Liaohe-CobotMagic-PnP
收藏资源简介:
Liaohe-CobotMagic-PnP数据集是一个针对工业应用的智能机器人模仿学习数据集,旨在解决工业环境下动态环境干扰引起的非线性交互问题。该数据集融合了多维度干扰特征,包括尺寸、颜色和光照变化,并使用高精度传感器同步收集视觉、扭矩和关节状态测量数据。数据集包括超过85%的几何相似度和标准化光照梯度的场景,以确保现实世界的代表性。通过ROS系统实现的微秒级时间同步和抗振数据采集协议保证了时间和操作的真实性。实验结果表明,该数据集提高了模型验证的鲁棒性,并提高了动态、干扰丰富的环境中的机器人操作稳定性。
The Liaohe-CobotMagic-PnP dataset is an intelligent robotic imitation learning dataset designed for industrial applications, aiming to address nonlinear interaction issues caused by dynamic environmental disturbances in industrial settings. This dataset integrates multi-dimensional disturbance features including variations in size, color, and illumination, and synchronously collects visual, torque, and joint state measurement data using high-precision sensors. The dataset includes scenarios with over 85% geometric similarity and standardized illumination gradients to ensure real-world representativeness. The microsecond-level time synchronization and vibration-resistant data acquisition protocol implemented via the ROS system ensure the temporal and operational authenticity of the collected data. Experimental results demonstrate that this dataset enhances the robustness of model validation and improves the operational stability of robots in dynamic, disturbance-rich environments.




