Exploring Navigation Maps for Learning-Based Motion Prediction

The prediction of surrounding agents' motion is a key for safe autonomous driving. In this paper, we explore navigation maps as an alternative to the predominant High Definition (HD) maps for learning-based motion prediction. Navigation maps provide topological and geometrical information on road-level, HD maps additionally have centimeter-accurate lane-level information. As a result, HD maps are costly and time-consuming to obtain, while navigation maps with near-global coverage are freely available. We describe an approach to integrate navigation maps into learning-based motion prediction models. To exploit locally available HD maps during training, we additionally propose a model-agnostic method for knowledge distillation. In experiments on the publicly available Argoverse dataset with navigation maps obtained from OpenStreetMap, our approach shows a significant improvement over not using a map at all. Combined with our method for knowledge distillation, we achieve results that are close to the original HD map-reliant models. Our publicly available navigation map API for Argoverse enables researchers to develop and evaluate their own approaches using navigation maps 4.

Authors

• Julian Schmidt
• Julian Jordan
• Franz Gritschneder
• Thomas Monninger
• Klaus Dietmayer

Keywords

• Training
• Automation
• Navigation
• Source coding
• Predictive models
• Cognition
• Trajectory
• Motion Prediction
• Navigation Map
• Prediction Model
• Geometric Information
• Learning-based Models
• High Definition
• OpenStreetMap
• Convolutional Neural Network
• Coordinate System
• Coordination Sphere
• Crowdsourcing
• Local Coordinate
• Map Information
• Student Model
• Road Segments
• Local Coordinate System
• Information Of Agents
• Trajectory Prediction
• Latent Embedding
• HSV Color
• Original Variant
• Improve Prediction Performance
• LiDAR Scans
• Global Coordinates
• Vehicle Motion
• Universal Transverse Mercator
• Distillation Loss
• Neural Network