DIDLM - 一个综合的多传感器SLAM数据集
收藏资源简介:
DIDLM是一个综合的多传感器数据集,旨在解决在具有挑战性的室内和室外环境中进行SLAM(Simultaneous Localization and Mapping,即同时定位与建图)时的数据稀缺问题。该数据集融合了红外摄像头、深度摄像头、激光雷达和4D毫米波雷达等多种传感器数据,以增强在极端条件下的感知能力,如雨雪天气和不平坦路面。该数据集由地面机器人和自动驾驶车辆在多变的自然与路面条件下采集,并对比了多种SLAM算法的真实性能。DIDLM首次将4D毫米波雷达与六种传感器结合,收集了大量极端条件下的数据,对机器人学和自动驾驶车辆研究具有显著价值,为SLAM技术的研究与优化提供了宝贵资源。
DIDLM is a comprehensive multi-sensor dataset designed to address the scarcity of data for SLAM (Simultaneous Localization and Mapping) in challenging indoor and outdoor environments. This dataset integrates data from various sensors including infrared cameras, depth cameras, LiDAR, and 4D millimeter-wave radar to enhance perception capabilities under extreme conditions such as rain, snow, and uneven terrain. Collected by ground robots and autonomous vehicles under varying natural and road conditions, the dataset compares the real-world performance of multiple SLAM algorithms. DIDLM is the first to combine 4D millimeter-wave radar with six other sensors, gathering extensive data under extreme conditions, which is of significant value to robotics and autonomous vehicle research, providing a valuable resource for the study and optimization of SLAM technology.
数据集概述
目前数据集详情页面未提供任何与数据集相关的信息。




