# Tiny-DroNeRF: Tiny Neural Radiance Fields aboard Federated Learning-enabled Nano-drones

arXiv: https://arxiv.org/abs/2603.01850 · HTML: https://mlsanity.com/paper/2603.01850/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 28 references that are in the map.
- First posted on arXiv 2 Mar 2026.

## Builds on (most cited first, 23 of 23)

- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [Instant Neural Graphics Primitives with a Multiresolution Hash Encoding](https://mlsanity.com/paper/2201.05989/) (2022)
- [Plenoxels: Radiance Fields without Neural Networks](https://mlsanity.com/paper/2112.05131/) (2021)
- [TensoRF: Tensorial Radiance Fields](https://mlsanity.com/paper/2203.09517/) (2022)
- [Mixed Precision Training](https://mlsanity.com/paper/1710.03740/) (2017)
- [K-Planes: Explicit Radiance Fields in Space, Time, and Appearance](https://mlsanity.com/paper/2301.10241/) (2023)
- [Baking Neural Radiance Fields for Real-Time View Synthesis](https://mlsanity.com/paper/2103.14645/) (2021)
- [Neuralangelo: High-Fidelity Neural Surface Reconstruction](https://mlsanity.com/paper/2306.03092/) (2023)
- [MobileNeRF: Exploiting the Polygon Rasterization Pipeline for Efficient Neural Field Rendering on Mobile Architectures](https://mlsanity.com/paper/2208.00277/) (2022)
- [Co-SLAM: Joint Coordinate and Sparse Parametric Encodings for Neural Real-Time SLAM](https://mlsanity.com/paper/2304.14377/) (2023)
- [3D Semantic Scene Completion: a Survey](https://mlsanity.com/paper/2103.07466/) (2021)
- [Binary Radiance Fields](https://mlsanity.com/paper/2306.07581/) (2023)
- [How Far Can We Compress Instant-NGP-Based NeRF?](https://mlsanity.com/paper/2406.04101/) (2024)
- [Bio-inspired Autonomous Exploration Policies with CNN-based Object Detection on Nano-drones](https://mlsanity.com/paper/2301.12175/) (2023)
- [Federated Learning for Large-Scale Scene Modeling with Neural Radiance Fields](https://mlsanity.com/paper/2309.06030/) (2023)
- [GAP9Shield: A 150GOPS AI-capable Ultra-low Power Module for Vision and Ranging Applications on Nano-drones](https://mlsanity.com/paper/2407.13706/) (2024)
- [A Sim-to-Real Deep Learning-based Framework for Autonomous Nano-drone Racing](https://mlsanity.com/paper/2312.08991/) (2023)
- [Federated Neural Radiance Fields](https://mlsanity.com/paper/2305.01163/) (2023)
- [Di-NeRF: Distributed NeRF for Collaborative Learning with Relative Pose Refinement](https://mlsanity.com/paper/2402.01485/) (2024)
- [DecentNeRFs: Decentralized Neural Radiance Fields from Crowdsourced Images](https://mlsanity.com/paper/2403.13199/) (2024)
- [Fully Onboard Low-Power Localization with Semantic Sensor Fusion on a Nano-UAV using Floor Plans](https://mlsanity.com/paper/2310.12536/) (2023)
- [Federated Multi-Agent Mapping for Planetary Exploration](https://mlsanity.com/paper/2404.02289/) (2024)
- [Multi-sensory Anti-collision Design for Autonomous Nano-swarm Exploration](https://mlsanity.com/paper/2312.13086/) (2023)

## Abstract (arXiv)

Sub-30g nano-sized aerial robots can leverage their agility and form factor to autonomously explore cluttered and narrow environments, like in industrial inspection and search and rescue missions. However, the price for their tiny size is a strong limit in their resources, i.e., sub-100 mW microcontroller units (MCUs) delivering ~100 GOps/s at best, and memory budgets well below 100 MB. Despite these strict constraints, we aim to enable complex vision-based tasks aboard nano-drones, such as dense 3D scene reconstruction: a key robotic task underlying fundamental capabilities like spatial awareness and motion planning. Top-performing 3D reconstruction methods leverage neural radiance fields (NeRF) models, which require GBs of memory and massive computation, usually delivered by high-end GPUs consuming 100s of Watts. Our work introduces Tiny-DroNeRF, a lightweight NeRF model, based on Instant-NGP, and optimized for running on a GAP9 ultra-low-power (ULP) MCU aboard our nano-drones. Then, we further empower our Tiny-DroNeRF by leveraging a collaborative federated learning scheme, which distributes the model training among multiple nano-drones. Our experimental results show a 96% reduction in Tiny-DroNeRF's memory footprint compared to Instant-NGP, with only a 5.7 dB drop in reconstruction accuracy. Finally, our federated learning scheme allows Tiny-DroNeRF to train with an amount of data otherwise impossible to keep in a single drone's memory, increasing the overall reconstruction accuracy. Ultimately, our work combines, for the first time, NeRF training on an ULP MCU with federated learning on nano-drones.