Ajay Sridhar

Berkeley College, University of California, Berkeley

Papers

8

Total Citations

234

H-Index

5

About

Ajay Sridhar is a robotics researcher whose work sits at the intersection of machine learning, computer vision, and autonomous navigation. His research focuses on developing generalizable, data-driven policies that enable robots to navigate complex real-world environments — including unfamiliar spaces and crowded social settings — without requiring task-specific programming from scratch. Sridhar's most impactful contribution, **NoMaD** (96 citations), introduces a unified diffusion-based policy framework that elegantly handles both goal-directed navigation and open-ended exploration within a single model. His work on **GNM** (71 citations) demonstrated that training across heterogeneous robot platforms could yield powerful, transferable navigation models — a significant step toward generalizable robotics. Complementing this, **ViNT** extends the foundation model paradigm to visual navigation, while **SACSoN** (31 citations) addresses the nuanced challenge of socially compliant robot behavior in human-populated environments. Through frameworks like **ExAug** and **SELFI**, Sridhar has also advanced techniques for data augmentation and autonomous self-improvement, tackling the persistent scarcity of diverse robotic training data. With over 230 cumulative citations across recent publications, his contributions are shaping the future of scalable, adaptable robotic navigation systems.

Research Focus

Key Achievements

5
H-Index
8
Papers
234
Total Citations
29
Avg Citations/Paper
🏆 Most Cited Paper
NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration
96 citations · 2024
📈 Most Prolific Year: 2023 (5 Papers)
🤝 Key Collaborators: 9
🏛 Institutions: Berkeley College, University of California, Berkeley

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago