# Evaluating Global Geo-Alignment for Precision Learned Autonomous Vehicle Localization Using Aerial Data

> **Draft note**: This page was auto-generated from the ICRA 2025 paper list abstract and public arXiv metadata. Human review is still needed.

| Item | Value |
| --- | --- |
| Session | Novel Methods for Mapping and Localization |
| arXiv | [http://arxiv.org/abs/2503.13896](http://arxiv.org/abs/2503.13896) |
| Authors | Yang, Yi;Zhao, Xuran;Zhao, Haicheng Charles;Yuan, Shumin;Bateman, Samuel;Huang, Tiffany A.;Beall, Chris;Maddern, Will |
| Source | ICRA 2025 / arXiv |

## TL;DR

- Recently there has been growing interest in the use of aerial and satellite map data for autonomous vehicles, primarily due to its potential for significant cost reduction and enhanced scalability.
- Despite the advantages, aerial data also comes with challenges such as a sensor-modality gap and a viewpoint difference gap.
- Learned localization methods have shown promise for overcoming these challenges to provide precise metric localization for autonomous vehicles.

## Task

* Localization
* Aerial Robotics
* Software Tools

## Keywords

* Localization
* Mapping
* Intelligent Transportation Systems

## AI依存度

* Hybrid

## 何を解決？

* Recently there has been growing interest in the use of aerial and satellite map data for autonomous vehicles, primarily due to its potential for significant cost reduction and enhanced scalability.

## 何が新しい？

* Recently there has been growing interest in the use of aerial and satellite map data for autonomous vehicles, primarily due to its potential for significant cost reduction and enhanced scalability.

## どうやってる？

* Learned localization methods have shown promise for overcoming these challenges to provide precise metric localization for autonomous vehicles.

## どこが強い？

* Finally, we evaluate a learned localization system using the data alignment methods on a comprehensive (1600km) autonomous vehicle dataset and demonstrate localization error below 0.3m and 0.5�?sufficient for autonomous vehicle applications.

## 弱そうなところ

* abstract ベースの初稿。前提モデル、計算量、失敗ケース、パラメータ感度は本文確認が必要。

## 関連研究との違い

* Recently there has been growing interest in the use of aerial and satellite map data for autonomous vehicles, primarily due to its potential for significant cost reduction and enhanced scalability.

## non-AIとして見る価値

* 学習要素は含むが、周辺のシステム設計や問題設定の切り方には実装上の学びがある。

## 難易度

★★★★★（abstract 初見ベース）

## 自分の理解/感想

* 初見では、学習と従来手法のつなぎ方を見る材料として良さそう。