CRS 评测方法：用关节轴分解机械臂操作策略的动作质量
Kinematic Constraint-Referenced Evaluation for Articulated Object Manipulation
机械臂操作评测只看成功率太粗了，这篇 arXiv 论文提出把动作投影到关节轴上，能看出策略到底是干活还是在瞎使劲。
论文提出 Constraint-Referenced Strategy（CRS）评测方法，将机械臂末端执行器的指令速度投影到物体的关节轴上，把沿关节的有效推进与偏离约束的浪费动作分开。在 74 个仿真铰接物体上，传统二值成功率能发现迁移失败但无法归因，而 CRS 的两个通道能为每种失败给出独特特征。实验显示，一个只用门训练的行为克隆策略在 110 次开抽屉任务中成功 59 次，其成功回合的离轴旋转速度约 0.5 rad/s 量级对比抽屉训练策略的 0.032 rad/s，差距约 40 倍。CRS 的判定在关节轴误差不超过 30° 时保持稳定，绝对阈值可容忍约 8° 的误差。
Kinematic Constraint-Referenced Evaluation for Articulated Object Manipulation
Manipulation policies are evaluated predominantly by binary task success, which cannot tell whether the commanded gesture is the one the object's kinematics admit. We introduce Constraint-Referenced Strategy (CRS), which projects the commanded end-effector twist onto the object's joint axis, separating progress along the joint from off-constraint waste. CRS requires a joint model and a joint-following scripted controller derived from it, but no recorded demonstration or learned judge. On 74 simulated articulated objects, success registers three transfer failures as lower rates but attributes none, while the two CRS channels give each a distinct signature. A door-trained behavior-cloning policy opens drawers in 59 of 110 episodes, and in these passes it commands about $40\times$ the off-axis rotation of a drawer-trained policy's passes ($1.27$ vs.\ $0.032$~rad/s). Relative to a scripted controller scored with the same axis, CRS verdicts are unchanged under joint-axis errors up to $30^\circ$; absolute thresholds tolerate about $8^\circ$.
