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

Accelerometer aiding #86

Closed
nplayle opened this issue Oct 21, 2014 · 2 comments
Closed

Accelerometer aiding #86

nplayle opened this issue Oct 21, 2014 · 2 comments

Comments

@nplayle
Copy link

nplayle commented Oct 21, 2014

Hi,

I'm currently using MSF with ETHZ PTAM in a masters project and have a few questions.

Are the accelerometers used in the determination of pitch and roll?

  • Currently I cannot see anywhere this is the case. Is the angle estimate corrected towards the estimate coming from PTAM? I am not 100% sure that my current setup of PTAM is estimating attitude correctly, especially if the initial coordinate frame of PTAM is not gravity aligned. If not, based on the MSF framework it doesnt look like accel aiding would be too difficult to implement?
    I believe my inaccurate angle estimates from PTAM might be incorrectly estimating accelerometer and gyro biases causing me to read incorrect velocity estimates if I pitch / roll my setup.

What corrects the orientation estimate in the case of GNSS only?

  • if PTAM is not available then there are no angle measurements available to correct the gyro based estimates (based on my current understanding)

What units is the pressure input?

  • In pose_pressure_sensor there looks to be a pressure sensor used, but I can't find anywhere that suggests the scale. Is it already computed into metres?

Thanks!
Nick

@simonlynen
Copy link
Contributor

@nplayle ,

Orientation + Accel:
the accelerometers are used in the estimator to predict the position. (We do not use them in a separate orientation filter or in a complementary filter to estimate orientation.)

The only assumption we have in the framework is that the frame of reference of PTAM is roughly gravity aligned. (Usually this can be made sure by initializing ptam s.t. the control grid is horizontal). If the ptam frame of reference is not gravity aligned the estimator will estimate this rotation (world-to-vision transformation) and compensate for it. Of course this is only meaningful if the initial guess is not more than 10-20 degrees off.

The first thing I would verify is the imu to camera transformation. We expect you to set this in the parameter file. Here you have to provide the position and orientation of the camera frame of reference w.r.t. the imu frame of reference expressed in the imu frame of ref.

GNSS

The global yaw is only observable if you translate in a direction orthogonal to the direction of gravity. The roll and pitch are observable. We usually set the yaw to a reasonable guess at the beginning which we take from a compass.

pressure sensor

The units are expressed as metric height above a reference point which is stored at the filter initialization time.

Simon

@simonlynen
Copy link
Contributor

Assuming this is clarified.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants