docs: record 2026-04-18 session findings across all doc surfaces

- next_steps.md: chronology entry for PRs #4-6 — trace harness, VO-only
  diagnostic (ORB 100% on EuRoC), harness ORB fix (vo_success 0→99/100);
  decision note on Mock vs ORB backend; next-step: ESKF init with synthetic
  GPS origin
- README.md adapters table: update EuRoC status to reflect new vo_success
  baseline

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Yuzviak
2026-04-18 14:42:00 +03:00
committed by Maksym Yuzviak
parent 1ed7729fc2
commit 2ccd7be6fb
2 changed files with 7 additions and 1 deletions
+1 -1
View File
@@ -109,7 +109,7 @@ Markers (`e2e`, `e2e_slow`, `needs_dataset`) are registered in `pyproject.toml`.
| Adapter | Platform | Raw IMU | GT | Real-run status |
|---|---|---|---|---|
| `SyntheticAdapter` | — | yes (zero motion) | exact | smoke test only, always runs |
| `EuRoCAdapter` | indoor MAV | 200 Hz ADIS16448 | Vicon | ran on first 100 frames — ATE ~10.9 km, xfail documented; VO+ESKF active but no satellite anchoring for indoor scene |
| `EuRoCAdapter` | indoor MAV | 200 Hz ADIS16448 | Vicon | ran on first 100 frames — vo_success 99/100 (ORB), ESKF not yet init'd in harness (no start_gps), ATE xfail documented |
| `VPAIRAdapter` | fixed-wing light aircraft | no (pose-only) | GNSS/INS ~1 m | ran once — ATE ~1770 km, xfail documented; VO alone diverges without anchoring |
| `MARSLVIGAdapter` | rotary (DJI M300 RTK) | yes | RTK | pending (requires pre-extracted ROS bag) |