Papers
8
Total Citations
109
H-Index
5
About
Leandro de Souza Rosa is a robotics researcher whose work sits at the intersection of human-robot interaction and efficient embedded systems. His primary research areas include interactive imitation learning (IIL), bimanual robotic manipulation, and hardware-optimized algorithms for autonomous navigation. Rosa’s major contribution is advancing how robots learn from humans in real-time: his 2022 survey on Interactive Imitation Learning (53 citations) established a foundational framework for allowing non-expert users to correct robot behavior during execution, rather than requiring pre-programmed demonstrations. He extended this concept to dual-arm systems in his 2023 paper on bimanual movement primitives (23 citations), addressing the critical challenges of synchronization and coordination for industrial and domestic applications. Earlier in his career, Rosa tackled the practical problem of deploying complex algorithms on resource-constrained hardware, developing evolutionary optimization methods for converting floating-point to fixed-point arithmetic—work that enabled efficient EKF-SLAM implementations on FPGAs for embedded robotics. His combined expertise in learning from human interaction and hardware-aware algorithm design makes him a distinctive voice in making robotics both more accessible and more deployable in real-world settings.
Research Focus
Key Achievements
Top Papers
- 1Interactive Imitation Learning in Robotics: A Survey53 citations · 2022
- 2Interactive Imitation Learning of Bimanual Movement Primitives23 citations · 2023
- 3
- 4
- 5Interactive Imitation Learning in Robotics: A Survey6 citations · 2022
- 6
- 7
- 8Interactive Imitation Learning in Robotics: A Survey3 citations · 2022