Robinson Piramuthu
Papers
6
Total Citations
124
H-Index
5
About
Robinson Piramuthu is a leading researcher at the intersection of embodied AI, human-robot interaction, and natural language processing. His work focuses on enabling robots to operate seamlessly in human spaces by combining vision, language, and decision-making. Piramuthu’s most impactful contribution is the **TEACh dataset** (89 citations), a collection of over 3,000 human-human interactive dialogues that train robots to understand and execute instructions while using conversation to resolve ambiguity and correct mistakes. He also developed **VISITRON**, a multimodal Transformer-based navigator that handles dynamic dialogue in vision-and-language navigation. His research on **uncertainty-aware reinforcement learning** (2024) addresses the critical challenge of when robots should request human help, balancing autonomy with assistance. Piramuthu’s work on **RREx-BoT** advances remote object localization in dynamic home environments, while his **anomaly detection dataset** (2024) enables robots to identify dangerous situations, like leaving milk out or forgetting to turn off the stove. With over 120 citations across his key papers, Piramuthu is shaping the future of helpful, conversational home robots that can learn, adapt, and communicate naturally with people.
Research Focus
Key Achievements
Top Papers
- 1TEACh: Task-Driven Embodied Agents That Chat89 citations · 2022
- 2
- 3VISITRON: Visual Semantics-Aligned Interactively Trained Object-Navigator10 citations · 2022
- 4RREx-BoT: Remote Referring Expressions with a Bag of Tricks8 citations · 2023
- 5
- 6TEACh: Task-driven Embodied Agents that Chat2 citations · 2021