Radu Cordorel

About

Radu Cordorel is a researcher at the intersection of robotics and machine learning, with a primary focus on bridging the gap between simulated training and real-world deployment—a challenge known as sim-to-real transfer. His most cited work, "Sim-to-Real Transfer Learning using Robustified Controllers in Robotic Tasks involving Complex Dynamics" (2019, 15 citations), addresses a critical bottleneck in deep reinforcement learning for robotics. Cordorel demonstrates how controllers trained in simulation can be made robust enough to handle the unpredictable dynamics of physical systems, effectively allowing robots to learn complex tasks in safe, controllable virtual environments before acting in the real world. This approach leverages the key advantages of simulation—such as full environmental control and the ability to pause motion during computation—while mitigating the notorious "reality gap." His contributions are particularly valuable for tasks involving intricate dynamics where direct real-world training would be costly or dangerous. By advancing methods that make simulated learning practical for physical robots, Cordorel is helping to accelerate the development of autonomous systems that can learn and adapt more efficiently.

Research Focus

Key Achievements

Top Papers

1. 1
   Sim-to-Real Transfer Learning using Robustified Controllers in Robotic Tasks involving Complex Dynamics

   15 citations · 2019

Key Collaborators

Contact & Links