NUFR-M3F
收藏资源简介:
NUFR-M3F数据集是由东北大学创建的多模态室内SLAM数据集,涵盖了机器人可能遇到的多种挑战性场景,如无特征空间、反射表面和多层序列。数据集包含视觉、惯性和激光雷达传感器数据,支持多模态SLAM评估。创建过程中,数据通过移动机器人平台在东北大学的ISEC建筑的多层楼中收集,特别关注由于楼层布局相似性导致的感知混淆问题。该数据集旨在推动室内SLAM研究的进展,特别是在处理连续多层数据和提高SLAM算法可靠性方面。
The NUFR-M3F dataset is a multimodal indoor SLAM dataset developed by Northeastern University (China). It encompasses a wide range of challenging scenarios that robots may encounter, including featureless spaces, reflective surfaces, and multi-floor sequences. The dataset comprises visual, inertial, and LiDAR sensor data, which supports multimodal SLAM evaluation. During the data collection process, a mobile robotic platform was used to gather data across multiple floors of the ISEC building at Northeastern University (China), with special emphasis on the perceptual confusion issue arising from similar floor layouts. This dataset is intended to promote the advancement of indoor SLAM research, especially in processing continuous multi-floor data and improving the reliability of SLAM algorithms.




