Longitudinal and rotational accelerations
Tommi
It seems that the Galaxy S II that we use with Talos is not capable to recognize a rotational acceleration or a linear acceleration along the same axis... we are trying to find out why this happens. A boat is constantly accelerating and rotating along its longitudinal axis (specially 1x) and if the device used exchanges rotational acc for longitudinal ones recorded data are affected by a considerable error. The GSII is equipped by a K3DH Acceleration sensor made by STMicroelectronics, it seems that it was designed mainly to record rotational accelerations. Have you some information about?
Thanks!
Michael Hosemann
This list claims that the K3DH of which there is no mention in STM's webpage is actually an LIS3DH.
http://www.steim.org/research/wp-content/uploads/2012/03/android_acceleros.pdf
The datasheet can be found here:
http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00274221.pdf
This accelerometer detects longitudinal acceleration along three axes. Depending on how the phone is mounted in the boat, the rotating accelerations of the hull will be translated into longitudinal accelerations of the boat.
For sensing rotational acceleration a gyroscope would be a better choice. Phones are nowadays equipped with these, too. This presentation describes some ideas on how to fuse the data from accelerometers and gyroscopes.
www.icg.tugraz.at/Members/gerhard/mvc/MVC_04_Sensors.pdf
It looks very applicable to our problems.
Kind regards,
Michael
It seems that the Galaxy S II that we use with Talos is not capable to recognize a rotational acceleration or a linear acceleration along the same axis... we are trying to find out why this happens. A boat is constantly accelerating and rotating along its longitudinal axis (specially 1x) and if the device used exchanges rotational acc for longitudinal ones recorded data are affected by a considerable error. The GSII is equipped by a K3DH Acceleration sensor made by STMicroelectronics, it seems that it was designed mainly to record rotational accelerations. Have you some information about?
Thanks!
--
You received this message because you are subscribed to the Google Groups "talos-rowing" group.
To view this discussion on the web visit https://groups.google.com/d/msg/talos-rowing/-/4b14ILSz2XgJ.
To post to this group, send email to talos-...@googlegroups.com.
To unsubscribe from this group, send email to talos-rowing...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/talos-rowing?hl=en.
Tommi
I think that there aren't problems in device
orientation or displacement: the phone was well fixed to boat's hull, we tried
to put it as near as possible to the supposed center of mass of the rower-boat
system to minimize the effect of the pitch changes. The y axis was aligned to boat's longitudinal axis,
the recorded value at rest was 0 m/s^2 (no angle between the two axis, of
course we disabled gravity filter). We used always a sensor log app as well as
Talos to avoid problems created by negative horizontal angles of the
phone/hull. The problem still affects measurements, open Talos and try to
move you're device back and forth (the app will record a longitudinal acceleration) then
try to rotate it without moving it back and forth: the recorded data are quite
similar... I hope I was clear, I'm doing any mistake?
Rotational accelerations recorded "as" longitudinal ones worsen the quality of
the signal in a not negligible
quantity. We are trying to use the data from the gyroscope to solve the problem
(in the way that I explained in the topic about filters).
Thank you!