Offline Navigation Dataset

该数据集是一个大规模的离线数据集，它收集了在各种户外环境中用于训练自监督导航模型的先前经验。该数据集不仅用于训练，还用于评估RECON方法及其在以前未见过的环境中发现用户指定目标的能力。该数据集包含20小时的离线数据，旨在支持在开放世界环境中的机器人探索与导航任务。This is a large-scale offline dataset that gathers prior experiences across diverse outdoor environments for training self-supervised navigation models. It is used not only for model training but also for evaluating the RECON method and its capacity to identify user-specified targets in previously unseen environments. This dataset contains 20 hours of offline data, and is designed to support robotic exploration and navigation tasks in open-world environments.