In this repository are all the files used to generate and analyze the data presented in:
Barnett, N. V., and Lammert, A. C. (December 5, 2022). "Dynamic Stability of Passive Dynamic Walking Following Unexpected Perturbations." ASME. J Biomech Eng. April 2023; 145(4): 044501. https://doi.org/10.1115/1.4056166.
The main files in this folder are perturb_pdw.m
and collect_data.m,
which
encompass the data generating and synthesizing process. heatmap_loop.m
is also a useful data generating function.
collect_data.m
saves data into the
PDW-perturbations/Data directory, within a folder named, for example,
Data_n50000g0.014p0.5d03-Jun22
. Where the number after n
is the number
of fall or nonfall trials completed (not counting early falls), the number
after g
is the gamma value in radians, the number after p
is
the perturbation size (referred to as d
are the day number the data was saved followed by the month and year.
Details of the saved files are in the code documentation.
heatmap_loop.m
saves two data sets into a folder in PDW-perturbations/Data
titled, for example, HeatmapData_n20000g0.014_0.019p0.02_0.5d01-Jun22
.
The gamma and perturbation ranges are shown after g
and p
, respectively.
Details of the saved files are in the code documentation.
Data generation takes a LONG time. This most likely can be improved, but ultimately is a fundamental component of the model given the very low percent yields at high perturbation magnitudes. Before generating large amounts of data, experiment with the speed on your machine and familiarize yourself with the approximate time smaller trial numbers take.
The main files in this folder are pert_percent_histograms.m
,
fall_step_histogram.m
, linear_classify_processing.m
,
kinematics_compare.m
, and heatmap_grey.m
.
These were the files used to create the figures included in the manuscript.
The other files are either auxilliary functions for those previously mentioned
or other pieces of analysis. The data files loaded in these are shown as an example of
what data each script is able to process.
This folder is copied from the GitHub repository DrosteEffect/BrewerMap to assist in creating quality figures.
Since the purpose of this paper is to house the code described in the paper, no substantial modifications are needed. Performance and UI improvements are welcome to be submitted through a PR.
This work is licensed under the GPL General Public License -- a copyleft license.
The foundation of the simulation code is derived from the horchler/Passive-Dynamic-Walking repository, which carries its own copyright.
The following files are associated with COPYRIGHT_HORCHLER.txt
:
File | Modifications |
---|---|
perturb_pdw.m | Major modifications |
simpwm_noviz.m | Added outputs, y0 input, and suppressed visualization |
wmview.m | Unedited |