00 +0000

I have been involved in CRC 1574 “Circular Factory for the Perpetual Product” from an early stage, initially as the working group leader of the Changeable Autonomous Production Systems group. In that role, I coordinated the automation-related activities on the shop floor and helped align the technical work across the different contributors involved in building the circular production demonstrators.

Within the CRC, I am also responsible for the Transformer Cell, a robot-based manufacturing system that is intended to autonomously reconfigure itself and autonomously plan and execute assembly, disassembly, and reprocessing steps. This system captures much of what makes the CRC unique for me: bringing together modular automation, intelligent planning, and circular manufacturing in one adaptive production environment.

Later, I took on the role of research coordinator for the entire collaborative research center. In this position, I coordinate research activities across the full cluster and help connect the different disciplines contributing to the vision of an integrated circular factory. The project brings together production technology, robotics, computer science, knowledge modelling, ergonomics, and materials research to enable industrial-scale circular production systems.

SDPräFlexBot

Thu, 26 Oct 2023 00:00:00 +0000

In the PräFlexBot project, my main contribution was on trajectory planning and system-level integration. What makes the setup special is the cooperation between two robots: one robot presents and repositions the part, while the other robot performs the actual manufacturing task in the pose where it can achieve the highest possible precision.

The project combines flexible robot kinematics with high-precision optical sensing and adaptive planning methods in order to improve manufacturing accuracy. My work focused on the planning side of this problem: developing and shaping the trajectory and configuration concepts that allow the two robots to cooperate effectively and place the working robot in accuracy-optimal configurations along the process path. The real-time control based on measurement feedback was carried out by our collaborators in Stuttgart, while my role centered on the planning logic and the overall system perspective required to make the setup effective.

SDMBot

Sat, 01 Apr 2023 00:00:00 +0000

SDMBot was built as a fair demonstrator to showcase not only our PyBullet Industrial simulation environment, but also the broader idea of planning manufacturing processes digitally before deploying the resulting paths on a real robot. Instead of treating simulation as a side tool, the demonstrator was designed to make digital process planning itself tangible and intuitive.

I designed this exhibit together with a colleague. The central idea was that visitors could step into the role of the planning system: using VR goggles, they digitally plan a welding application in the virtual environment and then see how that planned task is transferred to the physical robot. In that way, the demonstrator communicates both the capabilities of PyBullet Industrial and the larger vision of simulation-based manufacturing planning in an accessible, hands-on format.

TriPed Robot

Sun, 25 Oct 2020 00:00:00 +0000

The TriPed project was one of the robotics systems for which I was the main driver. This threelegged robot was meant as a plattform to explore novel locomotion strategies and I was deeply involved in turning that concept into a working system.

My contribution covered both the high-level and hands-on sides of the project. I was involved in major hardware and software design decisions, developed large parts of the control architecture, and contributed substantially to implementation work ranging from wiring and electrical integration to programming and simulation. In short, I was chiefly responsible for pushing the three-legged robot from concept toward a functional experimental platform.

Value Stream Kinematics

Sun, 25 Oct 2020 00:00:00 +0000

Value Stream Kinematics is a large research effort at our institute that explores how robot-like kinematics can be used to realize highly flexible and still productive manufacturing systems. The broader goal is to combine the adaptability of industrial robots with the efficiency normally associated with specialized machinery.

Within this effort, I am one of the people responsible for making the system more intelligent. My work has focused in particular on control, trajectory planning, and architectural decisions that shape how such systems can be operated effectively in practice. I contribute to the methods and system concepts that help translate the overall vision into a usable automation approach.