Romain Drouilly

Institut national de recherche en sciences et technologies du numérique, Centre Inria d'Université Côte d'Azur

Papers

Total Citations

H-Index

About

Romain Drouilly’s research lies at the intersection of autonomous navigation, semantic mapping, and dynamic environment understanding. His most influential work, “Semantic representation for navigation in large-scale environments” (2015, 25 citations), tackles the grand challenge of enabling robots to navigate like humans—moving beyond simple geometric maps to incorporate high-level semantic interpretations of their surroundings. This paper demonstrates how robots can infer routes within a global map by leveraging semantic cues, a crucial step toward more intelligent and adaptable autonomous systems. In his earlier work, “Local map extrapolation in dynamic environments” (2014, 3 citations), Drouilly introduced a generative mapping framework that intelligently integrates static and dynamic entity classes. This approach allows robots to extrapolate map information across different resolutions, moving beyond conventional mapping paradigms that treat all environmental elements uniformly. Drouilly’s contributions are particularly valuable for robotics researchers working on long-term autonomy in changing environments, as his semantic framework provides a foundation for robots that can understand not just where they are, but what they are seeing. His work continues to influence the development of more perceptive and context-aware navigation systems.