Lorenzo Moretti
Papers
1
Total Citations
2
H-Index
1
About
Lorenzo Moretti is a leading researcher in humanoid robotics, specializing in the intersection of model-based control and deep learning for dynamic locomotion. His work focuses on enabling robots to achieve stylistic, adaptive walking—a critical step toward deploying humanoids in unstructured human environments. Moretti’s most-cited paper, "Online DNN-driven Nonlinear MPC for Stylistic Humanoid Robot Walking with Step Adjustment" (2024), introduces a novel three-layered architecture that seamlessly integrates an autoregressive Deep Neural Network for trajectory generation with a Nonlinear Model Predictive Controller for real-time step adjustment. This approach allows a humanoid to not only walk with distinct styles but also to dynamically adjust its foot placement online, a breakthrough for robust locomotion. With 2 citations in its first year, this work is already influencing the field. Moretti’s contributions bridge the gap between data-driven trajectory planning and rigorous model-based control, offering a scalable framework for expressive, safe, and agile humanoid movement—a key enabler for future service and assistive robots.
Research Focus
Key Achievements
Top Papers
- 1