Improving vehicle ride comfort using an active and semi-active controller in a half-car model

Abstract

In this paper, fuzzy logic control of a vehicle suspension with magneto rheological damper on a random road is considered. To simulate road roughness height a filtered white noise is employed and a four degree-of-freedom, half-car model is used in the analysis. A sensor mounted in the front of the vehicle is responsible for measuring the road irregularities at some distances in the front of the vehicle. Some other sensors are used to measure relative velocities of the vehicle body with respect to unsprung masses in wheel travel spaces. All measurements are assumed to be conducted in a noisy environment. The state variable of the simulated vehicle are calculated, using a method which is similar to the Kalman filter. To have a tailored estimation of magneto rheological damper performance a model which is known as Bouc-Wen is adopted. In addition, utilization of genetic algorithm (GA) in elicitation of fuzzy rules and tuning membership function is exhibited. This controller is a combination of fuzzy controller as a feed back and preview controller as a feed-forward controller, so as to anticipate irregularities of the road. Furthermore, GA is used to determine the optimized preview time. The typical parameters of fuzzy controller are encoded into a chromosome represented as an integer string. Then, by evaluating a multi objective function using a non dominated sorting GA II the superior chromosomes are elected and as a result, the optimized fuzzy logic controller can be employed. Also, by implementation of the combined controller which is designed by GA, it is possible to obtain better performances for the Fuzzy Interface System or, at the same performances, a less complex structure for the system. The effectiveness of the proposed algorithm has been verified through a series of simulations for a former designed controller.