About

Cindy Cappelle is a researcher specializing in autonomous systems, robotic localization, and fault-tolerant cooperative navigation. Her work sits at the intersection of computer vision, multi-robot systems, and machine learning, addressing two critical challenges in modern robotics: reliable visual localization and the safety of collaborative positioning systems. Cappelle's early contributions focused on visual localization, most notably her 2019 paper combining Convolutional Networks with Locality-Sensitive Hashing for sequence-based place recognition, which has garnered 19 citations and demonstrated the power of deep image representations in real-world robotic navigation. This built upon her foundational 2016 work on multi-feature sequence matching for visual localization. More recently, her research has pivoted toward fault-tolerant cooperative localization for multi-robot systems, a domain where safety and integrity are paramount. She has explored diverse diagnostic approaches — from Jensen-Shannon divergence and decision trees to federated learning — to detect and exclude sensor faults in decentralized systems. Her comparative analyses of centralized versus federated learning architectures reflect a sophisticated understanding of scalable, safe robotics deployment. With a growing publication record spanning perception, probabilistic reasoning, and autonomous systems integrity, Cappelle represents an important voice in making multi-robot navigation both reliable and practically deployable.

Research Focus

Key Achievements

3
H-Index
7
Papers
39
Total Citations
6
Avg Citations/Paper
🏆 Most Cited Paper
ConvNet and LSH-Based Visual Localization Using Localized Sequence Matching
19 citations · 2019
📈 Most Prolific Year: 2023 (3 Papers)
🤝 Key Collaborators: 6
🏛 Institutions: Université de Bourgogne, École Centrale de Lille, Centre National de la Recherche Scientifique, Université Bourgogne Franche-Comté

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago