# InfraVLA: Extending Vision-Language-Action Navigation with Infrastructure Cameras

arXiv: https://arxiv.org/abs/2609.33647 · HTML: https://mlsanity.com/paper/2609.33647/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Generalist VLA models and benchmarks, one of 13 directions in Vision-language-action models.
- Builds on 27 of its 31 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Placed in

- Vision-language-action models / [Generalist VLA models and benchmarks](https://mlsanity.com/vla/generalist-vla-models-and-benchmarks/)

## Builds on (most cited first, 27 of 27)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [End-to-end Autonomous Driving: Challenges and Frontiers](https://mlsanity.com/paper/2306.16927/) (2023)
- [HG-DAgger: Interactive Imitation Learning with Human Experts](https://mlsanity.com/paper/1810.02890/) (2018)
- [CoBEVT: Cooperative Bird's Eye View Semantic Segmentation with Sparse Transformers](https://mlsanity.com/paper/2207.02202/) (2022)
- [NaVILA: Legged Robot Vision-Language-Action Model for Navigation](https://mlsanity.com/paper/2412.04453/) (2024)
- [DART: Noise Injection for Robust Imitation Learning](https://mlsanity.com/paper/1703.09327/) (2017)
- [PointVLA: Injecting the 3D World into Vision-Language-Action Models](https://mlsanity.com/paper/2503.07511/) (2025)
- [ForceVLA: Enhancing VLA Models with a Force-aware MoE for Contact-rich Manipulation](https://mlsanity.com/paper/2505.22159/) (2025)
- [Fighting Copycat Agents in Behavioral Cloning from Observation Histories](https://mlsanity.com/paper/2010.14876/) (2020)
- [NavDP: Learning Sim-to-Real Navigation Diffusion Policy with Privileged Information Guidance](https://mlsanity.com/paper/2505.08712/) (2025)
- [V2X-VLM: End-to-End V2X Cooperative Autonomous Driving Through Large Vision-Language Models](https://mlsanity.com/paper/2408.09251/) (2024)
- [End-to-End Autonomous Driving through V2X Cooperation](https://mlsanity.com/paper/2404.00717/) (2024)
- [OmniVLA: An Omni-Modal Vision-Language-Action Model for Robot Navigation](https://mlsanity.com/paper/2509.19480/) (2025)
- [CAST: Counterfactual Labels Improve Instruction Following in Vision-Language-Action Models](https://mlsanity.com/paper/2508.13446/) (2025)
- [Keyframe-Focused Visual Imitation Learning](https://mlsanity.com/paper/2106.06452/) (2021)
- [Efficient Exploration in Constrained Environments with Goal-Oriented Reference Path](https://mlsanity.com/paper/2003.01641/) (2020)
- [Robot-Relay : Building-Wide, Calibration-Less Visual Servoing with Learned Sensor Handover Network](https://mlsanity.com/paper/2310.15677/) (2023)
- [E-VLA: Event-Augmented Vision-Language-Action Model for Dark and Blurred Scenes](https://mlsanity.com/paper/2604.04834/) (2026)
- [MM-Nav: Multi-View VLA Model for Robust Visual Navigation via Multi-Expert Learning](https://mlsanity.com/paper/2510.03142/) (2025)
- [IndoorR2X: Indoor Robot-to-Everything Coordination with LLM-Driven Planning](https://mlsanity.com/paper/2603.20182/) (2026)
- [Anticipating Human Behavior for Safe Navigation and Efficient Collaborative Manipulation with Mobile Service Robots](https://mlsanity.com/paper/2410.05015/) (2024)
- [SurveilNav: Collaborative Object Goal Navigation with Robot and Surveillance System](https://mlsanity.com/paper/2606.25119/) (2026)

## Abstract (arXiv)

Many indoor environments in which robots operate, such as warehouses, offices, and hospitals, already have cameras installed. They observe parts of the building that the robot cannot see from where it stands, yet navigation policies, including recent vision-language-action (VLA) models, do not use them. We propose InfraVLA, an end-to-end method that adapts a pretrained navigation VLA to such static infrastructure views: a closed-circuit television (CCTV) encoder turns each external view into tokens of the input sequence. Because the views matter only at rare decision points, fine-tuning alone did not make the policy use them in our experiments; we therefore train in two stages, on demonstrations with upsampled counterfactual data and then on recovery data. We evaluate on two simulated warehouse tasks, finding an object named in the instruction and rerouting around blocked aisles, where the deciding information is often visible only to the infrastructure cameras. Tested in distribution, InfraVLA reached a success rate of 100% on both, against 34.0% and 73.6% for a baseline without CCTV input. On out-of-distribution test sets it reached 88.2% and 88.9%. On a real quadruped fine-tuned with under 10 minutes of demonstrations, the policy reached 83.3% against 29.2% for the on-board-only baseline.