Shivendra Agrawal

CAIRO Lab @ CU Boulder

prof_pic.jpg

University of Colorado

Boulder, Colorado

About myself: I am a final year CS Ph.D. student with Prof. Bradley Hayes at CU Boulder. I am interested in Robotics, Accessibility, and Human Robotics Interaction (HRI) and unifying them to create real-world Assistive Technology. My thesis involves creating foundational methods that enable assistive systems to utilize rich social and semantic cues to support visually impaired individuals. I enjoy teaching and mentoring students and have been fortunate to receive multiple Outstanding TA awards and Instructor awards for excellence in teaching.

I am on the job market for a faculty position starting Fall 2026.

Other interests: I enjoy playing tennis a lot (like a lot!!). On good sunny days, I also like to bike and run in the lovely city of Boulder. (Bonus) Boulder Creek Path Fall view through my bike. I also like to explore local food and beverages.

Minor-flex - I have more than 59 millions views on my Google Map contributions and have received a rather vibrant pair of socks from Google. And no, that was all I ever got from them.

Summary of some of my work:

  • Explainable Robotic Coaching: A framework for a robot to infer a human partner’s likely mental model by observing suboptimal actions to provide targeted coaching with justifications (HRI ‘19 Best Paper Runner-up). Project page

  • Designing Embodied AI for Social Context: Understanding that human environments are governed by social norms, I developed a perceptive robotic cane that used models from psychology to identify socially appropriate seating locations (optimizing for privacy and intimacy), deploying all perception locally on the edge. Project page

  • Learning Decision-Making Policies from Human Behavior: My “ShelfHelp” system learned an optimal guidance policy from human behavioral data to provide manipulation guidance (featured in national media). Project page

  • Robust Semantic Localization: Traditional methods falter in geometrically ambiguous (like retail aisles) environments. “ShelfAware” leverages semantic features within a probabilistic framework to handle quasi-static changes. It does global localization using onboard RGB-D/VIO sensors, eliminating the need for environmental augmentation. Project page

News 2026: