ACT — bluev2, GelSight 500x375 (native)
One point of a GelSight-resolution sweep on the DG-5F + UR5e bluev2 task. Runs differ only in GelSight resolution.
- Dataset: Kaz55/dg5f_ur5e_bluev2 — 90 episodes / 101,406 frames
- GelSight: 500x375 (native)
- RealSense: 640x480 (identical across the sweep)
- Policy: ACT, chunk_size=60, n_action_steps=60
- Training: 100,000 steps (~7.9 epochs), batch 8, seed 1000
Inputs
observation.state (26) + RealSense x2 + GelSight x2
observation.velocity and observation.effort exist in the dataset but are
deliberately excluded — feature auto-derivation would otherwise feed them to
the policy and add a second difference between runs.
Caveat
On the earlier combined, blue_180ep and newblue sweeps, training loss was
essentially unchanged at every GelSight resolution, including
no-GelSight-at-all. Treat these losses as a sanity check, not as evidence about
tactile resolution — that needs on-robot evaluation.
- Downloads last month
- 28