Research Projects
Most projects include a hands-on interactive demo you can run right in your browser — open a project to try it.
Micro-Localization: High-Precision Mobile Robot Localization Using Microscopic Ground Texture
This work demonstrates how microscopic images of natural ground texture can enable a mobile robot to localize with tens-of-micrometers precision. Microscope optics reveal distinctive surface details even on materials that appear relatively textureless at larger scales. Increased magnification also makes small physical displacements easier to measure, allowing precise localization without artificial markers or modifications to the surface.
Key Contributions
- A geometric analysis relating optical magnification and image-matching uncertainty to physical localization precision.
- An adaptation of Lightweight Ground Texture Localization (L-GROUT) for microscopic imagery, incorporating optical calibration, camera modeling, and autofocus.
Performance Highlights
- 100% localization success across 1,007 test queries per surface on printed wood grain, terrazzo, and concrete, with 28–73 µm translation root mean square error (RMSE) relative to a calibrated gantry reference.
- A mobile robot costing less than $600 localized 100% of queries on printed wood grain and terrazzo and 98.4% on concrete using gantry-built maps, with 21–26 µm RMSE relative to image-registration references on valid comparison queries.
- Demonstrated micrometer-scale displacement sensing by tracking the thermal expansion of an aluminum rod through repeated heating and cooling cycles.
Related Manuscript: Micro-Localization: High-Precision Mobile Robot Localization Using Microscopic Ground Texture PDF
Improved Bag-of-Words Image Retrieval for Ground Texture Localization
GT-BoW is a geometry-aware bag-of-words image-retrieval system for ground-texture localization. It uses the fixed scale and consistent orientation of downward-facing imagery to reject visually similar but geometrically inconsistent matches, improving both global localization and loop-closure detection.
Key Contributions
- An approximate k-means vocabulary with soft assignment
- Scale-binned vocabularies and orientation-based geometric verification
- Separate high-accuracy and high-speed variants for global localization and loop closure
Performance Highlights
- Increased global-localization mean average precision from 0.026 to 0.559
- Detected nearly 3× as many loop closures as a standard DBoW baseline
Related Publication: "Improved Bag-of-Words Image Retrieval with Geometric Constraints for Ground Texture Localization" (ICRA 2025) PDF
Lightweight Ground Texture Localization (L-GROUT)
L-GROUT is a lightweight global-localization system that matches natural ground texture observed by a downward-facing camera to a prebuilt map. It improves localization performance while reducing computation enough to run in real time on resource-constrained, single-board computers without GPU acceleration.
Key Contributions
- An improved method for selecting database features
- Locality-preserving dimensionality reduction that supports fast binary features
- A spatial-voting stage designed to retain accuracy in lightweight configurations
Performance Highlight
- Runs the complete localization pipeline at more than 4 Hz on a Raspberry Pi 4
L-GROUT provides millimeter-scale global localization without markers, installed infrastructure, or other environmental modifications.
Related Publication: "Lightweight Ground Texture Localization" (ICRA 2024) PDF