Sonia Raychaudhuri
   

I am an Embodied AI researcher specializing in robot navigation and multimodal spatial reasoning, with a focus on building robust human-centric autonomy for real-world environments. I received my Ph.D. from Simon Fraser University in 2026, advised by Prof. Angel Xuan Chang in the 3DLG Lab under GrUVi Lab.

My research interest lies at the intersection of computer vision, natural language grounding and robotics. In particular, my work focuses on building Embodied AI agents capable of following natural language instructions to navigate in unknown 3D indoor environments, with the long-term research goal of enabling robots to safely co-exist with humans, assisting them in daily life.

I had the opportunity to intern at the Robotics and AI Institute, with Prof. Bernadette Bucher and Duy Ta on combining foundation models with traditional SLAM approaches towards solving object-centric vision and language navigation. Previously, I worked as a Software Engineer with Ericsson and TCS, after I received a Bachelor of Engineering (B.E.) degree from IIEST in India.

Research Summary  /  Email  /  CV  /  github  /  Google Scholar  /  LinkedIn

News

Publications


LangNavBench

 

Survey on Semantic Mapping in Indoor Embodied AI

 

LIFGIF

 

R3DS: Reality-linked 3D Scenes for Panoramic Scene Understanding

 

MOPA: Modular Object Navigation with PointGoal Agents
MOPA: Modular Object Navigation with PointGoal Agents
Sonia Raychaudhuri, Tommaso Campari, Unnat Jain, Manolis Savva, Angel X. Chang
WACV 2024
webpage / paper / arxiv / code

 

Retrospectives on the Embodied AI Workshop
Retrospectives on the Embodied AI Workshop
Matt Deitke, Dhruv Batra, Yonatan Bisk, ... Sonia Raychaudhuri, ... Luca Weihs, Jiajun Wu
arXiv 2022
paper
ECCV 2021
3DV 2020

Teaching Experience