Abstract
The tires contacting with the road are important parts determining the vehicle behavior. Recently, the studies which want to increase the vehicle safety by observing the tire strains containing additional information about vehicle are in progress actively. This system called the intelligent tire system. For developing the intelligent tire system, the strain sensor measuring the strains of tire must be developed. However, there is no reliable sensor by reasons of adhesion of sensor and lack of reliability. In this research, strains of tire were estimated using the tire Finite Elements Model. And from the analysis using the FE model, the shapes of tire strains according to the driving conditions could be estimated in each traveling events. Based on the results of analysis, we suggest the method to develop the simplified numerical tire model to estimate the tire strain in real time. The tire model consists of neural network model which estimates decisive parameters and piecewise polynomial curve model which constructs whole tire strains based on the decisive parameters. The developed tire model is made by combination of matrix, therefore the real-time computation could be achived. The simplified analytical tire strain model can be used in vehicle simulation replaced with real tire sensor system.
Keywords: Intelligent Tire System, Tire Model, Finite Element Analysis, Neural Network Model, Piecewise Polynomials curve fitting