Juan Antonio Delgado-Guerrero
Papers
3
Total Citations
20
H-Index
3
About
Juan Antonio Delgado-Guerrero is a leading researcher in robot motion learning and non-rigid object manipulation, with a focus on sample-efficient, data-driven approaches. His work centers on using Gaussian process latent variable models and controlled dynamical systems to enable robots to learn complex tasks from minimal demonstrations—a critical step toward deploying robots in unstructured household environments. His most cited paper, "Controlled Gaussian process dynamical models with application to robotic cloth manipulation" (2023, 9 citations), tackles the open problem of handling deformable objects like cloth, where physical interaction is inherently uncertain. Earlier foundational work, "Sample-Efficient Robot Motion Learning using Gaussian Process Latent Variable Models" (2020, 8 citations), advanced kinesthetic teaching by allowing robots to learn motions from only a few human-guided examples. Delgado-Guerrero’s research directly addresses the challenge of making robots easy to instruct by non-experts, a key requirement for domestic robotics. His contributions to contextual policy search and covariate Gaussian process models further push the boundaries of adaptive, context-aware robot learning, positioning him as a rising figure in the intersection of probabilistic machine learning and robotic manipulation.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3