An extended model of the ISO-2631 standard to objectify the ride comfort in autonomous driving.

Abstract

With the development of autonomous driving, the occupants' comfort perception and their activities during the drive are becoming increasingly the focus of research. Especially in one of the first applications, a drive on a motorway, vertical dynamics play a major role. To be able to robustly objectify ride comfort, better models need to be developed. Initial studies have shown, that the current ISO-2631 standard creates good results in the objectification and can be regarded as benchmark. To increase the accuracy in objectification, an extended model with the occupants' head as additional measuring point is introduced. Instead of the known frequency filters, weighting (k-factors) is used to differentiate possible excitations. For comparing the model with the ISO-2631, a simulator study with 5 excitations and 50 inattentive subjects is carried out. Evaluating the study with the ISO-2631, 3 out of 5 excitations indicate a significant difference between the occupant's impression and the calculated comfort value. In comparison the extended model has no significant difference. The results further show, that inattentive occupants move their heads significantly more. By measuring accelerations of the head, the extended model creates equivalent or more accurate comfort values than the ISO-2631.