Hacktoberfest 2026: le issue che i maintainer hanno segnato per ottobre, aperte e adatte ai principianti. Sfoglia le issue Hacktoberfest

EPIC: Time-varying sensor fusion

Aperta
  • Qualcosa non va?
#346 1 commento 0 reazioni 0 assegnatari Vedi su GitHub

I maintainer di solito rispondono entro 2 giorni

Nessuno ha ancora preso questa issue.

Valutazione

Difficoltà
5/5
Tempo stimato
Più di una settimana
Idoneità per principianti
20/100
Tipo di issue
Funzionalità
Chiarezza
Da chiarire
Stato di attività
Attiva
Stack tecnologico
c, matlab
Ambito
embedded-iot

Direzione di ricerca

Start by reading the existing sensor pipeline and the sensor interrupt handlers; the issue does not name specific files or tests. Before implementation, agree on a smaller task and define the filter state, data-timing behavior, and acceptance criteria. Done should mean the chosen task has clear criteria and can be verified without relying on stale sensor data.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Descrizione

enhancement

Version: 2.8.0 maybe

Describe the feature you'd like
We can write an Extended Kalman Filter to process our sensor data in a way that doesn't reuse stale data. Our current sensor pipeline processes and logs data mostly all at once, which is fine for just logging, but we get different data rates from our sensors:

IMU: super fast, like 1000 Hz
GPS: around 10 Hz, but we get different types of messages with different data
Mag: up to 30 Hz
Baro: about 200 Hz

If we try to combine all the data during each sensor dump like we do now, we use old data that doesn't necessarily reflect the current state of the FC, especially GPS. We could set up our interrupt handlers for the sensors to set a flag that the data is ready along with a timestamp, then only process data that's ready when we do a sensor dump.

There are a lot of options on what the state vector of the filter will actually contain, but at a bare minimum, for example, something for vertical translation should output position, velocity, and acceleration. Also consider using an indirect Kalman filter, which estimates error and applies a feedback correction.

It might be desirable to write a complicated filter like this in MATLAB because otherwise we will need some solution to how to do linear algebra in our code. Other things, like interrupt service routine changes can be done in smaller tasks.

Describe alternatives you've considered
Leave most things the way they are and check if previous data is duplicated
CMSIS has linear algebra support, but we don't have SIMD so it wouldn't be super beneficial.

Additional context
I might update this with resources, but I think Phil's Lab has some stuff of this type of filtering.

Indirect Kalman filter (basically exactly what we're looking for): https://mars.cs.umn.edu/tr/reports/Trawny05b.pdf - this uses JPL quaternions but we use Hamilton quaternions

Points of Contact
Firmware Lead/Sensor SME: @NArmistead
Avionics Lead/ISR SME: @ETSells

Lingua principale
C
Stelle
8
Fork
6
Merge medio
2g 17h
PR unite (30g)
8

Preparare l'ambiente

  • Nessun Dockerfile né file Docker Compose
  • Ha un modello di pull request
  • Nessuna guida per i contributori

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di SunDevilRocketry/Flight-Computer-Firmware

Tutte le issue di SunDevilRocketry/Flight-Computer-Firmware

Issue simili

Altre issue su C

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.