Efficient Localisation Using Images and OpenStreetMaps

The ability to localise is key for robot navigation. We describe an efficient method for vision-based localisation, which combines sequential Monte Carlo tracking with matching ground-level images to 2-D cartographic maps such as OpenStreetMaps. The matching is based on a learned embedded space representation linking images and map tiles, encoding the common semantic information present in both and providing potential for invariance to changing conditions. Moreover, the compactness of 2-D maps supports scalability. This contrasts with the majority of previous approaches based on matching with single-shot geo-referenced images or 3-D reconstructions. We present experiments using the StreetLearn and Oxford RobotCar datasets and demonstrate that the method is highly effective, giving high accuracy and fast convergence.

Authors

• Mengjie Zhou
• Xieyuanli Chen
• Noe Samano
• Cyrill Stachniss
• Andrew Calway

Keywords

• Three-dimensional displays
• Monte Carlo methods
• Navigation
• Scalability
• Roads
• Urban areas
• Semantics
• 3D Reconstruction
• Faster Convergence
• Major Approaches
• Latent Space
• Exhaustive Search
• Autonomous Vehicles
• Semantic Features
• Local Map
• Motion Model
• Particle Filter
• Convergence Time
• Yaw Angle
• Map Representation
• Discrete Locations
• Plan View
• Local Grid
• Breadth-first Search
• Panoramic Images
• Visual Odometry
• Complex Routes
• Aerial Spraying
• Tracking Framework
• Wall Street
• Variety Of Sensors
• OpenStreetMap Data
• Regular Grid
• Position Estimation
• Frontal Direction