FLAG: Feature-based Localization between Air and Ground

In GPS-denied environments, robot systems typically revert to navigating with dead-reckoning and relative mapping, accumulating error in their global pose estimate. In this paper, we propose Feature-based Localization between Air and Ground (FLAG), a method for computing global position updates by matching features observed from ground to features in an aerial image. Our method uses stable, descriptorless features associated with vertical structure in the environment around a ground robot in previously unmapped areas, referencing only overhead imagery, without GPS. Multiple-hypothesis data association with a particle filter enables efficient recovery from data association error and odometry uncertainty. We implement a stereo system to demonstrate our vertical feature based global positioning approach in both indoor and outdoor scenarios, and show comparable performance to laser-scan-matching results in both environments.

Authors

• Xipeng Wang
• Steve Vozar
• Edwin Olson

Keywords

• Image edge detection
• Feature extraction
• Three-dimensional displays
• Cameras
• Robot kinematics
• Buildings
• Image Features
• Global Positioning System
• Aerial Images
• Vertical Structure
• Particle Filter
• Feature Matching
• Vertical Features
• Ground Robots
• Local Features
• Satellite Images
• Global Features
• Point Cloud
• Local Method
• Indoor Environments
• Robust Features
• Local Solution
• Changes In Appearance
• 3D Point Cloud
• Edge Features
• Simultaneous Localization And Mapping
• Stereo Camera
• Speeded Up Robust Features
• Stereo Images
• Stereo Image Pairs
• Factor Graph
• Scale-invariant Feature Transform
• Priority Map
• Robot Pose
• Robot Localization
• Grid Position