BeStable project

Benchmarking System for Assessment of Balance Performance (BeStaBle) is a funded project of the Eurobench cover project.

Copyright BeStable 2020

This is an example of Performance Indicator implemented in Octave. It is prepared to be used within the Eurobench Benchmarking Software.

Purposes

Characterize the gait performances (performance indicators - PI) of a walking human subject (step length, step width, step time and target error).

Testbed data collection

Before starting the measurement, the operator inputs subject information and selects perturbation parameters (protocol condition) in D-Flow testbed software (Runtime Console).

The software checks if perturbation parameters (stored in subject_X_condition_Y.yaml) are the same as in previous run (if previous run exists). If yes, the system increase run number (Z=Z+1), while the condition number (Y=Y) remains the same. If not, the system increase condition number (Y=Y+1) and starts with new run (Z=1). Note that only previous condition file is checked with the current protocol.

Here is flowchart diagram of the data collection and file naming.

The BeStable testbed outputs the following files:

1. subject_X_info.yaml
2. subject_X_condition_Y.yaml
3. subject_X_cond_Y_run_Z_platformData.csv
4. subject_X_cond_Y_run_Z_gaitEvents.csv (not used for PI computation)

Files 1-3 need to be uploaded for the performance indicators calculation process, where the following files are generated:

1. base_step_length_{left, right}.yaml
2. base_step_width_{left, right}.yaml
3. base_step_time_{left, right}.yaml
4. free_step_length_{left, right}.yaml
5. free_step_width_{left, right}.yaml
6. free_step_time_{left, right}.yaml
7. pert_{left, right}_{fw, fwiw, fwow, iw, ow}_step_length.yaml
8. pert_{left, right}_{fw, fwiw, fwow, iw, ow}_step_width.yaml
9. pert_{left, right}_{fw, fwiw, fwow, iw, ow}_step_time.yaml
10. pert_{left, right}_{fw, fwiw, fwow, iw, ow}_target_error.yaml
11. pert_{left, right}_{fw, fwiw, fwow, iw, ow}_success_rate.yaml
12. data (Octave file containing all data)
13. plot_results.pdf (plotted results with boxplots)

Performance indicators collected in files 1-3 relate to nonperturbed baseline walking (before enabling perturbations). Performance indicators collected in files 4-6 relate to nonperturbed walking inbetween perturbations (free walking). Performance indicators collected in files 7-11 relate to perturbed walking, 4 steps from perturbation onset. More details are provided in README

Installation

To run the application the following dependencies are needed:

sudo apt install fig2dev
pkg load statistics

Usage

from octave

Assuming we are in an Octave terminal, located at the root of this repository, the computation can be launched as follows:

# add source location to the octave path
addpath('src')
# we assume that a folder test_output has been previously created
computePI('test_data/input/subject_19_cond_2_run_1_platformData.csv', 'test_data/input/subject_19_condition_2.yaml', 'test_data/input/subject_19_info.yaml','test_output')

All Performance indicator files will be placed in the indicated folder test_output.

using the script

The script run_pi launches this PI from the shell of a machine with Octave installed. The permissions of this file must be changed in order to be executable:

chmod 755 run_pi

Assuming folder ./test_data/input/ contains the input data, and that ./test_output exists and will contain the resulting files, the shell command is:

./run_pi ./test_data/input/subject_19_cond_2_run_1_platformData.csv  ./test_data/input/subject_19_condition_2.yaml ./test_data/input/subject_19_info.yaml ./out_tests

Build docker image

(only tested under linux)

Run the following command in order to create the docker image for this PI:

docker build . -t pi_bestable

Launch the docker image

Assuming the test_data/input contains the input data, and that the directory out_tests/ is already created, and will contain the PI output:

docker run --rm -v $PWD/test_data/input:/in -v $PWD/out_tests:/out pi_bestable ./run_pi /in/subject_19_cond_2_run_1_platformData.csv /in/subject_19_condition_2.yaml /in/subject_19_info.yaml /out

Acknowledgements

Supported by Eurobench - the European robotic platform for bipedal locomotion benchmarking. More information: Eurobench website

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 779963.

The opinions and arguments expressed reflect only the author‘s view and reflect in no way the European Commission‘s opinions. The European Commission is not responsible for any use that may be made of the information it contains.