Daniele Bernardini

Papers

Total Citations

H-Index

About

Daniele Bernardini is a robotics researcher whose work bridges the critical gap between simulation and real-world application, with a focus on robotic manipulation and constrained reinforcement learning. His most impactful contribution is **6IMPOSE** (18 citations), a novel framework that tackles the persistent "reality gap" in 6D pose estimation for robotic grasping. By enabling deep learning models to generalize beyond synthetic benchmarks to real-world scenarios, this work directly enhances the reliability of autonomous robotic systems in industrial and service settings. More recently, Bernardini has advanced the field of safe reinforcement learning with his work on generating all feasible actions for cyber-physical systems, addressing the critical challenge of enforcing safety and operational constraints in complex, data-driven control. This research is foundational for deploying RL agents in high-stakes environments like autonomous driving and manufacturing. Through these contributions, Bernardini is helping to make robotic systems both more perceptive and more trustworthy, paving the way for their broader adoption in real-world applications where precision and safety are paramount.