azaion/gps-denied-onboard
1
0
Fork 0
mirror of https://github.com/azaion/gps-denied-onboard.git synced 2026-06-22 07:21:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update autodev state documentation to reflect progress in the Decompose phase, changing the current step from 5 to 6. Revise sub-step details to indicate a shift to phase 2, focusing on module layout for the Satellite Service and Tile Manager, and awaiting confirmation before product task decomposition. Additionally, enhance problem documentation to clarify the original still-image sample limitations and introduce the Derkachi representative fixture for improved data validation. Update references to the Tile Manager and Satellite Service throughout the documentation for consistency.

Browse Source
This commit is contained in:
Oleksandr Bezdieniezhnykh
2026-05-03 10:27:44 +03:00
parent 35547e9b65
commit 5bf2dbd85f
39 changed files with 5543 additions and 181 deletions
Download Patch File Download Diff File Expand all files Collapse all files
_docs/02_document/tests/performance-tests.md
+9 -7
View File
@@ -25,23 +25,25 @@
### NFT-PERF-02: BASALT + Wrapper Replay Latency
**Summary**: Validate relative VIO hot-path latency using synchronized public or representative camera/IMU data.
**Summary**: Validate relative VIO hot-path latency using synchronized Derkachi video/telemetry and public or representative camera/IMU data.
**Traces to**: AC-2.1a, AC-4.1, AC-4.2
**Metric**: Per-frame VIO latency, completion rate, and memory usage.
**Preconditions**:
- MUN-FRL/ALTO/EPFL/Kagaru or representative synchronized dataset slice is pinned.
- OpenVINS reference replay is available for comparison.
- Derkachi `flight_derkachi.mp4` and `data_imu.csv` are mounted and pass fixture validation.
- MUN-FRL/ALTO/EPFL/Kagaru or another representative synchronized dataset slice is pinned for calibrated final comparison.
- OpenVINS reference replay is available for comparison when the dataset supports it.
| Step | Consumer Action | Measurement |
|------|-----------------|-------------|
| 1 | Replay synchronized camera/IMU stream through BASALT + wrapper | Measure per-frame processing time |
| 2 | Compare emitted trajectory against ground truth | Measure completion rate and error distribution |
| 3 | Monitor memory | Track CPU/GPU shared memory peak |
| 1 | Replay Derkachi video at target 3 fps and stress rates from the 30 fps source | Measure per-frame processing time, dropped frames, and telemetry alignment |
| 2 | Replay synchronized camera/IMU stream through BASALT + wrapper | Measure VIO processing time and completion rate |
| 3 | Compare emitted trajectory against Derkachi `GLOBAL_POSITION_INT` and calibrated dataset ground truth where available | Measure completion rate and error distribution |
| 4 | Monitor memory | Track CPU/GPU shared memory peak |
**Pass criteria**: Normal-frame VO registration >95%; p95 processing latency <400 ms for the hot path; memory <8 GB shared.
**Pass criteria**: Normal-frame VO registration >95% on calibration-supported segments; p95 processing latency <400 ms for the hot path; memory <8 GB shared; Derkachi replay maintains stable 3-video-frames-per-telemetry-row alignment with <=10% dropped frames under sustained target-rate replay.
**Duration**: Dataset-dependent; at least one normal segment and one challenging segment.