Shop OBEX P1 Docs P2 Docs Learn Events
Math, Gyro & Accelerometers...? — Parallax Forums

Math, Gyro & Accelerometers...?

JavalinJavalin Posts: 892
edited 2009-06-09 07:47 in General Discussion
Hello all,

I've got a pcb for data logging gps, gyro, accelerometer data.· Thats all well and good, but·I want to integrate the three data sources to give me a better precision and higher update rate(s).

Sadly my maths is the let-down here.· Can anybody point me to a simple source of information on how I might do this?· (i.e. not kalman filters, etc)

(updated) I guess what I am trying to solve is:
1) Gyro Drift
2)·Integration of accelerometers (etc) taking account of gyro input (orientation) etc.

Fusion with the GPS seems easy enough - im just adding extra points between GPS readings.

Cheers,

James

Post Edited (Javalin) : 6/8/2009 8:12:25 PM GMT

Comments

  • LeonLeon Posts: 7,620
    edited 2009-06-08 16:06
    A Kalman filter is generally used for sensor fusion, but artificial neural networks can also be used, as well as other methods. I don't think you can avoid a lot of maths, whatever technique you use.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • JavalinJavalin Posts: 892
    edited 2009-06-08 20:42
    ah.· Found this http://obex.parallax.com/objects/319/ - its good but far too heavy for what I am after...

    James
  • LeonLeon Posts: 7,620
    edited 2009-06-08 21:30
    That is a Kalman filter!

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • JavalinJavalin Posts: 892
    edited 2009-06-09 07:47
    I know - its too heavy!

    James
Sign In or Register to comment.