Papers
5
Total Citations
110
H-Index
5
About
Yinglong Miao is a robotics researcher whose work spans motion planning, robot manipulation, and learning-based planning systems. His research bridges classical planning algorithms with modern machine learning approaches, with a particular focus on solving computationally demanding problems in real-world robotic settings. Miao's most impactful contribution is MPC-MPNet (57 citations), a model-predictive motion planning network that achieves near-optimal solutions for kinodynamic motion planning — a notoriously difficult problem requiring robots to simultaneously satisfy kinematic and dynamic constraints. This work significantly advances the state of the art in fast, efficient planning under complex physical constraints. Complementing this, his earlier work on Motion Planning Networks (MPNet) demonstrated how neural networks can learn generalizable near-optimal heuristics applicable to both seen and unseen environments. Beyond motion planning, Miao has made notable contributions to robot manipulation in challenging environments. His research on prehensile rearrangement in cluttered and confined spaces (27 citations) addresses practical warehouse and shelf-stocking scenarios, while his work on online object model reconstruction enables robots to recognize and reuse knowledge of previously encountered objects — a critical capability for lifelong robot learning. His continued focus on occlusion-aware manipulation further demonstrates his commitment to bridging theoretical advances with practical robotic deployment.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5