VisualNav : Visually Grounded Natural Language Crawler Robot Navigation
Lakmina Gamage, Haritha Weerathunga, Vishwani Geeganage, Chinthaka Premachandra et autres
The integration of Vision-Language-Action (VLA) models into robotic systems promises to bridge the gap between high-level semantic intent and low-level control. However, deploying these computationally intensive models on resource-constrained mobile platforms while ensuring open-world generalization remains a significant challenge. This paper presents …
lk, jp (code pays fourni par la source)