Shrihari Vasudevan

About

Shrihari Vasudevan is a robotics researcher whose work sits at the intersection of spatial cognition, probabilistic modeling, and autonomous mobile systems. His research has made significant contributions to two core challenges in robotics: enabling robots to understand and represent their environments in human-compatible ways, and modeling complex terrains for field deployment. Vasudevan's most celebrated work centers on cognitive and semantic mapping for mobile robots. His 2007 paper on object-based cognitive maps, with 225 citations, established a foundational framework for how robots can organize spatial knowledge using objects as conceptual anchors. Complementing this, his Bayesian space conceptualization work (2008, 91 citations) advanced semantic place classification, pushing robots closer to human-like environmental understanding — a vision he began articulating as early as 2005 through probabilistic hierarchical spatial representations. A second major thread of his research addresses large-scale terrain modeling. Leveraging Gaussian processes, including heteroscedastic and multi-sensor dependent variants, Vasudevan developed statistically rigorous methods for handling uncertainty and incompleteness in outdoor terrain data — critical capabilities for deploying ground robots in unstructured real-world environments. Together, these contributions reflect a sustained effort to make mobile robots spatially intelligent, perceptually robust, and genuinely field-ready.

Research Focus

Key Achievements

Top Papers

1. 1
   Cognitive maps for mobile robots—an object based approach

   225 citations · 2007
2. 2
   Bayesian space conceptualization and place classification for semantic maps in mobile robotics

   91 citations · 2008
3. 3
   Gaussian Process modeling of large scale terrain

   58 citations · 2009
4. 4
   Heteroscedastic Gaussian processes for data fusion in large scale terrain modeling

   24 citations · 2010
5. 5
   A Tale of Two Object Recognition Methods for Mobile Robots

   22 citations · 2008
6. 6
   Towards a Cognitive Probabilistic Representation of Space for Mobile Robots

   19 citations · 2006
7. 7
   Large-scale terrain modeling from multiple sensors with dependent Gaussian processes

   15 citations · 2010
8. 8
   Evaluation of Three Vision Based Object Perception Methods for a Mobile Robot

   13 citations · 2012
9. 9
   Towards a multilevel cognitive probabilistic representation of space

   13 citations · 2005
10. 10
    A bayesian conceptualization of space for mobile robots

    11 citations · 2007

Key Collaborators

Contact & Links