What do you log during OpenArm teleoperation so the run stays useful for replay, training, and later failure analysis?
This thread is about OpenArm teleop data logging for builders and integrators.\n\nWhat metadata, retries, and failure notes do you keep so the run is still valuable after the excitement of the session is gone?\n\nIf you reply, include one field that became useful only later.
Teams often save the trajectory and forget the session context. That makes later replay much weaker than it should be.
Retries, intervention points, and calibration state often become the most valuable signals when training starts to fail.
Share one logging field you did not think mattered at first but ended up using during evaluation or retraining.
Pick your closest symptom to follow the right troubleshooting path.
Reproduce "[OpenArm] Teleop data logging for builders integrators (intermediate)" in a controlled loop, then compare baseline vs current measurements before applying partial fixes.
python tools/reproduce_issue.py --case current_thread
python tools/validate_fix.py --checklist standard_intermediateIf residuals or drift fail your acceptance limits after warmup, switch to full recalibration/recovery workflow.
Use them as a checklist template first. Confirm interface names, fixture IDs, and safety conditions in your own cell before execution.