Dalton Omens
Papers
2
Total Citations
83
H-Index
2
About
Dalton Omens is a researcher at the forefront of differentiable physics and robotics simulation. His primary contributions lie in developing high-performance, feature-complete physics engines that bridge the gap between traditional simulation and modern machine learning. Omens is best known for creating **Nimble** (nimblephysics.org), a differentiable physics engine for articulated rigid bodies with hard contact constraints. This work, detailed in his most-cited paper (2021, 60 citations), enables gradient-based optimization through complex physical interactions—a capability critical for robotic control, policy learning, and system identification. A follow-up publication (2021, 23 citations) further refined these methods, solidifying Nimble’s status as a comprehensive tool that supports Lagrangian dynamics and contact-rich scenarios often missing from other engines. By making physics simulation fully differentiable and computationally efficient, Omens has empowered researchers to train robots in simulation and transfer skills to the real world with unprecedented accuracy. His work is a cornerstone for anyone exploring model-based reinforcement learning or sim-to-real transfer in robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2