Max Robotics — Global Service Robot Directory + US Certification

About

Arjun Majumdar is a leading researcher at the intersection of embodied AI, computer vision, and robotics, with a focus on building agents that can perceive, reason, and act in the physical world. His work has been instrumental in advancing vision-and-language navigation (VLN) and object-goal navigation (ObjectNav). In a landmark contribution, Majumdar introduced ZSON (Zero-Shot Object-Goal Navigation using Multimodal Goal Embeddings), a scalable approach that enables robots to find open-world objects without task-specific training—a paper with over 40 citations that has shaped the field of zero-shot navigation. He also pioneered sim-to-real transfer for VLN, bridging the gap between simulation and real-world deployment, and recently led the development of OpenEQA, a modern framework for embodied question answering that leverages foundation models to answer natural language queries about environments. Beyond navigation, Majumdar explores socially impactful applications, such as using mobile manipulators to lead physical therapy exercise games for Parkinson’s disease patients. His work consistently pushes the boundaries of how robots can understand and interact with human spaces, with his most cited papers accumulating over 100 citations and influencing both academic research and real-world robotic systems.