Multi-objective optimization by genetic algorithms in H∞/LPV control of semi-active suspension

Anh Lam Do (1) , Olivier Sename (1) , Luc Dugard (1) , Soualmi Boussaad (2)

1 GIPSA-SLR - GIPSA - Systèmes linéaires et robustesse
2 LAMIH - Laboratoire d'Automatique, de Mécanique et d'Informatique industrielles et Humaines - UMR 8201

In semi-active suspension control, comfort and road holding are two essential but conflicting performance objectives. In a previous work, the authors proposed an LPV formulation for semi-active suspension control of a realistic nonlinear suspension model where the nonlinearities (i.e the bi-viscous and the hysteresis) were taken into account; an H1/LPV controller to handle the comfort and road holding was also designed. The present paper aims at improving the method of Do et al. (2010) by using Genetic Algorithms (GAs) to select the optimal weighting functions for the H1/LPV synthesis. First, a general procedure for the optimization of the weighting functions for the H1/LPV synthesis is proposed and then applied to the semi-active suspension control. Thanks to GAs, the comfort and road holding are handled using a single high level parameter and illustrated via the Pareto optimality. The simulation results performed on a nonlinear vehicle model emphasize the efficiency of the method.

Domaines: Automatique

Anh Lam Do, Olivier Sename, Luc Dugard, Soualmi Boussaad. Multi-objective optimization by genetic algorithms in H∞/LPV control of semi-active suspension. IFAC WC 2011 - 18th IFAC World Congress, Aug 2011, Milan, Italy. pp.n.c. ⟨hal-00606438⟩