Anirudh Vemula

About

Anirudh Vemula is a robotics researcher whose work spans human-robot interaction, motion planning, and autonomous navigation. His research tackles some of the most pressing challenges in deploying robots in real-world environments — particularly how machines can move safely and intelligently alongside people. Vemula's most celebrated contribution is his work on social attention mechanisms for crowd navigation. His 2018 paper, "Social Attention: Modeling Attention in Human Crowds," has amassed over 700 citations, establishing him as a leading voice in socially aware robot navigation. This work introduced attention-based models that enable robots to predict human trajectories by capturing the implicit cooperative dynamics within crowds — a critical step toward robots that feel natural and safe in human spaces. Beyond crowd navigation, Vemula has made notable contributions to path planning in dynamic environments, developing adaptive dimensionality approaches that make planning more computationally tractable without sacrificing safety. His work on planning with inaccurate models, including the CMAX++ framework, addresses the critical gap between simulated and real-world robot behavior, offering provable performance guarantees during execution. Taken together, Vemula's research bridges theoretical rigor with practical robotics, making him a valuable contributor to the field's ongoing effort to bring capable, trustworthy robots into everyday human environments.

Research Focus

Key Achievements

Top Papers

1. 1
   Social Attention: Modeling Attention in Human Crowds

   703 citations · 2018
2. 2
   Modeling cooperative navigation in dense human crowds

   14 citations · 2017
3. 3
   Path Planning in Dynamic Environments with Adaptive Dimensionality

   14 citations · 2021
4. 4
   Robotics: Science and Systems XVI

   12 citations · 2020
5. 5
   Social Attention: Modeling Attention in Human Crowds

   8 citations · 2017
6. 6
   Path Planning in Dynamic Environments with Adaptive Dimensionality

   6 citations · 2016
7. 7
   CMAX++ : Leveraging Experience in Planning and Execution using Inaccurate Models

   5 citations · 2021
8. 8
   Improved Soft Duplicate Detection in Search-Based Motion Planning

   3 citations · 2022
9. 9
   Planning, Learning and Reasoning Framework for Robot Truck Unloading

   2 citations · 2020
10. 10
    Planning and Execution using Inaccurate Models with Provable Guarantees

    2 citations · 2020

Key Collaborators

Contact & Links