search

thomas-haslwanter / scikit-kinematics

Python functions for working with 3D kinematics.
Other
128 stars 45 forks source link

Data import error and unit for gyroscope data #9

Closed jankodie closed 6 years ago

jankodie commented 7 years ago

Hi Thomas,

if I try to set sample rate, acceleration and gyroscope data by the in_data parameter of the IMU_Base constructor I get an error. The error is caused by the function call self._set_info() in line 199 of imus.py (last call in _set_data that is called with in_data in function __init__). But the function expects all parameters in its definition (def _set_info(self, rate, acc, omega, mag): line 295 of imus.py). I solved it for me in the rather crude way to call it with self._set_info(rate = self.rate, acc = self.acc, omega = self.omega, mag = self.mag) from _set_data (line 199).

Besides that I have a question regarding the unit of the gyroscope data. In the comment for the IMU_Base class (line 65) the unit is mentioned as [deg/s] but in nearly all other cases as [rad/s] - as far as I've seen from a first look. I also didn't noticed a conversion between rad/s and deg/s. That's why I suppose the gyroscope data should be provided in [rad/s]. Is this correct?

Best regards, dj

ivan866 commented 6 years ago

_set_info() in line 199 is called with no args, while it should have 4.

thomas-haslwanter commented 6 years ago

Thanks for pointing that problem out: it crept in when I made the code more modular (for simpler implementation of new sensor), and forgot to update the manual data input. Should be fixed in Ver 0.6.4.

ivan866 commented 6 years ago

I provide omega in deg/s and acc in m/s^2 to imus.in_data(), then call imus.calc_position() and quat.quat2deg(imus.quat), and I get very large values in imus.pos (17000 to 1250000). scipy.integrate.cumtrapz calculates cumulative antiderivative instead of a finite antidifference. How can I get discrete position? Is numpy.diff(cumtrapz(acc,axis=0),axis=0) sufficient to get velocity? Also, there is no way in 0.6.8 to provide CStype argument to imus.analytical(), it is always 'bf'.

thomas-haslwanter commented 6 years ago

1) In second order approximation, cumtrapz is ok. However, you have to eliminate gravity first. 2) In order to get velocity, you should use scipy.signal.savgol_filter 3) IMUs measure the inertial signal with respect to the sensor. So the analysis has to be "bf"