Identifikasi Parameter Koefisien Gesek Memakai Metode Jaringan Saraf Tiruan Untuk Kontrol Dinamika Kendaraan

Midriem Mirdanies, Estiko Rijanto

Abstract


Dynamical control of vehicle in order to stabilize its maneuver depends on the friction force between tire and road surface. In this paper it will be described a parameter identification method of a   non-linear model using backpropagation, which is one method in an artificial neural network (ANN). The non-linear model used in this paper is friction coefficient between tire and road surface,in which it has two parameters that should be defined. The ANN architecture used in this paper has one input, one hidden layer and one output. Computer simulation has been done using GNU Octave software for three road conditions: slippery road, wet road and dry road. From the simulation results the two parameter values have been obtained so that the equation of friction coefficient on the slip road is µ = 0.4 sin(1,5761.tan-1(10,901.s)), on the wet road is µ = 0.6 sin(1,6443.tan-1(9,5941.s)) and on the dry road is µ = 0.9 sin(-1,6354.tan-1(-10,279.s)).

Full Text:

PDF


DOI: http://dx.doi.org/10.14203/j.inkom.91

Refbacks

  • There are currently no refbacks.


INKOM - Jurnal Informatika, Sistem Kendali dan Komputer