Soochul Yoo
Papers
2
Total Citations
8
H-Index
2
About
Soochul Yoo is a robotics researcher whose work lies at the intersection of geometric deep learning, motion planning, and control. His most notable contribution is the tutorial survey on SE(3)-equivariant robot learning and control, which synthesizes and advances the application of symmetry-aware neural networks to robotic manipulation and navigation—a critical step toward more sample-efficient and generalizable robot policies. This work has already garnered 6 citations shortly after its 2025 publication, signaling its timely impact. Yoo also introduced a novel approach to collision-free motion planning through denoising heat-inspired diffusion with insulators, addressing a key limitation of diffusion-based planners: their heavy reliance on inference-time obstacle detection and extra sensors. By embedding collision avoidance directly into the diffusion process, his method offers a more self-contained and efficient planning paradigm. Through these contributions, Yoo is helping to bridge the gap between theoretical geometric learning and practical robot autonomy, making his research particularly relevant for students and researchers interested in principled, data-efficient approaches to robot learning and control.
Research Focus
Key Achievements
Top Papers
- 1SE(3)-equivariant Robot Learning and Control: A Tutorial Survey6 citations · 2025
- 2