pegfail op• Data kinds: signal → table
• Call: import fullseye as fs; fs.ledger.insertion_stall_detect(depth, window: 'int' = 30, min_advance: 'float' = 5e-05, arm_advance: 'float' = 0.0005) -> 'dict' (to call the implementation directly, import pegfail; pegfail.insertion_stall_detect(depth, window: 'int' = 30, min_advance: 'float' = 5e-05, arm_advance: 'float' = 0.0005) -> 'dict'; from the registry, opsdrive.get("insertion_stall_detect"))
Detects stalls from a depth time series [m] (at each tick i, stalled if depth[i] − depth[i − window] < min_advance).
Before the descent starts (while holding the pose) the depth is constant and a naive detector fires —— it arms only after advancing arm_advance (onset is the first stalled after that). Returns: stalled (sequence of bool, same length), onset (index of the first stall, −1 if none), armed_at (index at which it armed; −1 = never armed = never advanced), n_stalled. It does not fire on a monotonic descent (speed ≥ min_advance/window) (gate). Raises ValueError: window < 1, min_advance ≤ 0, empty series.
• Sample-data catalog (download URLs / licences) — 2-D uses skimage.data (BSD/public domain) plus synthetic images; 3-D lists download URLs for real data sources (Stanford, PDS, …).
• Operator provenance and references — the sources of the research/methods this op family came from.
• The canonical algorithm (author, year) and its uses are named in the family usage guide above.
• poc_peg_failure_recovery — py -3.11 examples/poc_peg_failure_recovery.py
table as input)course_layout · course_occupancy · course_contains · world_build · world_camera · world_move · lidar_scan · rss_longitudinal_same
pegfail)insertion_failure_table · insertion_failure_validate · insertion_signature · insertion_failure_classify · insertion_recovery_primitive · jamming_parallelogram_planar · jamming_force_check · wedging_risk
*Provenance: pegfail.py — DRIVE operator registry. This per-op note is generated by tools/opdocs.py md (do not hand-edit).*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.