Using fuzzy logic controller and evolutionary genetic algorithm for automotive active suspension system

This study designs a fuzzy logic controller (FLC) for an active automobile suspension system in which the membership functions and control rules are optimized using a genetic algorithm (GA). The objective of the FLC is to strike an optimal balance between the ride comfort and the vehicle stability. The values of the crossover and mutation parameters in the GA are adapted dynamically during the convergence procedure using a fuzzy control scheme. The convergence state of the GA is determined by using a support vector machine (SVM) method to identify the variation in each of the genes of the best-fit GA chromosome following each iteration loop. The feasibility of the proposed GA-assisted FLC scheme is verified by performing a series of numerical simulations in which the characteristics of the controlled plant are compared with those observed in a passive suspension system and obtained under an optimal linear feedback controller. The results demonstrate that the GA-assisted FLC results in a lower suspension deflection, a reduced sprung mass acceleration and a lower bouncing distance between the tire and the ground.