Learning Singularity Avoidance - Data using a real world 7 link sawyer robot and simulated 3 link planar system

This dataset contains both real world and simulated data. The real world data consists of 50 trajectories of 7DOF Jointspace data recorded kinaesthetically using the Sawyer robot. Each demonstration has 1) a task space component where the sawyer moves from a starting position to anywhere on a drawer and 2) a null space component where the system is closing the drawer. The constraint in the null space limits movement of the system to the x-axis. These can be read as text files. The simulated data consists of 3 sets of 50 trajectories of 3DOF Jointspace data in a planar system. In each set the system is constrained by one of the 3 constraints: 1) X and Y, 2) X and Theta and 3) Y and Theta. These are stored as matlab data.For further details please refer to the paper: Learning Singularity Avoidance Manavalan, J. & Howard, M. J. W., 2019, IEEE/RSJ International Conference on Intelligent Robots and Systems.

本数据集涵盖真实世界与仿真两类实验数据。真实世界数据包含50条7自由度关节空间（7DOF Jointspace）轨迹数据，采用Sawyer机器人通过动觉示教方式采集得到。每条演示数据包含两个分量：1）任务空间分量：Sawyer机器人从初始位置移动至抽屉表面的任意位置；2）零空间分量：系统执行抽屉闭合动作。该零空间约束将系统运动限制于X轴方向。此类数据可通过文本文件读取。仿真数据包含3组各50条平面系统下的3自由度关节空间（3DOF Jointspace）轨迹数据。每组数据分别受3种约束之一限制：1）X轴与Y轴约束；2）X轴与角度（Theta）约束；3）Y轴与角度（Theta）约束。此类数据以Matlab格式存储。如需进一步了解细节，请参阅论文《Learning Singularity Avoidance》，作者为Manavalan, J. 与 Howard, M. J. W.，发表于2019年IEEE/RSJ国际智能机器人与系统会议（IEEE/RSJ International Conference on Intelligent Robots and Systems）。