Adrian Ragobar

Papers

Total Citations

H-Index

About

Adrian Ragobar is a robotics researcher whose work bridges the gap between data-driven grasp planning and real-world reliability. His key research areas include robotic manipulation, grasp synthesis, and human-robot interaction. Ragobar's most notable contribution is the creation of the EPFL Grasp Dataset, a massive collection of 37,000 human-planned, six-degree-of-freedom robotic grasps. This dataset, detailed in his highly cited 2020 paper, addresses a critical limitation in deep learning-based grasp planning: the persistent failure rate of approximately one in ten attempts. By providing a large-scale, human-annotated resource, Ragobar has enabled researchers to train more robust and practical grasping algorithms, moving the field closer to truly autonomous manipulation. His work has garnered 5 citations, reflecting its foundational role in improving grasp success rates. Ragobar's achievements highlight his commitment to enhancing the reliability of robotic systems, making him a key figure in advancing data-driven approaches to complex manipulation tasks.