Evolutionary algorithm-based PID controller tuning for nonlinear quarter-car electrohydraulic vehicle suspensions

The basic challenge associated with the design of vehicle suspension system is the attainment of an optimal trade-off between the various design objectives. This study presents the design of proportional-integral-derivative (PID) controller for a quarter-car active vehicle suspension system (AVSS) using evolutionary algorithms (EA) such as the particle swarm optimization (PSO), genetic algorithm (GA) and differential evolution (DE). Each of the EA-based PID controllers showed overall improvement in suspension travel, ride comfort, settling time and road holding from the manually tuned controller and the passive vehicle suspension system. These improvements were, however, achieved at the cost of increased actuator force, power consumption and spool-valve displacement. DE-optimized PID control resulted in the best minimized suspension performance, followed by the GA and PSO, respectively. Frequency-domain analysis showed that all the signals were attenuated within the whole body vibration frequency range and the EA-optimized controllers had RMS frequency weighted body acceleration of the vehicle within allowable limits for vibration exposure. Robustness analysis of the DE-optimized PID-controlled AVSS to model uncertainties is carried out in the form of variation in vehicle sprung mass loading, tyre stiffness and speed.