Ying‐Sheng Luo
Papers
2
Total Citations
5
H-Index
2
About
Ying-Sheng Luo is a robotics researcher advancing the frontier of agile, real-world robot locomotion through reinforcement learning. His work centers on two critical challenges: enabling versatile multi-gait movement and ensuring smooth, hardware-safe control policies. In his highly cited 2023 paper, Luo introduced the transition-net, a robust strategy for seamlessly switching between dedicated locomotion policies, allowing a single robot to perform diverse gaits in complex real-world settings. This approach distributes the complexity of different movement modes, dramatically expanding operational versatility. Building on this, his 2024 benchmarking study systematically identifies, categorizes, and compares methods to mitigate high-frequency oscillations in deep RL policies—a notorious problem that causes hardware damage and instability. By defining clear metrics for smoothness, Luo provides a vital toolkit for deploying RL controllers on physical robots. With over 5 citations in just two years, his contributions are quickly shaping best practices for reliable, dynamic robot control, making him a rising authority in bridging simulation-trained policies with robust real-world hardware deployment.
Research Focus
Key Achievements
Top Papers
- 1
- 2