About Me

I am a PhD candidate in Robotics at the University of Michigan, working in the Fluent Robotics Lab with Prof. Christoforos Mavrogiannis. I work on motion planning, control, and human-robot interaction, mostly social navigation, getting robots to move among people as easily as people move around each other. Previously I worked with Prof. Harold Soh at NUS’s CLeAR Lab on diffusion-based planning, and with Prof. Howie Choset and Prof. Robin Murphy at CMU’s Biorobotics Lab during my undergrad at BITS Pilani Goa.

My research focuses on navigation under actionable uncertainty in human environments, such as whether a person has noticed the robot or what lies around a blind corner. I develop methods that model this uncertainty and plan motion to reduce it, using the robot’s movement both to gather information and to make its intent clear to the people around it. In Rethinking Legibility, we found that signaling an immediate choice, like which side to pass on, beats signaling a final goal, and holds up even when people are distracted. A related thread is human motion prediction and the data behind it, including how prediction quality shapes navigation and the datasets (ACME, Bi³) and testing tools (SocRATES) these models need. You can reach me at prgoyal@umich.edu.

News

HRI 2026 (LBR). Rethinking Legibility in Social Robot Navigation
HRI 2026. How Human Motion Prediction Quality Shapes Social Robot Navigation Performance
IEEE RA-L. SocRATES, Automated Scenario-based Testing of Social Navigation Algorithms
ICRA 2026. Bi³, A Biplatform, Bicultural, Biperson Dataset for Social Robot Navigation
RSS 2024 workshop. Runner-up Best Paper Award for our scenario-testing paper