Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Brandons baked beans adding notes #32

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions features/Alpha-Beta-Gamma-filter.md
Original file line number Diff line number Diff line change
Expand Up @@ -22,3 +22,8 @@ on Gyro:
* `gyro_abg_half_life` (OSD: GYRO ABG HL) Range 0-10000, default 50

From testing Alpha Beta Gamma filter should only be used in conjunction with a lowpass filter. Results of the lowpass Alpha Beta Gamma filter combo do show better results than just a lowpass filter on its own, and is thus still worth persuing.

ABG is more of a movement promotor than it is a filter. It can be used to track motion with near zero delay response, but doesn't require a simulated system model to do it. Blackbox logs show that the lower you set the Alpha value, the lower the noise floor but you get a stronger peak on movement band (as it promotes movement). It should be noted that any existing tuning oscillations typically are amplified under the alpha value.
The lower the Alpha value, the more the output is filtered. Disable by setting Alpha value to 0.
ABG boost will decrease ABG latency, but higher values may feel unnatural.
ABG has memory and it decays based on abg halflife, where the value is in ms.
4 changes: 3 additions & 1 deletion features/Emu-Boost.md
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,8 @@ EmuBoost is similar to FeedForward in that it changes your stick feel. One way t

0.3.0+: Feed Forward was removed completely.

1.0.0: Feed Forward returns. Boost Limit adjustment is built into the Emuboost value now.

### How does EmuBoost work?
EmuBoost is a boost to the error used in calculating your pterm and iterm. This means that your moves will be more aggressive when running a high EmuBoost. Small craft like whoops are quite happy running an EmuBoost of almost 1000!

Expand Down Expand Up @@ -63,4 +65,4 @@ set f_roll = 35
set f_yaw = 35
```

***@nerdCopter Tip (0.2.0+):*** As a less experienced freestyle pilot, I find high FF/Boost values a bit unbridled. Using an X-Lite controller, I prefer low RC-Rates in conjunction with Expo. Having higher FF/Boost values seems to unsettle this and produce too quick of a response when moving sticks off center. I prefer a smooth and subtle change. For me, higher values also make it easier to overshoot my end of roll/flip target. So for now I'm choosing Low FF/Boost values. Also notice that many popular presets have zero FF/Boosts.
***@nerdCopter Tip (0.2.0+):*** As a less experienced freestyle pilot, I find high FF/Boost values a bit unbridled. Using an X-Lite controller, I prefer low RC-Rates in conjunction with Expo. Having higher FF/Boost values seems to unsettle this and produce too quick of a response when moving sticks off center. I prefer a smooth and subtle change. For me, higher values also make it easier to overshoot my end of roll/flip target. So for now I'm choosing Low FF/Boost values. Also notice that many popular presets have zero FF/Boosts.
4 changes: 2 additions & 2 deletions features/Feathered-PIDs.md
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ parent: Features
nav_order: 5
---

# Featherd PIDs
# Feathered PIDs

* Feathered PIDs changes how dTerm (kD) is calculated.
* 0 is error-based dTerm and 100 is measurement based-dTerm.
Expand All @@ -14,4 +14,4 @@ nav_order: 5
* Measurement-based is smoother but less responsive than error-based.
* Any value in-between is a percentage of each.
* Some quads prefer error-based, while others prefer measurement-based.
* 50 is an exact balance between both.
* 50 is an exact balance between both.
4 changes: 4 additions & 0 deletions features/RC-Smoothing.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,6 +8,10 @@ nav_order: 10
# RC Smoothing

RC Smoothing is a personal preference and may be specific to components used. RC Smoothing is CPU intensive. Generally, BiQuad is smoother at cost of minor delay and PT1 faster at cost of minor less smoothness.
Starting in 1.0.0, interpolation is depreciated, as is biquad and has been replaced with pt2. RC Prediction has been added and works with smoothing to reduce input delay.

set rc_predictor = 100
other rc predictor parameters, discuss tuning

***@3dracingman Tip:*** Racers should leave RC Smoothing defaults. For freestylers, if not using BBL to analyze RC Command, then Interpolation is a good start. I have spent many hours of testing and concluded the following personal preference:
```
Expand Down