Deliberate Before You Fly: Vision-Guided Spatial Deliberation for UAV See-and-Reach Navigation

2026-08-05

AI Summary

This work addresses the challenges of semantic-control misalignment, action inconsistency, and unreliable termination in language-guided drone navigation within target-visible environments. To this end, the authors propose DBFly, a novel framework that introduces an explicit vision-guided spatial reasoning chain prior to waypoint generation. This chain comprises target-direction anchoring, spatial diagnosis, and maneuver decision-making, complemented by implicit flight corridor modeling and a terminal convergence-aware stopping strategy to reliably bridge high-level linguistic intent with continuous low-level control. Experimental results demonstrate that DBFly achieves a 25.07 percentage point improvement in average success rate over the strongest baseline across both seen and unseen objects and scenes, significantly enhancing navigation stability and reliability.

Abstract

UAV see-and-reach navigation requires an aerial agent to approach a language-specified target visible in its initial view and stop reliably near it. Existing methods typically map vision-language representations directly to action outputs without explicitly modeling intermediate fine-grained spatial decisions. This direct mapping causes semantic-control misalignment, leading to inconsistent maneuvers and unreliable termination. To address this issue, we propose DBFly, a vision-language waypoint prediction framework that introduces explicit vision-guided spatial deliberation before waypoint generation. Specifically, DBFly introduces a spatial maneuver decision chain that progressively performs target-direction anchoring, spatial diagnosis, and maneuver decision, enabling high-level maneuver intent to explicitly guide continuous waypoint generation. DBFly further constructs an implicit flight corridor by transforming the initial target-direction prior into a persistent geometric reference and deriving an online corridor state from the UAV's current position, thereby providing soft geometric guidance for spatial diagnosis and maneuver correction. In addition, DBFly develops a terminal-convergence-aware stopping strategy that characterizes terminal states through both target proximity and short-horizon motion convergence, enabling more reliable stopping near the target. Extensive experiments across seen, unseen-object, and unseen-scene test sets demonstrate that DBFly improves the success rate over the SOTA baseline by an average of 25.07 percentage points. The project homepage is available at https://xuefanfu.github.io/DBFly-Page.

Problem

• UAV navigation
• see-and-reach
• vision-language grounding
• spatial decision-making
• semantic-control misalignment

Innovation

• vision-language navigation
• spatial deliberation
• waypoint prediction
• flight corridor
• terminal convergence