patial Evolution of a road network at the UC San Diego Campus in both satellite image [6] and vectorized centerline map. The left side of the figure depicts the area prior to 2021, while the right side presents the current state of the road network. Below the satellite image, the corresponding vector map from the respective time period is displayed
Abstract
High Definition (HD) Maps have long been favored for their precise depictions of static road elements. However, their accessibility constraints and vulnerability to rapid environmental changes impede the widespread deployment of highly map-reliant autonomous driving tasks, such as motion forecasting. In this context, we propose to leverage Open-StreetMap (OSM) as a promising alternative to HD Maps for long-term motion forecasting. The contributions of this work are threefold: firstly, we extend the application of OSM to long-horizon forecasting, doubling the forecasting horizon compared to previous studies. Secondly, through an expanded observation landscape and the integration of intersection priors, our OSM-based approach exhibits competitive performance, narrowing the gap with HD-map-based models. Lastly, we conduct an exhaustive context-aware analysis, providing deeper insights in motion forecasting across diverse scenarios as well as conducting class-aware comparisons. This research not only advances long-term motion forecasting with coarse map representations but additionally offers a scalable solution within the domain of autonomous driving
High Definition (HD) Maps have long been favored for their precise depictions of static road elements. However, their accessibility constraints and vulnerability to rapid environmental changes impede the widespread deployment of highly map-reliant autonomous driving tasks, such as motion forecasting. In this context, we propose to leverage Open-StreetMap (OSM) as a promising alternative to HD Maps for long-term motion forecasting. The contributions of this work are threefold: firstly, we extend the application of OSM to long-horizon forecasting, doubling the forecasting horizon compared to previous studies. Secondly, through an expanded observation landscape and the integration of intersection priors, our OSM-based approach exhibits competitive performance, narrowing the gap with HD-map-based models. Lastly, we conduct an exhaustive context-aware analysis, providing deeper insights in motion forecasting across diverse scenarios as well as conducting class-aware comparisons. This research not only advances long-term motion forecasting with coarse map representations but additionally offers a scalable solution within the domain of autonomous driving