Andrej Orsula

Papers

Total Citations

H-Index

About

Andrej Orsula is an emerging researcher at the forefront of autonomous robotic manipulation, with a particular focus on deep reinforcement learning and generative models for robotic grasping. His work bridges the gap between terrestrial robotics and extraterrestrial exploration, addressing one of the most fundamental challenges in robotics: enabling machines to reliably interact with unknown objects in unstructured environments. Orsula's most recognized contribution, "Learning to Grasp on the Moon from 3D Octree Observations with Deep Reinforcement Learning" (2022, 20 citations combined), tackles the ambitious problem of autonomous robotic manipulation for lunar rovers, using octree-based 3D representations to enable sample collection and scientific data gathering without constant human oversight. This work positions him at a compelling intersection of space exploration and AI-driven autonomy. His more recent work, "GraspLDM: Generative 6-DoF Grasp Synthesis Using Latent Diffusion Models" (2024, 15 citations), demonstrates his evolution toward cutting-edge generative approaches, leveraging latent diffusion models to synthesize diverse, task-relevant grasps — a critical capability for practical deployment of robotic systems. Together, these contributions mark Orsula as a promising voice in next-generation autonomous manipulation research.