Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note3 min read

VoLN: Vision-Only Long-Horizon Navigation---Paradigm, Benchmark, and Method

Source

arXiv

arxiv.org/abs/2607.21400v1

Read the source

Source

Plain-Language Brief

The text discusses a new navigation system called Vision-Only Long-Horizon Navigation (VoLN) for autonomous agents. In this system, the agent navigates using visual cues instead of relying on route-level instructions or GPS. The authors present VoLN-UAV, a benchmark for testing this system in five environments. They also introduce VoLN-MLLM as an initial baseline model that uses self-supervised learning to predict waypoints based on the agent's observations, goal views, and visual-semantic tokens. The results show that the model performs poorly, achieving success rates of 7.4%, 4.5%, and 1.8% on Easy, Normal, and Hard episodes respectively.

What the Document Says

  • VLN is a navigation system for embodied agents to follow natural-language instructions.
  • VoLN is a new formulation that shifts route-relevant information from external instructions to locally observable in-scene cues.
  • Goal views specify the destination in VoLN.
  • Route-relevant information is available only through locally observable in-scene cues.
  • VoLN-UAV is a benchmark for aerial navigation, consisting of 7,210 episodes.
  • The benchmark includes long-horizon goal-directed flight, continuous 3D motion, large viewpoint changes, and context-dependent beacon selection.
  • VoLN-MLLM aligns self-supervised visual features with a structured semantic space.
  • It predicts short-horizon waypoint segments from observation history, goal views, retrieved visual--semantic tokens, and proprioception.
  • On the Test-Unseen split, VoLN-MLLM obtains success rates of 7.4%, 4.5%, and 1.8% on Easy, Normal, and Hard episodes respectively.
  • The results reveal challenges in long-horizon evidence integration, cross-view goal matching, and closed-loop stability.

Key Concepts Mentioned

  • Vision-and-Language Navigation (VLN)
  • Vision-Only Long-Horizon Navigation (VoLN)
  • VoLN-UAV
  • benchmark
  • aerial navigation
  • self-supervised learning
  • waypoint segments
  • goal views
  • visual--semantic tokens
  • proprioception

Claims Made by the Document

  • VoLN provides an alternative to VLN that does not rely on externally supplied instructions and global guidance.
  • VoLN-UAV is a benchmark for testing the performance of VoLN.

Evidence Used

  • The text describes VoLN as shifting route-relevant information from external sources to local observations.
  • The description of VoLN-UAV indicates it is a benchmark designed for testing VoLN.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Vision-and-Language Navigation (VLN) enables embodied agents to follow natural-language instructions. However, route-level instructions commonly encode spatial priors, such as orientation, distance, and layout, that are not explicitly available from onboard sensing at deployment in open, GPS-denied environments. Benchmark performance under such interfaces therefore jointly reflects visual navigation ability and the use of route structure explicitly supplied by the task description. As a complementary formulation, we propose Vision-Only Long-Horizon Navigation (VoLN), which shifts route-relevant information from externally supplied instructions and global guidance to locally observable in-scene cues. In VoLN, goal views specify the destination, while route-relevant information is available only through locally observable in-scene cues that the agent must detect, interpret, and select online. We instantiate VoLN for aerial navigation through VoLN-UAV, a 7,210-episode benchmark that combines long-horizon goal-directed flight, continuous 3D motion, large viewpoint changes, and context-dependent beacon selection. We further provide VoLN-MLLM as an initial reference baseline. It aligns self-supervised visual features with a structured semantic space and predicts short-horizon waypoint segments from observation history, goal views, retrieved visual--semantic tokens, and proprioception. On the five-environment Test-Unseen split, it obtains success rates of 7.4%, 4.5%, and 1.8% on Easy, Normal, and Hard episodes, respectively. These results provide an initial evaluation of VoLN and reveal substantial remaining challenges in long-horizon evidence integration, cross-view goal matching, and closed-loop stability. Project page: https://admire-ljb.github.io/VoLN-UAV/