Juan Miguel Santos

Universidad de Buenos Aires, Instituto Tecnológico de Buenos Aires (ITBA)

Papers

Total Citations

H-Index

About

Juan Miguel Santos is a pioneering researcher in robotics and reinforcement learning, with a career spanning foundational work in autonomous systems to applied medical robotics. His key research areas include robot learning, multi-robot coordination, and rehabilitation robotics. Santos made major contributions to reinforcement learning for real-world robots, most notably through his 1999 paper "Exploration tuned reinforcement function" (14 citations), which addressed critical challenges in generalization and exploration for physical robots. His work on dynamic reinforcement function updates (7 citations) further advanced robot learning theory. In multi-robot systems, Santos developed innovative strategies for robot formations as emergent collective tasks and team coordination in robot soccer, exemplified by his "UBA-Sot" approach (9 citations). More recently, he has applied his expertise to medical robotics, designing a control strategy for a tethered follower robot for pulmonary rehabilitation (4 citations), helping COPD patients with walking exercises. With a career bridging theoretical foundations and practical applications, Santos has shaped how robots learn and collaborate, from soccer fields to rehabilitation clinics.