Shubham Agrawal
Papers
1
Total Citations
4
H-Index
1
About
Shubham Agrawal advances the frontier of robotic intelligence by bridging physics-based modeling with machine learning. His research centers on differentiable programming for rigid body dynamics, enabling robots to reason about physical interactions through automatic differentiation and continuous sensitivity analysis. In his seminal 2020 work, Agrawal developed a framework that allows gradient-based optimization through complex physical simulations—a critical capability for learning dynamics models directly from data. This approach equips robots with the ability to predict the consequences of their actions in dynamic environments, a foundational step toward truly intelligent behavior. Though early in his career, his contributions have already garnered attention, with his most-cited paper accumulating 4 citations and influencing subsequent work in model-based control and robotic learning. By making physics differentiable, Agrawal provides researchers with powerful tools to train robots that understand cause and effect, moving beyond black-box learning toward physically grounded artificial intelligence. His work sits at the intersection of robotics, control theory, and scientific computing, offering a principled path to machines that can reason about and adapt to the physical world.
Research Focus
Key Achievements
Top Papers
- 1