Max Robotics — Global Service Robot Directory + US Certification

About

Zehan Ma is a robotics researcher whose work centers on scaling robot learning through large-scale data collection, deformable object manipulation, and generative design for assembly. His most impactful contribution is the creation of **DROID**, a large-scale, in-the-wild robot manipulation dataset that has already garnered over 100 citations. This dataset provides diverse, high-quality trajectories collected across varied environments, serving as a critical resource for training more robust and generalizable robotic manipulation policies—a foundational step toward real-world deployment. Ma also tackles challenging industrial tasks, such as automating **deformable gasket assembly**, a long-horizon, high-precision problem common in manufacturing. More recently, he introduced **Blox-Net**, a generative design-for-robot-assembly framework that leverages vision-language models and physics simulation to translate natural language instructions into feasible assemblies. By bridging data-driven learning with practical automation challenges, Ma’s work advances both the science and engineering of robot manipulation, offering scalable solutions for complex, real-world tasks.