Title :
A genetic algorithm predictor for vehicular rollover
Author :
Trent, Victor ; Greene, Michael
Author_Institution :
Archangel Syst. Inc., Auburn, AL, USA
Abstract :
A model-based genetic algorithm predictor is developed to estimate the potential for vehicular rollover. The model is based on a 1997 Jeep Cherokee Sports Utility Vehicle (SUV) and is discretized at 100 Hz. Preliminary simulation results indicate rollover prediction of 400 ms in advance of the actual event. Such advance prediction would then allow the use of active anti-roll algorithms and systems such as differential braking or active suspension control.
Keywords :
braking; genetic algorithms; mechanical variables control; road vehicles; transport control; 100 Hz; 400 ms; Jeep Cherokee Sports Utility Vehicle; active anti-roll algorithms; active suspension control; differential braking; genetic algorithm predictor; tire deflection estimate; vehicular rollover; Control systems; Current measurement; Discrete event simulation; Genetic algorithms; Prediction algorithms; Predictive models; Time measurement; Tires; Vehicle dynamics; Velocity measurement;
Conference_Titel :
IECON 02 [Industrial Electronics Society, IEEE 2002 28th Annual Conference of the]
Print_ISBN :
0-7803-7474-6
DOI :
10.1109/IECON.2002.1185235
