# Flying in Clutter on Monocular RGB by Learning in 3D Radiance Fields with Domain Adaptation

arXiv: https://arxiv.org/abs/2512.17349 · HTML: https://mlsanity.com/paper/2512.17349/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Dec 2025.

## Built on by (most cited first, 5 of 5)

- [NavDreamer: Video Models as Zero-Shot 3D Navigators](https://mlsanity.com/paper/2602.09765/) (2026)
- [Zero-Shot UAV Navigation in Forests via Relightable 3D Gaussian Splatting](https://mlsanity.com/paper/2602.07101/) (2026)
- [GaussFly: Contrastive Reinforcement Learning for Visuomotor Policies in 3D Gaussian Fields](https://mlsanity.com/paper/2604.05062/) (2026)
- [Efficient Transceiver Design for Aerial Image Transmission and Large-scale Scene Reconstruction](https://mlsanity.com/paper/2604.11098/) (2026)
- [ParkingWorld: End-to-End Autonomous Parking Reinforcement Learning from Corrective Experience in 3DGS Simulation](https://mlsanity.com/paper/2605.25029/) (2026)

## Abstract (arXiv)

Modern autonomous navigation systems predominantly rely on lidar and depth cameras. However, a fundamental question remains: Can flying robots navigate in clutter using solely monocular RGB images? Given the prohibitive costs of real-world data collection, learning policies in simulation offers a promising path. Yet, deploying such policies directly in the physical world is hindered by the significant sim-to-real perception gap. Thus, we propose a framework that couples the photorealism of 3D Gaussian Splatting (3DGS) environments with Adversarial Domain Adaptation. By training in high-fidelity simulation while explicitly minimizing feature discrepancy, our method ensures the policy relies on domain-invariant cues. Experimental results demonstrate that our policy achieves robust zero-shot transfer to the physical world, enabling safe and agile flight in unstructured environments with varying illumination.