Papers
10
Total Citations
106
H-Index
6
About
Yongle Luo is a robotics researcher whose work sits at the intersection of reinforcement learning, robot manipulation, and sports robotics—particularly table tennis. His most significant contribution is the development of **Relay Hindsight Experience Replay (RHER)** , a self-guided continual reinforcement learning framework that enables robots to master sequential object manipulation tasks under sparse rewards. This work, which has garnered 38 citations, addresses one of the hardest challenges in robotics: learning long-horizon tasks without dense feedback. Luo has also pioneered vision-based manipulation systems that operate without calibration, achieving occlusion-aware monocular control—a breakthrough for unstructured environments. In the domain of table tennis robotics, he has developed novel methods for ball spin estimation from trajectory data and control strategies for returning high-speed spinning balls, earning him recognition in both the sports engineering and robotics communities. His work on **Dense2Sparse reward shaping** tackles the fundamental trade-off between learning efficiency and effectiveness in deep reinforcement learning, proposing adaptive reward functions that balance exploration and exploitation. With over 100 total citations across his publications, Luo’s research is shaping how robots learn from sparse feedback and interact with dynamic, real-world environments—from factory floors to the ping-pong table.
Research Focus
Key Achievements
Top Papers
- 1
- 2A Novel Trajectory-Based Ball Spin Estimation Method for Table Tennis Robot16 citations · 2023
- 3
- 4
- 5
- 6D2SR: Transferring Dense Reward Function to Sparse by Network Resetting8 citations · 2023
- 7
- 8
- 9
- 10