Juan Wilches
Papers
2
Total Citations
36
H-Index
2
About
Juan Wilches is a robotics researcher whose work centers on enabling robots to acquire and generalize complex manipulation skills through self-supervised learning. His primary contributions lie in developing data-driven approaches that allow robots to learn from their own experiences, reducing reliance on human demonstrations. His most cited paper, "Robot gaining accurate pouring skills through self-supervised learning and generalization" (2020, 31 citations), presents a method where a robot learns precise pouring by repeatedly practicing and refining its actions, achieving robust performance across different containers and liquids. In related work, "Generalizing Learned Manipulation Skills in Practice" (2020, 5 citations), Wilches introduces an RNN-based skill model that learns from demonstrations and then adapts to new settings, addressing a core challenge in robotics: transferring learned behaviors to novel environments. His research has significant implications for domestic and industrial robotics, where adaptability is key. By focusing on practical, generalizable skill acquisition, Wilches is helping to move robots closer to autonomous operation in unstructured, real-world settings.
Research Focus
Key Achievements
Top Papers
- 1
- 2Generalizing Learned Manipulation Skills in Practice5 citations · 2020