Nipuni Karumpulli

Singapore University of Technology and Design

Papers

Total Citations

H-Index

About

Nipuni Karumpulli is a researcher at the forefront of human-robot interaction, specializing in multi-modal communication and gesture-based comprehension systems. Her work addresses a fundamental challenge in robotics: enabling machines to accurately interpret ambiguous human instructions by leveraging naturally-generated gestures. Karumpulli’s most cited paper, "Gesture Enhanced Comprehension of Ambiguous Human-to-Robot Instructions" (2020, 11 citations), introduces M2Gestic, a neural-based system that integrates pointing gestures as an additional input modality. This pioneering approach significantly improves collaborative task performance by resolving linguistic ambiguity, demonstrating that gestures can serve as a robust complement to verbal commands. Her contributions have implications for assistive robotics, manufacturing, and service automation, where clear human-robot communication is critical. Though early in her career, Karumpulli’s work has already shaped discussions on multi-modal interaction, earning recognition for its practical feasibility and innovative fusion of natural user behavior with AI-driven comprehension. Her research continues to push boundaries in making robotic agents more intuitive and responsive to human partners.