Max Robotics — Global Service Robot Directory + US Certification

About

Abhiram Maddukuri is an emerging robotics researcher whose work centers on robot learning, large-scale dataset construction, and generalist robotic manipulation. His contributions have played a meaningful role in advancing the field's understanding of how diverse, high-quality data can unlock more capable and adaptable robotic systems. Maddukuri has been a contributing collaborator on some of the most impactful recent projects in embodied AI. His involvement in the Open X-Embodiment initiative (119 citations) helped demonstrate that large, cross-platform robotic datasets can train generalizable models analogous to foundation models in NLP and computer vision. His work on DROID (108 citations), a large-scale in-the-wild robot manipulation dataset, tackled the significant logistical challenges of collecting diverse real-world robotic data at scale. He also contributed to RoboCasa (27 citations), which advocates for realistic physical simulation as a scalable alternative to costly real-world data collection. More recently, Maddukuri has explored sim-and-real co-training strategies, investigating how simulation and real-world data can be jointly leveraged to train robust vision-based manipulation policies. Collectively, his research reflects a clear focus on bridging the data scarcity gap that remains one of robotics' most pressing challenges.