Jingzhou Liu
Papers
5
Total Citations
351
H-Index
5
About
Jingzhou Liu is a leading researcher in robot learning and simulation, whose work is shaping how robots acquire complex manipulation skills. His key research areas include simulation-to-reality transfer, dexterous manipulation, and surgical robotics. Liu is best known for developing **Orbit**, a unified simulation framework powered by NVIDIA Isaac Sim that has garnered over 226 citations. Orbit provides a modular, high-fidelity environment for interactive robot learning, enabling researchers to create photo-realistic scenes with both rigid and deformable body physics. Building on this foundation, he introduced **Orbit-Surgical**, an open-source framework specifically designed for learning surgical augmented dexterity, addressing the critical need for fast and accurate surgical simulation. Liu’s work on **DeXtreme** has been equally impactful, demonstrating how deep reinforcement learning can achieve agile, in-hand manipulation of complex objects in simulation and then successfully transfer those skills to real-world robotic hands—a milestone with 88 citations. By bridging the sim-to-real gap and providing accessible, powerful simulation tools, Jingzhou Liu is accelerating progress in both general-purpose and surgical robotics, making him a pivotal figure in the field.
Research Focus
Key Achievements
Top Papers
- 1
- 2DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality88 citations · 2023
- 3
- 4
- 5