ViSA-Enhanced Aerial VLN: A Visual-Spatial Reasoning Enhanced Framework for Aerial Vision-Language Navigation
This work addresses the problem of inadequate spatial reasoning and linguistic ambiguities in aerial Vision-Language Navigation (VLN) for drone navigation systems.
Existing aerial Vision-Language Navigation (VLN) methods struggle with spatial reasoning and linguistic ambiguities. This paper proposes a Visual-Spatial Reasoning (ViSA) enhanced framework that uses structured visual prompting to enable Vision-Language Models (VLMs) to reason directly on image planes, achieving a 70.3% improvement in success rate on the CityNav benchmark.
Existing aerial Vision-Language Navigation (VLN) methods predominantly adopt a detection-and-planning pipeline, which converts open-vocabulary detections into discrete textual scene graphs. These approaches are plagued by inadequate spatial reasoning capabilities and inherent linguistic ambiguities. To address these bottlenecks, we propose a Visual-Spatial Reasoning (ViSA) enhanced framework for aerial VLN. Specifically, a triple-phase collaborative architecture is designed to leverage structured visual prompting, enabling Vision-Language Models (VLMs) to perform direct reasoning on image planes without the need for additional training or complex intermediate representations. Comprehensive evaluations on the CityNav benchmark demonstrate that the ViSA-enhanced VLN achieves a 70.3\% improvement in success rate compared to the fully trained state-of-the-art (SOTA) method, elucidating its great potential as a backbone for aerial VLN systems.