Oscar Gustavsson
KTH Royal Institute of Technology, Eckert & Ziegler (United States)
Papers
2
Total Citations
33
H-Index
2
About
Oscar Gustavsson is a roboticist whose research focuses on enabling robots to learn and adapt in dynamic, real-world environments—particularly in industrial and domestic settings. His key contributions lie at the intersection of behavior tree learning, context-aware planning, and cloth manipulation. In his highly cited 2022 paper, “Combining Context Awareness and Planning to Learn Behavior Trees from Demonstration” (25 citations), Gustavsson addresses a critical challenge for small- to medium-sized manufacturers: how to quickly generate robot programs that can react to unpredictable, collaborative tasks. By integrating context awareness with planning from demonstration, his work allows robots to adapt their behavior trees on the fly, drastically reducing programming time and effort. In parallel, Gustavsson has tackled the notoriously difficult problem of cloth manipulation. His 2022 study on “Cloth manipulation based on category classification and landmark detection” (8 citations) leverages deep learning to classify garment types and detect key landmarks, laying the groundwork for automated laundry and textile handling. Through these contributions, Gustavsson is helping to bridge the gap between rigid industrial robotics and the flexible, perception-driven autonomy needed for the factories and homes of tomorrow.
Research Focus
Key Achievements
Top Papers
- 1
- 2