Robert Gieselmann

Papers

Total Citations

H-Index

About

Robert Gieselmann is a robotics researcher whose work sits at the intersection of machine learning, deformable object manipulation, and motion planning. His most impactful contribution is **ReForm**, a robot learning sandbox for deformable linear object manipulation (23 citations), which addresses the critical gap between rigid-body assumptions in most learning-based control algorithms and the reality of manipulating cables, ropes, and hoses. He also pioneered **Planning-Augmented Hierarchical Reinforcement Learning** (21 citations), a hybrid approach that merges model-free RL with graph-based planning to solve long-horizon decision-making problems where environment dynamics are unknown. Additionally, Gieselmann has explored the use of deep generative models—including GANs and VAEs—for density estimation in configuration spaces, offering a systematic study of their benefits and limitations for sampling-based motion planning. His work is notable for bridging theoretical advances in learning with practical robotic challenges, particularly in domains where traditional rigid-body assumptions fail. With a growing citation record and a focus on deformable objects—a notoriously difficult problem in robotics—Gieselmann is establishing himself as a key contributor to the next generation of adaptive, learning-driven robotic systems.