Research
Research experience in robot learning and vision-language-action models.
CV:HCI Lab & H²T Lab, Karlsruhe Institute of Technology
Jun 2026 – Present
Research Assistant
- Supporting VLA foundation model adaptation for ARMAR-6/7 humanoids and bimanual Franka Emika platforms via Isaac Sim workflows, hardware interfacing, and teleoperation support.
- Contributing to research on reducing manual human supervision needed to adapt VLA models to new robot embodiments.
- Employed at the CV:HCI Lab (Prof. Rainer Stiefelhagen); supervised by Dr.-Ing. David Schneider, working within the H²T group (Prof. Tamim Asfour).
Intuitive Robots Lab (IRL), Karlsruhe Institute of Technology
Oct 2025 – Present
Research Intern — supervised by Pankhuri Vanjani
Project: Imitation Learning with State Space Models
- Investigating whether SSM recurrence, used as a primary temporal backbone rather than a shallow add-on, improves long-horizon, memory-dependent robot manipulation.
- Co-developed a state space model–based architecture jointly processing vision, language, and proprioception, with a flow-matching action head for chunked prediction and recurrent memory that carries task state across a trajectory.
- Ran controlled architecture ablations over memory depth, trunk depth, and visual token compression using a reproducible evaluation harness, evaluated in simulation and on real-robot memory tasks on a Franka Emika Panda.
- Results and a preprint will be released soon.
- Conducting this research under the supervision of Pankhuri Vanjani and TT-Prof. Dr. Rudolf Lioutikov.