Raghu Rajan

Papers

Total Citations

H-Index

About

Raghu Rajan is a researcher working at the intersection of robotics, machine learning, and autonomous systems, with a particular focus on enabling robots to understand and interact with the physical world through learned representations. Their most notable work, "T3VIP: Transformation-based 3D Video Prediction" (2022), addresses one of the fundamental challenges in autonomous robotics: equipping robots with the ability to reason about future physical outcomes by learning from their own past experiences. By developing a transformation-based framework for 3D video prediction, Rajan's research pushes the boundaries of how robots can model the dynamic, three-dimensional rules that govern the real world — a critical capability for autonomous skill acquisition. This approach reflects a broader commitment to grounding robot learning in physically meaningful representations rather than purely abstract statistical patterns. While the work is still accumulating citations in its early stages, the problem it tackles — enabling robots to predict and plan within complex physical environments — sits at the heart of modern robotics research. Rajan's contributions position them as an emerging voice in the field of robot learning and world modeling.