FUSEGPS meaning of parameter Rpos

2 views (last 30 days)
Alberto CIPPELLETTI
Alberto CIPPELLETTI on 7 Feb 2024
Commented: Aishwarya on 10 Feb 2024
hello everyone!
i'm working on matlab to reconstruct a route with data from the IMU and GPS (sensors connected to an arduino board).
I have a doubt about the meaning of the Rpos parameter for the fusegps command: what does it mean on a 'practical' level?
thank you very much,
Alberto.

Answers (1)

Aishwarya
Aishwarya on 9 Feb 2024
Hi Alberto,
The "Rpos" parameter in the "fusegps" function refers to the position measurement covariance matrix of the GPS sensor.
To further explain the Position Measurement Covariance Matrix:
  • This matrix represents the uncertainty associated with GPS position measurements.
  • It quantifies how much the observed GPS position (latitude, longitude and altitude) can vary due to measurement errors.
  • The matrix has dimensions 3x3 beacuse it corresponds to geodetic latitude, longitude and altitude components.
To answer your question regarding its practical significance:
  • Usually, this covariance matrix is considered a diagonal matrix where the diagonal elements correspond to the variance of latitude, longitude and altitude measurements.
  • Do take note that latitude and longitude are measured in degrees with north and east being positive, while altitude is measured in meters.
  • You can determine the appropriate values for the variance based on GPS sensor specifications or through experimental validation. You can also make educated guesses about the noise properties of the sensor.
Refer to the following documentation for more details about the "fusegps" function: https://www.mathworks.com/help/nav/ref/marggpsfuser.fusegps.html
I hope this clarifies your question!
  2 Comments
Alberto CIPPELLETTI
Alberto CIPPELLETTI on 9 Feb 2024
ok thank you.
so if i have a gps sensor and i have to reconstruct a route with the data i collect, in the Rpos parameter i put the error of the sensor measurement that i find on the datasheet?
Aishwarya
Aishwarya on 10 Feb 2024
Yes, you can refer the datasheet if the data about the measurement noise is available. Ensure that variance of the measurements need to be entered in the matrix. Usually an approximate value is used when the data is not available.
There are also different ways to estimate the position covariance matrix in literature.

Sign in to comment.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!