Papers
8
Total Citations
172
H-Index
7
About
Natalia Lyubova is a robotics researcher specializing in developmental robotics, active object learning, and human-robot interaction. Her work centers on enabling humanoid robots to autonomously perceive, learn, and recognize objects through curiosity-driven exploration and interaction — drawing inspiration from how children naturally develop cognitive and perceptual abilities. Lyubova's most influential contribution, "Object Learning Through Active Exploration" (2014, 68 citations), introduced a cognitive architecture that allows robots to incrementally build visual object representations without relying on predefined image databases. This foundational work, alongside complementary studies on curiosity-driven manipulation using the iCub humanoid robot, established her as a leading voice in biologically inspired machine learning for robotics. Her research within the MACSi project further advanced developmental frameworks where robots expand their knowledge through caregiver interaction and self-identification, bridging perception, motor behavior, and social learning. More recently, she has extended her expertise to long-term human-robot interaction, proposing multi-modal incremental Bayesian networks for open-world user identification (2021). With over 170 cumulative citations, Lyubova's work has meaningfully shaped how autonomous robots can learn adaptively in unpredictable, human-centered environments — making her research particularly relevant to researchers exploring embodied cognition, continual learning, and assistive robotics.
Research Focus
Key Achievements
Top Papers
- 1Object Learning Through Active Exploration68 citations · 2014
- 2
- 3
- 4
- 5Multi-modal Open World User Identification13 citations · 2021
- 6Developmental approach for interactive object discovery11 citations · 2012
- 7
- 8Developmental approach of perception for a humanoid robot2 citations · 2013