الملخص
Off-road traversability is direction-dependent and vehicle specific, yet most global maps assign a single isotropic cost to each location. Existing learned estimators are also commonly trained independently for each vehicle; this preserves vehicle-specific behavior but prevents vehicles from sharing common terrain representations. DGT-MAP addresses both limitations through a self-supervised framework that learns global, directional, and vehicle-conditioned traversability costmaps from RGB-D observations and locomotion signals. A shared multi-task backbone learns common terrain features across training vehicles while vehicle-specific prediction heads preserve platform-dependent responses. At inference, DGT-MAP produces a heading-indexed costmap that can be used by a direction-aware planner. We evaluate DGT-MAP in simulation by integrating it into a Hybrid A* navigation stack and measuring downstream task success on challenging terrains, including slopes that are traversable downhill but not uphill and a ridge obstacle that is traversable by some vehicles, but not by others. Across evaluated tasks, DGT-MAP achieves the highest or tied-highest navigation success rate when compared against geometric, binary, and learned direction-agnostic baselines.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Singh, J., Noori, K. K., Xiao, J., & Chamzas, C. (2026). DGT-Map: Directional Global Traversability Mapping Utilizing Multi-Task Learning for Heterogeneous Vehicles. https://omanscience.com/ar/articles/dgt-map-directional-global-traversability-mapping-utilizing-multi-task-learning-for-heterogeneous-vehicles
MLA 9
Singh, Jaskrit, et al. "DGT-Map: Directional Global Traversability Mapping Utilizing Multi-Task Learning for Heterogeneous Vehicles." https://omanscience.com/ar/articles/dgt-map-directional-global-traversability-mapping-utilizing-multi-task-learning-for-heterogeneous-vehicles.
شيكاغو (المؤلف–التاريخ)
Singh, Jaskrit, Kashif K. Noori, Jing Xiao, and Constantinos Chamzas. 2026. "DGT-Map: Directional Global Traversability Mapping Utilizing Multi-Task Learning for Heterogeneous Vehicles." https://omanscience.com/ar/articles/dgt-map-directional-global-traversability-mapping-utilizing-multi-task-learning-for-heterogeneous-vehicles.
هارفارد
Singh, J., Noori, K. K., Xiao, J. and Chamzas, C. (2026) 'DGT-Map: Directional Global Traversability Mapping Utilizing Multi-Task Learning for Heterogeneous Vehicles', Available at: https://omanscience.com/ar/articles/dgt-map-directional-global-traversability-mapping-utilizing-multi-task-learning-for-heterogeneous-vehicles.
فانكوفر
Singh J, Noori KK, Xiao J, Chamzas C. DGT-Map: Directional Global Traversability Mapping Utilizing Multi-Task Learning for Heterogeneous Vehicles. https://omanscience.com/ar/articles/dgt-map-directional-global-traversability-mapping-utilizing-multi-task-learning-for-heterogeneous-vehicles
IEEE
J. Singh, K. K. Noori, J. Xiao, and C. Chamzas, "DGT-Map: Directional Global Traversability Mapping Utilizing Multi-Task Learning for Heterogeneous Vehicles," https://omanscience.com/ar/articles/dgt-map-directional-global-traversability-mapping-utilizing-multi-task-learning-for-heterogeneous-vehicles.