Max Robotics — Global Service Robot Directory + US Certification

About

Avinava Dubey is a leading researcher at the intersection of robotics, computer vision, and natural language processing, whose work is fundamentally reshaping how robots understand and interact with the world. His most impactful contribution is the groundbreaking RT-2 model, a vision-language-action model that transfers web-scale knowledge directly to robotic control, enabling robots to perform emergent semantic reasoning and generalize to novel tasks with over 267 citations. This work represents a paradigm shift in end-to-end robotic learning. Dubey has also pioneered modular, safe embodied AI systems for dual-arm robots that can follow open-ended natural language instructions, achieving zero-shot learning for long-horizon collaborative tasks. To address the critical challenge of deploying large models on physical robots, he developed SARA-RT (Self-Adaptive Robust Attention for Robotics Transformers), introducing the novel "up-training" method that efficiently converts pre-trained transformers for on-robot deployment. With a research portfolio spanning from foundational vision-language models to practical deployment strategies, Dubey’s work is not only highly cited but is directly enabling the next generation of capable, safe, and generalizable robotic systems.