Sergey Kolyubin

Quadrupedal robots can achieve up to 20% lower energy consumption by learning speed-dependent gaits with a hierarchical RL policy that explicitly optimizes cost of transport.

Forget fixed datasets: OSMa-Bench++ uses prompt-generated synthetic scenes to create controllable, extensible benchmarks for semantic mapping in robotic manipulation.

Achieve state-of-the-art zero-shot 3D visual grounding by intelligently combining geometric reasoning and on-demand image rendering, outperforming existing methods without task-specific 3D training.

Unlock real-time dynamic scene understanding: R5DGS achieves 11 FPS extrapolation by intelligently applying rigid body constraints to semantically-aware 4D Gaussians.

Semantic SLAM can now understand free-form language queries and ground them in 3D space using only a monocular video feed, opening the door to robots that truly understand and interact with the world around them.