research Robot manipulation with VLA models Imitation and reinforcement learning pipelines for real-world robot manipulation built on Vision-Language-Action models. Autonomous solar-panel assembly for lunar bases Dual-arm robotic assembly of solar panels towards full autonomous lunar base construction (MSc thesis, SII 2025). Low-cost teleoperation with force feedback Augmenting the GELLO teleoperation system with force feedback for better data collection (SII 2025). Resolving uncertainty with VLMs and conformal prediction Vision-Language-Model planning with uncertainty quantified via conformal prediction, executed by an imitation learning policy.
