Nora Etxezarreta

Papers

Total Citations

H-Index

About

Nora Etxezarreta is a researcher at the forefront of modular robotics and artificial intelligence, pioneering methods to make machines more adaptable and intelligent. Her work centers on the convergence of hardware modularity and deep reinforcement learning (DRL), with a key focus on developing hierarchical learning frameworks that allow modular robots to simultaneously master multiple tasks. This approach, she argues, is critical for achieving true reconfigurability in robotic systems. In her most-cited works, each garnering 4 citations, she proposed a novel end-to-end DRL framework that trains robots directly from joint states using traditional robotic tools, and introduced a hierarchical method enabling multi-task learning in environments composed of different modular robots. Etxezarreta has also advanced the concept of the "self-adaptable robot," arguing that hardware modularity is the key to moving from programming machines to training them, leading to more capable and adaptable systems. Her research lays the groundwork for a future where robots can autonomously reconfigure and learn, promising significant impacts on the field of autonomous systems.