README.md

---
title: README
emoji: 👁
colorFrom: yellow
colorTo: indigo
sdk: static
pinned: false
license: mit
---

<h1 align="center">Towards Capability-Aware Traversability Navigation for Unstructured
Environments</h1>

<p align="center">
  <b>Gianluca Capezzuto, Felipe Tommaselli, Matheus P. Angarola, Ricardo V. Godoy, and Marcelo Becker</b><br>
  <i>Mobile Robotics Group, University of São Paulo, São Carlos, SP, Brazil</i>
</p>

<p align="center">
  🎉 <b>Accepted at IROS 2026</b> 🎉
</p>

<p align="center">
  <a href="https://arxiv.org/abs/xxxx.xxxxx"><strong>📄 Paper</strong></a>
  &nbsp;&nbsp;&nbsp;&nbsp;|&nbsp;&nbsp;&nbsp;&nbsp;
  <a href="https://cat-iros-2026.github.io"><strong>🌐 Website</strong></a>
  &nbsp;&nbsp;&nbsp;&nbsp;|&nbsp;&nbsp;&nbsp;&nbsp;
  <a href="https://huggingface.co/datasets/CAT-IROS-2026/data_cat_iros_2026"><strong>🤗 Dataset</strong></a>
</p>

<hr>

<p align=justify>This repository serves as the central data hub for the <b>Capability-Aware Traversability (CAT)</b> framework, 
specifically organizing the <b>datasets and pre-trained</b> models utilized in our research. It does not contain 
the source code implementation; rather, it hosts the dense, trajectory-grounded supervision masks collected 
from multiple robot embodiments, alongside the final model weights needed to reproduce our capability-conditioned 
traversability predictions.</p>

<hr>