These scripts require the moteus python library, installed via pip3 install moteus. See the python reference and the full Moteus reference for details.
actuator.py Shared abstraction layer for the moteus motor controller using the USB to canFD interface. Used by all other test scripts
record_max_torque.py torque ramps with configurable duration and peak
record_speedramp.py Speed ramp with configurable duration and peak
record_torque_constant.py Torque ramp in combination with a mini40 loadcell (untested)
record_torqueramp.py Torque ramps for play and stiffness. multiple repetations at 2 different peaks. Used for play and stiffness estimation
record_trajectory.py For a trajectory of postions with configurable speed limit. Repeats trajectory at increasing accelerations.
record-run-in.py Run actuator in alternating directions indefinetly while saving CSV data.
These files are jupyter notebook files requiring the jupyter python library, or can be opened using vs-code. Furthermore, they require some libraries installed by pip install numpy matplotlib pandas
analyse_anti-backlash.ipynb jupyter notebook for the play, stiffness, and friction tests and all resulting plots. Saved plots as SVG are the ones used in the the paper.
analyse_performance.ipynb jupyter notebook for the maximum torque and velocity tests and plots, as well as the pendulum trajectory tracking tests and corresponding plots
./test_data Resulting test data, all files have timestamps and short descriptions in the name. All of the used data is aready added in the analysing notebook files
./figures Resulting plots as SVG files from the analysis notebooks