Papers
9
Total Citations
140
H-Index
6
About
Riccardo Polvara is a robotics researcher whose work spans autonomous aerial systems, agricultural robotics, and long-term robot deployment in dynamic environments. He has made significant contributions to the field of UAV autonomy, most notably pioneering the application of deep reinforcement learning for sim-to-real quadrotor landing, combining sequential Deep Q-Networks with domain randomization to bridge the gap between simulation and real-world performance (38 citations). His earlier work on autonomous UAV landing using fiducial markers for ship deck operations further established his expertise in precision aerial navigation (35 citations). Polvara's research has increasingly turned toward agricultural robotics, where he has developed frameworks for human detection in field environments, multi-criteria robotic exploration strategies, and robust long-term robot localization in continuously changing outdoor settings. His contribution to the Bacchus Long-Term dataset (24 citations) represents a landmark resource for the agricultural robotics community, enabling rigorous benchmarking of autonomous platforms in vineyard environments. With work addressing challenges from gas detection planning to selective grape harvesting, Polvara demonstrates a rare breadth across perception, navigation, and autonomous decision-making, making him a versatile and impactful voice in modern field robotics research.
Research Focus
Key Achievements
Top Papers
- 1
- 2Towards autonomous landing on a moving vessel through fiducial markers35 citations · 2017
- 3
- 4
- 5
- 6
- 7
- 8A Next-Best-Smell Approach for Remote Gas Detection with a Mobile Robot4 citations · 2018
- 9