6533b7dbfe1ef96bd127159e

RESEARCH PRODUCT

Mathematical modeling and parameters estimation of a car crash using data-based regressive model approach

Hamid Reza KarimiKjell G. RobbersmyrWitold Pawlus

subject

Estimationregressive models parameters estimation viscoelastic modeling virtual experimentComputer sciencebusiness.industrySpeech recognitionApplied MathematicsVDP::Technology: 500::Mechanical engineering: 570CrashMachine learningcomputer.software_genreVDP::Mathematics and natural science: 400::Mathematics: 410Modeling and SimulationModelling and SimulationVirtual experimentArtificial intelligencebusinesscomputer

description

Author's version of an article in the journal: Applied Mathematical Modelling. Also available from the publisher at: http://dx.doi.org/10.1016/j.apm.2011.04.024 n this paper we present the application of regressive models to simulation of car-to-pole impacts. Three models were investigated: RARMAX, ARMAX and AR. Their suitability to estimate physical system parameters as well as to reproduce car kinematics was examined. It was found out that they not only estimate the one quantity which was used for their creation (car acceleration) but also describe the car's acceleration, velocity and crush. A virtual experiment was performed to obtain another set of data for use in further research. An AR model to predict the behavior of a low-speed car impacting a rigid barrier was created and verified.

yearjournalcountryeditionlanguage
2011-10-01Applied Mathematical Modelling
10.1016/j.apm.2011.04.024http://dx.doi.org/10.1016/j.apm.2011.04.024