Article

DGT-Map: Directional Global Traversability Mapping with Multi-Task Learning for Heterogeneous Vehicles

DGT-Map proposes a self-supervised framework that learns global, directional, and vehicle-conditioned traversability costmaps from RGB-D observations and locomotion signals. It uses a shared multi-task backbone to learn common terrain features across vehicles, while vehicle-specific prediction heads preserve platform-dependent responses. At inference, it outputs a heading-indexed costmap for use by direction-aware planners.

Part of the event: DGT-Map: Heterogeneous Vehicle Directional Traversability Map

This page provides a summary and source link only.