Carlos Maestre

Papers

Total Citations

H-Index

About

Carlos Maestre’s research lies at the intersection of developmental robotics, autonomous skill acquisition, and sensorimotor learning. His work addresses a fundamental challenge in robotics: how machines can discover and interact with objects in open, unpredictable environments without exhaustive pre-programming. In his most cited paper, “Bootstrapping interactions with objects from raw sensorimotor data: A novelty search based approach” (8 citations), Maestre pioneered a method enabling robots to autonomously explore and learn object affordances—the possible actions an object offers—directly from raw sensory and motor data. This approach uses novelty search to drive discovery, bypassing the need for predefined object models. Building on this, his follow-up work, “Iterative affordance learning with adaptive action generation” (3 citations), introduced a framework where robots iteratively refine their interaction skills by adapting their own actions, moving beyond designer-imposed limitations. Maestre’s contributions are significant for advancing autonomous robot learning, offering a pathway toward more adaptive, self-sufficient systems. His research is particularly impactful for students and researchers interested in embodied cognition, open-ended learning, and the future of robots that can learn like living organisms.