READ ME File For: Dataset in support of the article 'Optimal Algorithms For Improving Pressure Sensitive Mat Centre of Pressure Measurements' Dataset DOI: 10.5258/SOTON/D3384 Date that the file was created: 20/02/2025 ------------------- GENERAL INFORMATION ------------------- ReadMe Author: Alexander Dawid Bincalar, University of Southampton [ORCID: 0009-0003-4038-0503] Date of data collection: 12/12/2024 Data Collection Location: Southampton, United Kingdom -------------------------- SHARING/ACCESS INFORMATION -------------------------- Licenses: CC-BY Cite: Bincalar, A.D., Freeman, C. & schraefel, m.c. Dataset in support of the article 'Optimal Algorithms For Improving Pressure Sensitive Mat Centre of Pressure Measurements'. University of Southampton 2025. https://doi.org/10.5258/SOTON/D3384 This dataset supports the publication: https://www.mdpi.com/1424-8220/25/5/1283 AUTHORS: Alexander Dawid Bincalar, Chris Freeman, m.c. schraefel TITLE: Optimal Algorithms for Improving Pressure-Sensitive Mat Centre of Pressure Measurements JOURNAL: MDPI Sensors DOI: 10.3390/s25051283 -------------------- DATA & FILE OVERVIEW -------------------- This dataset contains: "Dataset.xlsx" - An Excel document with spreadsheets containing the raw simulation data and the computations to produce the average errors reported in the linked paper. "Code.zip" - A compressed folder containing the code that demonstrates the work in the paper. It was also used to generate the raw data used for the paper. "Code.zip" contains: "GenerateAnOptimalGeometry.py" - this script demonstrates how optimal geometries can be computed using our algorithms. "OptimalGeometryComputations.py" - this script computes the base errors for a default geometry, as well as the errors for an optimal geometry used in our paper. This script then saves the data as csv files within the "OutputData" folder. "FootprintFittingComputations.py" - this script implements the footprint fitting algorithm, saving the data as csv files within the "OutputData" folder. "requirements.txt" - The Python modules used to run the program. "README.md" - A summary of how to run the Python scripts, a disclaimer and and a reference to the license. "LICENSE" - A text file containing the CC-BY license. "InputData" folder - a folder that holds "pressure_map.csv". "pressure_map.csv" - a file containing an artificial pressure map of a foot which is used in the scripts. "OutputData" folder - a folder where the scripts will save csv files containing raw data. "1.txt" - a dummy file so that "OutputData" could be put into a zip folder. -------------------------- METHODOLOGICAL INFORMATION -------------------------- The details on how the raw data was generated are explained in great detail within the paper this dataset was produced for. The mathematics within the paper were used to write the Python scripts that generated the dataset. The processing of the dataset is also described within the paper. The way X percentage error, Y percentage error, absolute percentage error, and average absolute percentage error are calculated can also be seen within the spreadsheets by looking at the formulas used within columns 5, 6, 7, and 8 (respectively). Microsoft 365 Excel was used to produce the dataset, however, other spreadsheet software was as Google Documents should suffice. Python 3.12 was used to run the Python scripts. The exact modules and their versions can be found in "requirements.txt". The Python scripts require the "InputData" folder and the "OutputData" folder to be within the same directory, with "InputData" containing "pressure_map.csv". The Python scripts will produce csv files into the "OutData". These csv files will match the raw data (columns 1 to 4) within the "Dataset.xlsx" document. -------------------------- Dataset.xlsx -------------------------- The "Dataset.xlsx" file is an Excel document that contains multiple spreadsheets. Inside this document, the spreadsheets "Uniform Side Weight Shift", "Uniform Front Weight Shift", "Uniform Foot Slides" are the base results without using optimisations, so the layout is uniform. "Non-Uniform Side Weight Shift", "Non-Uniform Front Weight Shift", "Non-Uniform Foot Slides", are the results from an optimised track geometry that has a non-uniform layout. "Fitting Side Weight Shift", "Fitting Front Weight Shift", "Fitting Foot Slides" are from a uniform layout, but using the known footprint optimisation. In the aforementioned spreadsheets, the first 4 columns ("Real X", "Real Y", "Estimated X", "Estimated Y") in each sheet contains raw data from the simulations. The other columns use the raw data to compute the average absolute percentage error, with columns 5 and 6 containing the percentage error of the X and Y CoP, respectively. Column 7 computes the absolute percentage error of the CoP. Column 8 then holds the average absolute percentage error across all time steps. There are 51 time steps in each spreadsheet, other than "Overall Averages". The spreadsheet titled "Overall Averages" contains a summary of the average absolute percentage errors for each case and each scenario.