INAV PID Tuner

Analyze INAV Blackbox logs and generate practical PID/filter tuning suggestions.

This is a small FastAPI web tool for FPV builders who tune INAV aircraft from real flight data. Upload a .bbl or .bfl Blackbox log, review the extracted metrics, then copy the suggested CLI changes into INAV Configurator.

Download the review snapshot and see PROJECT_STATUS.md for validation notes. Blackbox logs, rule suggestions, and tuning feedback are welcome through the current Blackbox-log request.

What It Does

• Accepts INAV Blackbox logs in .bbl and .bfl format.
• Decodes logs through blackbox_decode.
• Extracts PID error, D-term noise, and gyro-noise metrics.
• Produces rule-based tuning suggestions for P, I, D-term LPF, and gyro LPF.
• Generates CLI-style output that can be reviewed before applying changes.

Why It Exists

PID tuning is easier when the decision is tied to real flight data. This tool is meant to make the first tuning pass less manual by turning Blackbox metrics into clear recommendations.

It is not a magic auto-tuner. Treat the output as guidance, make small changes, and always test carefully.

Requirements

• Python 3.10+
• blackbox_decode installed and available in PATH
• INAV Blackbox logs from a real flight

Download blackbox_decode from:

https://github.com/iNavFlight/blackbox-tools/releases

Run

pip install -r requirements.txt
python main.py

Open:

http://localhost:8000

Usage

1. Fly your aircraft and save the Blackbox log.
2. Upload the .bbl or .bfl file.
3. Review the metrics and recommendations.
4. Copy the generated CLI lines into INAV Configurator only after checking them.
5. Fly again and compare the next Blackbox log.

Current Recommendation Rules

Signal	What the tool looks for	Suggested direction
P error RMS	High or unusually low axis error	Increase or decrease P by small steps
I error RMS	High accumulated axis error	Increase I by small steps
D noise RMS	High D-term noise	Lower D-term LPF cutoff
Gyro noise RMS	High gyro noise	Lower gyro LPF cutoff

API

The web app exposes one analysis endpoint:

POST /analyze

Upload a .bbl or .bfl file as multipart form data under the file field.

Project Structure

File	Purpose
main.py	FastAPI app and upload endpoint
decoder.py	Blackbox decoding wrapper
metrics.py	Metric extraction
rules.py	Rule-based tuning recommendations
frontend/	Browser UI
tests/	Test files

Related Projects

• INAV firmware experiments
• SkyPilot H743 flight controller
• INAV Blackbox tools

Feedback Wanted

Useful feedback includes:

• Example INAV Blackbox logs from different airframes.
• Better tuning rules for fixed wing, 5-inch freestyle, and long-range builds.
• UI improvements that make the recommendations easier to review safely.

Start here: INAV Blackbox log request or discussion.

Good first issues and review paths: https://19379353560.github.io/contribute.html.

See ROADMAP.md for the current validation, analysis, and product plan.

Related guide: INAV Blackbox log sharing guide.

Safety notes: SAFETY.md.