Arturo Ribes

Imperial College London, Consejo Superior de Investigaciones Científicas, Artificial Intelligence Research Institute

Papers

Total Citations

H-Index

About

Arturo Ribes is a developmental roboticist whose work sits at the intersection of active learning, sensorimotor anticipation, and human-robot interaction. His research focuses on enabling robots to autonomously acquire models of their own bodies and environments through incremental, interactive learning. In his most cited work (16 citations), Ribes introduced an active learning framework for the humanoid robot iCub, allowing it to listen to a human musical performance and then incrementally learn to imitate the sequence using a virtual instrument—all under real-time constraints. This work demonstrated how robots can efficiently balance exploration and task performance. Ribes has also made notable contributions to mobile robotics, developing methods for incremental learning of optical flow models that enable sensorimotor anticipation, allowing robots to predict the consequences of their actions and avoid collisions. His earlier work on object-based place recognition using panoramas helped adapt general object classification methods to the challenges of mobile robotics. Across his publications, Ribes consistently addresses the developmental robotics goal of creating robots that learn like humans—through embodied interaction, incremental model building, and active curiosity.