2
$\begingroup$

Magnetometer is used to measure yaw angle which will be used on AHRS system.When I simulate the magnetometer in matlab, I feel confused about what unit should be used for measurements magnetometer?Should I use μT(microtesla),rad or degree?

I know in real world, the magnetometer is to measure earth magnetic field, so it should be μT. the reason that I feel confused is that:In some papers, their simulation says $B_{i}$(earth's magnetic field) is given by $[\frac{1}{\sqrt{2}},0,\frac{1}{\sqrt{2}}]$ or $[1;0;1]$,symmetry-preserving observer.page11 and they didn't give the unit of above vector. I also find a book, called Small Unmanned Aircraft: Theory and Practice, it introduced how to simulate IMU sensors and says the standard deviation and bias error of Magnetometer could be 0.3 degrees and 1 degree.

So I am totally confused about the unit and value of the earth's magnetic field or Magnetometer measurements.

$\endgroup$
2
  • 1
    $\begingroup$ It is impossible to answer this question without knowing what you want to achieve with your simulation. Are you analysing the electronics, are you developing navigation software, are you investigating the effect of magnetic anomalies on navigation performance, are you modeling the dynamic response of a magneto meter to a changing magnetic field, something else? $\endgroup$
    – DeltaLima
    Feb 5 at 9:09
  • 1
    $\begingroup$ I am designing a estimator for the navigation system. so the measurements to my estimator are positions(GPS),acceleration(accelerometer),magnetic field(magnetometer).Right now I want to add some noise and bias to the measurements and see the estimation performance of the estimator. so in my case, I want to figure out the unit and value range of the magnetometer so that I can add the right noises. $\endgroup$
    – DarkKnight
    Feb 5 at 15:51
0
$\begingroup$

The unit does not really matter if you measure the right quantity. But obviously angle magnetic induction and angle are different quantities. Typically—but it might depends on the type of magnetometer used and is not true e.g. of mechanical compass—there will be two or three magnetometers detecting components of the magnetic induction vector, and then you calculate the bearing to magnetic north, and if you have three components the inclination from those.

The raw output of an electric vector magnetometer, measuring one axis of the magnetic induction vector, will be a voltage. You can convert it to the inductance in μT or whatever unit, but for navigation you only care about the direction and the units cancel out when you calculate that so you don't have to.

For magnetometers intended for nagivation, the required accuracy is only specified for the angle, in degrees, because the magnitude is not used.

$\endgroup$
4
  • $\begingroup$ For my case, the magnetometer would be a MEMS electric magnetometer. If a paper says $B_{i}$(earth's magnetic field) is given by $[\frac{1}{\sqrt{2}},0,\frac{1}{\sqrt{2}}]$ or $[1;0;1]$. Since they didn't give the unit, do you know the unit according to the values they give? is it μT, rad or degree? $\endgroup$
    – DarkKnight
    Feb 5 at 15:46
  • $\begingroup$ @DarkKnight no, that would require more context. $\endgroup$
    – Jan Hudec
    Feb 5 at 20:51
  • $\begingroup$ hello, I have added more context,could you help me take a look? $\endgroup$
    – DarkKnight
    Feb 7 at 16:34
  • $\begingroup$ I think I have found the answer why some papers didn't give the magnetic field vector. They have done some normalization. $\endgroup$
    – DarkKnight
    Feb 8 at 2:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.