Using Multiple Correspondence Analysis for large driving signals database exploration. Example with lane narrowing and curves

Abstract

In most driving studies, several factors (at least two, i.e. individual and time) and many variables are collected via multidimensional signals (MS). This article suggests starting the analysis while keeping the three main aspects of time, i.e. simultaneity, chronology and duration. To achieve this aim, with the possibility to show nonlinear relationships, a MS set exploratory investigation is performed using the pair space-time windowing/Multiple Correspondence Analysis. This article shows how intra and inter-individual differences can be underscored.