ActionMap (50 demos per task)· Iteration 1
EnvironmentDatum-corner alignment
- Episodes
- OpenVLA-OFT 30 ActionMap (50 demos per task) 30
- Total duration
- OpenVLA-OFT 10:38 ActionMap (50 demos per task) 10:49
- Mean duration
- OpenVLA-OFT 21.3 s ActionMap (50 demos per task) 21.6 s
- Success
- OpenVLA-OFT 20% ActionMap (50 demos per task) 27%
- Manipulation density
- OpenVLA-OFT Share of time manipulating, five definitionsJoints moving98.2%End-effector moving88.6%Gripper engaged0.0%Contact force > 5 N68.3%Gripper engaged or in contact68.3% ActionMap (50 demos per task) Share of time manipulating, five definitionsJoints moving98.3%End-effector moving90.6%Gripper engaged0.0%Contact force > 5 N74.9%Gripper engaged or in contact74.9%
- SPARC smoothness
- OpenVLA-OFT 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. ActionMap (50 demos per task) 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
0:000:23