Hongyi Dong
Papers
2
Total Citations
67
H-Index
2
About
Hongyi Dong is a leading researcher in indoor robotics localization, whose work bridges computer vision and probabilistic robotics to solve one of the field's most persistent challenges: reliable global localization in sparse, repetitive indoor environments. His most influential contribution, the 2019 paper "A Robust Indoor Localization System Integrating Visual Localization Aided by CNN-Based Image Retrieval with Monte Carlo Localization" (65 citations), introduces a pioneering multi-sensor fusion framework that combines deep learning-based place recognition with Monte Carlo localization. This system operates in three stages—coarse place recognition, fine localization, and re-localization—achieving robust performance where traditional methods fail. Dong's earlier work, "Global Localization Using Object Detection in Indoor Environment Based on Semantic Map" (2018), further demonstrates his innovative approach by leveraging semantic object detection to create distinctive environmental signatures, addressing the particle filter convergence problem at robot startup. His research has significantly advanced the practical deployment of autonomous mobile robots in real-world indoor settings, from warehouses to hospitals, by providing computationally efficient solutions that maintain accuracy despite environmental symmetries and visual ambiguities.
Research Focus
Key Achievements
Top Papers
- 1
- 2