Yiwen Chen
Papers
2
Total Citations
46
H-Index
2
About
Yiwen Chen is a robotics researcher whose work bridges the critical gap between simulation and real-world robotic control. Their primary research areas include dynamic system identification, deep reinforcement learning, and sim-to-real transfer for robotic manipulation. Chen's most significant contribution is the development of a global optimization approach for identifying the physical dynamic parameters of the KUKA LBR iiwa robot, a problem fundamental to computing accurate link mass matrices. This work, published in 2020 and garnering 42 citations, enables more precise model-based control by extracting a minimal set of dynamic parameters through least squares methods. More recently, Chen has advanced the field of robotic visual insertion tasks by proposing a novel Real2Sim policy adaptation framework using deep reinforcement learning. This 2023 work addresses the persistent challenge of transferring policies learned in simulation to real-world robotic systems, offering a bidirectional approach that improves both simulation fidelity and real-world deployment. Chen's research is particularly valuable for students and practitioners working on robot control, manipulation, and the practical implementation of learning-based methods in industrial robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2