Rongshun Juan
Papers
6
Total Citations
163
H-Index
5
About
Rongshun Juan is a leading researcher in robot learning, specializing in deep reinforcement learning, sim-to-real transfer, and imitation learning. His work addresses the critical challenge of enabling robots to learn complex behaviors efficiently and safely, bridging the gap between simulation and the real world. Juan’s most impactful contribution, “Transferring policy of deep reinforcement learning from simulation to reality for robotics” (2022), has garnered 128 citations, establishing a foundational framework for deploying simulated policies in physical robots. He has advanced imitation learning through generative adversarial imitation learning (GAIL), as seen in his 2023 work on GAN-based interactive learning, which allows robots to surpass expert performance by combining demonstrations with human evaluative feedback. His innovative approaches include shaping progressive networks for policy transfer (2021) and model-based adversarial imitation learning (2023), both of which integrate human reward to enhance sample efficiency. Juan has also contributed to autonomous navigation with his AD-RRT* path planning method for underwater gliders (2025). His research consistently pushes the boundaries of sim-to-real transfer, as demonstrated in his adaptive forward dynamics model (2023), which improves policy robustness. With a growing citation impact, Juan’s work is pivotal for students and researchers aiming to develop practical, real-world robot learning systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6