aswinkumar99/LeRobot-SO101-task3-single-sponge-with-distractors-fixed-locations

这是一个真实世界的SO101拾取放置数据集，针对单个海绵且场景中存在干扰物的情况。该数据集使用了6个预定义位置的固定工作区布局。数据集结构包括数据、元数据和视频文件夹。任务为拾取放置单个海绵，场景中存在干扰物。分割名称为固定位置，之前称为已知，工作区布局为6个固定位置。元数据包括使用的机器人、集数、帧数、帧率、摄像头和光照条件。数据集旨在用于模仿学习、行为克隆、策略评估和布局变化下的鲁棒性研究。限制包括仅包含一个任务变体、仅使用海绵对象、摄像头位置固定以及光照仅限于收集的早晨和夜晚设置。创建者为Aswinkumar。

Real-world SO101 pick-and-place dataset for a single sponge with distractor objects present. This split uses fixed workspace layouts drawn from 6 predefined locations. Dataset structure includes data, meta, and videos folders. The task is pick-and-place for a single sponge with distractor objects present in the scene. Split name is fixed-locations, previously known, with workspace layouts of 6 fixed locations. Metadata includes the robot used, number of episodes, frames, fps, cameras, and lighting conditions. The dataset is intended for imitation learning, behavior cloning, policy evaluation, and robustness studies under layout changes. Limitations include only one task variant, only sponge objects, fixed camera placement, and lighting limited to collected morning and night settings. Creator is Aswinkumar.