Doctoral Research · Space Robotics Inspection with a Free-Flying Space Manipulator
A Doctoral Research Journal Aerospace Engineering

CHAIN_13 D3 — schedule-κ oracle: calibration, then certification

Pre-registered in tasks/chain13_criteria.md (D3); oracle laws + recorded idealizations in validation/schedule_kappa_oracle.py’s docstring. Stage (a) asks: rolled out at the 8 flown configurations, does the oracle put each run’s top-5 logged κ-peak phases inside (±0.02 phase) a predicted σ₆ < 0.05 interval?

run peak hits gate predicted spans pred frac<0.025 logged frac<0.025 track err med/max [m] per-peak phase dist
r0_6dof_path 5/5 PASS 2 0.96 0.76 0.0832 / 0.611 [0.0, 0.0, 0.0, 0.0, 0.0]
r0_6dof_minus2pct 5/5 PASS 2 0.96 0.76 0.0830 / 0.611 [0.0, np.float64(0.0), np.float64(0.0), 0.0, 0.0]
r0_6dof_plus2pct 5/5 PASS 2 0.96 0.76 0.0829 / 0.611 [0.0, 0.0, 0.0, np.float64(0.0), np.float64(0.0)]
r1_7dof_path 0/5 fail 0 0.00 0.27 0.0000 / 0.566 [inf, inf, inf, inf, inf]
r1_7dof_minus2pct 0/5 fail 0 0.00 0.13 0.0000 / 0.566 [inf, inf, inf, inf, inf]
r1_7dof_plus2pct 0/5 fail 0 0.00 0.01 0.0000 / 0.566 [inf, inf, inf, inf, inf]
r2_7dof_reactive 0/5 fail 0 0.00 0.38 0.0000 / 0.566 [inf, inf, inf, inf, inf]
r3_7dof_anchor 0/5 fail 0 0.00 0.69 0.0000 / 0.566 [inf, inf, inf, inf, inf]

Stage (a) gate: 3/8 runs pass (bar ≥ 7/8) → FAIL — NO stage-(b) prediction is citable (pre-registered stop)

regen: /Users/antoniahoffman/miniforge3/envs/new-pin-env/bin/python validation/schedule_kappa_oracle.py