Max Robotics — Global Service Robot Directory + US Certification

About

Eugenio Chisari is a robotics researcher whose work spans interactive imitation learning, robotic manipulation, and autonomous systems. He has emerged as a notable voice in the field of Interactive Imitation Learning (IIL), co-authoring a widely referenced survey on the topic that has accumulated over 60 citations across multiple venues, establishing a comprehensive foundation for how human feedback can be leveraged during robot execution to enable online behavioral improvement. His 2022 paper "Correct Me If I am Wrong" (37 citations) addresses the practical challenge of sample efficiency in deep reinforcement learning for manipulation, proposing an interactive framework that reduces reliance on costly trial-and-error. More recently, his CenterGrasp framework (2024, 19 citations) advances 6-DoF grasp estimation by combining object-aware implicit representations with shape reconstruction, pushing beyond traditional clutter-removal approaches. His contribution to the AMZ Driverless autonomous racing system further demonstrates his breadth, tackling real-world perception and control at high speeds. Through his focus on Bayesian scene representations and compact visual policies, Chisari consistently addresses sample efficiency and generalization — two of robotics' most pressing challenges — making his work highly relevant for researchers bridging learning theory and real-world deployment.