roboticslab-uc3m / tools

Additional software tools.
https://robots.uc3m.es/tools/
0 stars 0 forks source link

RealToSimControlboard config files for robot end-effectors #16

Open jgvictores opened 4 years ago

jgvictores commented 4 years ago

RealToSimControlboard config files for robot end-effectors, as a follow-up of https://github.com/roboticslab-uc3m/tools/issues/15

NOTE: Before closing, edit https://github.com/roboticslab-uc3m/teo-configuration-files/blob/develop/share/applications/teoSimBase.xml (https://github.com/roboticslab-uc3m/teo-configuration-files/blob/5e0773dba56cd61ba98088a11965a44cde8a03f9/share/applications/yarpmanager/teoSimBase.xml) to invoke corresponding files (may split into none vs each end-effector).

jgvictores commented 4 years ago

Note port names (/teo/rightHand for either lacquey or dextra):

jgvictores commented 4 years ago

Additionally pending some ":smile:2:smile:" on interfaces (currently exposes IPositionControl) and maybe even some port naming.

jgvictores commented 4 years ago
  • Dextra: pending

From Drive: sign-language > iROS 2019 > csvs > backups:

  1. Using files from dextra_joint_record_20190211.zip:
    dextra_joint_record_step01_pinky_1_fwd.csv > dextra_joint_record_step01_pinky_fwd_all.csv
    dextra_joint_record_step01_pinky_2_fwd.csv >> dextra_joint_record_step01_pinky_fwd_all.csv
    dextra_joint_record_step01_pinky_3_fwd.csv >> dextra_joint_record_step01_pinky_fwd_all.csv
    ...

    Until 199 lines/file * 10 files = 1990 lines.

  2. Then, use calc to sort.
jgvictores commented 4 years ago

https://machinelearningmastery.com/moving-average-smoothing-for-time-series-forecasting-python/

from pandas import read_csv
from matplotlib import pyplot
series = read_csv('dextra_joint_record_step01_pinky_fwd_all_sorted.csv', header=0, index_col=0)
# Tail-rolling average transform
rolling = series.rolling(window=50)
rolling_mean = rolling.mean()
print(rolling_mean.head(10))
# plot original and transformed dataset
series.plot()
rolling_mean.plot(color='red')
pyplot.show()