Hannah Steele

Papers

Total Citations

H-Index

About

Hannah Steele is a robotics researcher whose work pushes the boundaries of how robots are designed and how they behave across radically different scales. Her primary research areas include soft robotics, automated design, and scale-invariant robot behavior. Steele’s most significant contribution is in scalable sim-to-real transfer for soft robot design, where she uses machine learning to automatically propose, test, and refine soft robot designs in simulation before transferring the most promising ones to physical hardware—a process that dramatically accelerates development and overcomes the notorious challenges of manual design. This work, her most cited paper (67 citations), has the potential to democratize soft robotics by replacing labor-intensive trial-and-error with automated optimization. In parallel, Steele has pioneered the concept of scale-invariant robot behavior using fractals, exploring whether robots can maintain identical functionality when deployed at vastly different sizes—an idea that could revolutionize deployment in diverse environments, from micro-robotics to large-scale infrastructure. Her work is notable for its ambition to create robots that are not just optimized for a single task or scale, but are fundamentally adaptable and scalable.