Raman Arora

Johns Hopkins University, University of Wisconsin–Madison

Papers

Total Citations

H-Index

About

Raman Arora’s research lies at the intersection of robotics, computer vision, and signal processing, with a unifying focus on leveraging mathematical structure—particularly group theory—to solve complex perception and planning problems. His work on **visual robot task planning** (2019, 8 citations) introduces a novel framework where robots generate sequences of high-level actions directly from a single input image, emphasizing the role of prospection in tackling unfamiliar environments. This approach moves beyond reactive control toward more anticipatory, intelligent behavior in autonomous systems. Earlier, Arora made foundational contributions to **group theoretical methods in signal processing** (2009, 8 citations), demonstrating how learning transformations and invariants can uncover hidden structure in complex, interconnected data. He also pioneered **navigation using a spherical camera** (2008, 4 citations), applying the special Euclidean motion group to relate 3D scenes captured on a sphere, enabling robust autonomous navigation. Though his citation counts are modest, Arora’s work is notable for its theoretical depth and cross-disciplinary ambition, blending abstract algebra with practical robotics. His research offers a compelling roadmap for students interested in mathematically principled approaches to perception, planning, and learning in intelligent systems.