Authors: Robert Goldman, Moustafa El-Gindy, Bohdan Kulakowski
Addresses: Naval Warfare Center, Division Newport, Newport, RI 02840, USA. ' Applied Research Laboratory, The Pennsylvania State University, 201 Transportation Research Building, University Park, PA 16802, USA. ' The Pennsylvania Transportation Institute, 206 Transportation Research Building, The Pennsylvania State University, University Park, PA 16802, USA
Abstract: This paper documents the research and development of a software-based rollover-warning device (RWD) to be used for road vehicles, with plans for hardware implementation. Although the RWD development concept is fairly general, the design described in this paper was geared towards heavy vehicles with high center-of-gravity height to track-width ratios. The RWD uses artificial neural networks to learn the dynamic input/output response of a road vehicle and estimate the instantaneous roll stability using inputs that are relatively easy to measure. The state of roll stability is quantified using a convenient measure called the load-transfer ratio (LTR) and used in conjunction with the rate of change of LTR as inputs to the RWD based upon a fuzzy logic rule-base for determination of an output warning level. Although the current RWD is based purely on computer simulation, experimental validation was performed and will be published at a later date.
Keywords: rollover dynamics; rollover warning device; heavy vehicles; simulation; vehicle dynamics; artificial neural networks; road vehicles; roll stability; load-transfer ratio; fuzzy logic; rollover accidents; heavy trucks; vehicle safety.
International Journal of Heavy Vehicle Systems, 2005 Vol.12 No.4, pp.282 - 306
Published online: 30 Nov 2005 *Full-text access for editors Access for subscribers Purchase this article Comment on this article