Yusuke Seto

Papers

Total Citations

H-Index

About

Yusuke Seto is a robotics researcher whose work centers on data-driven modeling and control of legged robotic systems, with a particular focus on hybrid dynamics and gait analysis. His most-cited paper, "Data-driven robot gait modeling via symbolic time series analysis" (2016, 6 citations), introduces a novel approach to modeling hidden-mode hybrid systems (HMHS) by combining symbolic time series analysis with Bayesian mode estimation. Seto’s key contribution lies in enabling robots to autonomously infer and switch between distinct locomotion modes—such as different walking gaits—using only observational data, without requiring explicit pre-programmed models. He experimentally validated this framework on a six-legged T-hex robot, demonstrating how a library of gaits can be learned and deployed for adaptive locomotion. This work bridges the gap between theoretical hybrid systems and practical robotic control, offering a scalable method for robots to handle complex, unstructured environments. Seto’s research has implications for field robotics, particularly in search-and-rescue and exploration missions where adaptive gait selection is critical. Though early in his career, his data-driven approach to mode estimation represents a meaningful step toward more autonomous and resilient legged robots.