Daniele Cattaneo

Papers

Total Citations

H-Index

About

Daniele Cattaneo is a robotics and computer vision researcher whose work sits at the intersection of continual learning and autonomous navigation. His research addresses one of the most pressing challenges in deploying learning-based systems in the real world: the inability of neural networks to generalize across diverse, unseen environments without catastrophic forgetting. Cattaneo's most notable contributions center on developing adaptive frameworks for core navigation tasks. His work on CoVIO (2023, 28 combined citations) pioneered online continual learning for Visual-Inertial Odometry, enabling mobile and robotic systems to continuously adapt to new environments without losing previously acquired knowledge — a critical capability for real-world deployment. Complementing this, his Continual SLAM framework (2022, 10 citations) extended these principles to Simultaneous Localization and Mapping, directly tackling the domain gap that hampers learning-based SLAM methods when robots venture into unfamiliar terrain. Collectively, Cattaneo's research pushes the boundary of lifelong autonomous systems, offering practical solutions for robots that must operate reliably across ever-changing conditions. His work is increasingly influential among researchers building the next generation of adaptive, deployment-ready robotic platforms.