Davide Celestini
Papers
3
Total Citations
41
H-Index
2
About
Davide Celestini’s research lies at the intersection of model predictive control (MPC), bio-inspired path planning, and reinforcement learning, with a strong focus on advancing autonomous robotics for precision agriculture. His most impactful work, “Transformer-Based Model Predictive Control” (2024, 35 citations), introduces a novel framework that leverages sequence modeling to solve the highly non-convex trajectory optimization problems central to MPC. This approach promises to make constrained control more efficient and generalizable for real-world robot autonomy. Celestini also developed a bio-inspired complete coverage path planner (2023, 4 citations), which mimics neural activity dynamics to enable robust terrain mapping, inspection, and spraying in dynamic agricultural environments. Further, his work on generalizing reinforcement learning through artificial potential fields (2023, 2 citations) addresses key challenges in deploying unmanned ground vehicles (UGVs) for smart farming, such as improving sample efficiency and safety. Collectively, Celestini’s contributions are shaping the next generation of autonomous systems, making them more adaptive, efficient, and ready for the demands of Agriculture 4.0.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3