TE-NeXt：基于LiDAR的3D稀疏卷积网络的可遍历性估计
摘要
本文提出了TE-NeXt，这是一个 novel and efficient architecture for Traversability Estimation(TE) from sparse LiDAR point clouds based on a residual convolution block. TE-NeXt block fuses notions of current trends such as attention mechanisms and 3D sparse convolutions. TE-NeXt aims to demonstrate high capacity for generalisation in a variety of urban and natural environments, using well-known and accessible datasets such as SemanticKITTI, Rellis-3D and SemanticUSL. Thus, the designed architecture ouperforms state-of-the-art methods in the problem of semantic segmentation, demonstrating better results in unstructured environments and maintaining high reliability and robustness in urbans environments, which leads to better abstraction. Implementation is available in a open repository to the scientific community with the aim of ensuring the reproducibility of results. 本文提出了TE-NeXt，这是一个基于残差卷积块的用于从稀疏LiDAR点云进行可遍历性估计(TE)的 novel and efficient architecture。TE-NeXt模块融合了注意力机制和3D稀疏卷积等当前趋势。TE-NeXt旨在在各种城市和自然环境中展现出高泛化能力，利用诸如SemanticKITTI、Rellis-3D和SemanticUSL等著名且可接触的数据集。因此，所设计的架构在语义分割问题上超越了最先进的方法，在非结构化环境中表现更好，在城市环境中保持高可靠性和鲁棒性，从而实现更好的抽象。该实现已公开于一个开放仓库，供科学界使用，以确保结果的可重复性。
引用
@article{arxiv.2406.01395,
title = {TE-NeXt: A LiDAR-Based 3D Sparse Convolutional Network for Traversability Estimation},
author = {Antonio Santo and Juan J. Cabrera and David Valiente and Carlos Viegas and Arturo Gil},
journal= {arXiv preprint arXiv:2406.01395},
year = {2025}
}