Patch Policy (CLS versus Dense)· Iteration 2
EnvironmentPlace spray bottles upright, nozzle forward (bimanual handover)
- Episodes
- Patch Policy (CLS versus Dense) 30 CLS token 30
- Total duration
- Patch Policy (CLS versus Dense) 33:31 CLS token 33:55
- Mean duration
- Patch Policy (CLS versus Dense) 67.0 s CLS token 67.8 s
- Success
- Patch Policy (CLS versus Dense) 27% CLS token 0%
- Manipulation density
- Patch Policy (CLS versus Dense) Share of time manipulating, five definitionsJoints moving100.0%End-effector moving99.9%Gripper engaged75.4%Contact force > 5 N63.8%Gripper engaged or in contact89.3% CLS token Share of time manipulating, five definitionsJoints moving100.0%End-effector moving99.0%Gripper engaged77.1%Contact force > 5 N63.7%Gripper engaged or in contact87.6%
- SPARC smoothness
- Patch Policy (CLS versus Dense) Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015. CLS token Spectral arc length of the motionHow directly the arm moved, read from the shape of its speed spectrum. Unitless measure ≤ -1; the closer to -1, the smoother.The measure ignores how large a movement is, so a large and a small one compare directly. Shown here as the mean over this environment's measured episodes.Cutoff10 HzAmplitude threshold5%Zero-padding16×Signalend-effector translationBalasubramanian et al., On the analysis of movement smoothness, J Neuroeng Rehabil 2015.
Episode
Held-out benchmark environment
0:001:13