Study on the Evaluation Model of Vehicle Comfort Based on the Neural Network

Abstract

Due to the subjective perception of the driver, the comfort cannot be described by the objective indexes, meanwhile the normal subjective evaluation methods require plenty of human and material resources. Therefore, in this paper, in order to evaluate the comfort when the vehicle occurs low frequency longitudinal vibration during Tipin/out operations, a comfort evaluation method based on the neural network model is developed. First of all, through the mechanism analysis of the low frequency longitudinal vibration, three basic signals of the objective evaluation are determined. During different Tipin/out operations, the basic signals are collected by the experiment instruments and the subjective evaluation grades are determined by professionals. After that, the basic signals are weighted filtered and transformed into three objective evaluation indexes. Finally, the neural network model is trained by a large number of subjective evaluation grades and the objective evaluation indexes. The comfort performance of the vehicle during Tipin/out operations can be evaluated by the evaluation method.