Claudio Fantacci
Papers
10
Total Citations
255
H-Index
7
About
Claudio Fantacci’s research bridges the critical gap between perception and action in robotics, with a focus on autonomous navigation, human-robot collaboration, and generalist manipulation. His early work on unscented Kalman filters for autonomous underwater vehicles (96 citations) established foundational techniques for robust state estimation in challenging environments. Fantacci has since made significant contributions to industrial robotics, notably developing adaptable workstations that dynamically reconfigure to improve worker ergonomics and productivity—a direct response to the high costs of musculoskeletal disorders (94 citations). His research extends to advanced manipulation, where he tackles the complex problem of robotic stacking of diverse shapes using reinforcement learning, moving beyond simple pick-and-place operations. Fantacci also contributed to the development of RoboCat, a self-improving generalist agent capable of mastering novel skills across different robotic embodiments, representing a step toward foundation models for robotics. His work on MaskUKF, combining instance segmentation with Kalman filtering for 6D object pose tracking, exemplifies his integration of deep learning with classical estimation methods. With over 250 total citations, Fantacci’s research consistently addresses real-world deployment challenges, from underwater navigation to collaborative manufacturing and general-purpose robotic intelligence.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapes16 citations · 2021
- 4
- 5RoboCat: A Self-Improving Generalist Agent for Robotic Manipulation9 citations · 2023
- 6
- 7
- 8
- 9Gemini Robotics: Bringing AI into the Physical World4 citations · 2025
- 10Lossless Adaptation of Pretrained Vision Models For Robotic Manipulation3 citations · 2023