Comparative multivariable statistical analysis of the dynamics of motor vehicles when they move in urban and extraurban environments

Abstract: Based on multivariable statistical analysis, the paper presents some functional particularities of the dynamics of vehicles when moving them in urban and extraurban areas. There are several aspects related to the study of vehicle dynamics by applying the methods of multivariable statistics. The study is based on experimental data obtained from tests of Ford Focus and Volkswagen Touareg II 3.0 V6 TDI cars equipped with Diesel engine, as well as tests of Logan Laureate car equipped with gasoline injection engine, all equipped with on-board computer, sensors and built-in actuators. The concepts and algorithms specific to multivariable statistics are presented, which are applied in the case of large data sets, as is the case of vehicles equipped with a computer on board. For this purpose, spatial correlation analysis, canonical correlation analysis, principal component analysis, factor analysis, group analysis, discriminatory analysis, data paternity classification and multidimensional scaling are used. These processes have ensured the detection of central trends in evolution, reduced data and better readability by reducing the space with more dimensions in the plan. Comparative conclusions are drawn on the dynamics of motor vehicles in urban and extraurban areas.

Keywords: vehicle dynamics, multivariable statistics, spatial correlation, canonical correlation analysis, principal component analysis.