室外社区视觉导航避障数据

面向基于视觉的机器人环境建模与定位导航基础研究，为突破复杂环境下机器人视觉导航避障技术瓶颈，基于双目RGB彩色相机，在中科院奥运村园区（场地占地面积约40000平方米，存在动态目标，光线条件较为复杂，以自然光和人工光照混合条件为主）进行数据采集。该数据集包含1880张彩色影像（分辨率为640×360像素,采样频率5Hz），数据量为658M，主要构建复杂场景下的社区视觉导航避障据集。This dataset is developed for basic research on vision-based robotic environmental modeling, localization and navigation, aiming to break through the technical bottlenecks of robotic visual navigation and obstacle avoidance in complex environments. Data collection was carried out using a binocular RGB color camera at the Olympic Village Campus of the Chinese Academy of Sciences, which covers an area of approximately 40,000 square meters, has dynamic targets, and features complex lighting conditions mainly composed of mixed natural and artificial lighting. This dataset contains 1880 color images with a resolution of 640×360 pixels and a sampling frequency of 5 Hz, with a total data volume of 658 MB, and is primarily constructed for visual navigation and obstacle avoidance in complex community scenarios.

搜集汇总

背景与挑战

背景概述

该数据集为室外社区视觉导航避障数据，包含1880张彩色影像（分辨率为640×360像素），数据量为658MB，主要用于复杂场景下的社区视觉导航避障研究。数据采集于中科院奥运村园区，环境复杂，包含动态目标和混合光照条件。