Advanced mechatronic design using a multi-objective genetic algorithm optimization of a motor-driven four-bar system Zouhaier Affi a , Badreddine EL-Kribi a , Lotfi Romdhane b, * a Laboratoire de Ge ´nie Me ´canique, Ecole Nationale d’Inge ´nieurs de Monastir, Monastir 5019, Tunisia b Laboratoire de Ge ´nie Me ´canique, Ecole Nationale d’Inge ´nieurs de Sousse, Sousse 4000, Tunisia Received 2 June 2006; accepted 5 June 2007 Abstract In this work we present a genetic algorithm based method to design a mechatronic system. First, we present the sequential approach where we optimize the geometry of the mechanism, for a given path, and then solve the dynamic problem where we take into account the characteristics of the motor along with the inertia of the different links of the mechanism. Several types of objective functions are tested. We show, however, that this sequential method does not yield acceptable results for the dynamic behavior due to the fact that the geom- etry is assumed fixed when optimizing the dynamics. This led us to formulate a global optimization problem where all the parameters of the mechanism are considered simultaneously. The problem is then presented as a multi-objective optimization one where the geometry and the dynamics are considered simultaneously. The obtained solutions form what is called a ‘‘Pareto front’’ and they are analyzed for several different design conditions. This paper also shows the advantages of a multi-objective optimization approach over the single- objective one. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Advanced mechatronic design; Genetic algorithms; Mechanism synthesis; Multi-objective optimization 1. Introduction The traditional approach, for robot design process, is usually made of two sequential phases: the mechanical design, during which the geometry of the mechanism is determined, followed by the control system design, where the problem of the dynamic behavior is considered. The mechanical structure of a mechanism is usually determined in advance without considering the corresponding control system. This approach yields an electromechanical system, with high mechanical performance, in which the dynamic behavior is not usually acceptable. Control action may be hindered by hardware limitations even if much effort has been made on the sophisticated advanced controller [3,9]. The research trend in modern mechanical systems devel- opment is the mechatronic approach. The key idea of the mechatronic methodology is to create an integrated design environment that enables strictly simultaneous designs for mechanical structures, the control strategy and even the interaction of the whole system with its environment. The goal of this philosophy is to achieve an optimal mecha- tronic system performance. Recently, the mechatronic system design attracted sev- eral researches [1,4,6]. An integrated design and PD control of high-speed four-bar mechanism is presented in [1]. In this work, the authors showed that the result of the con- troller can be significantly improved by using the Design For Control (DFC) strategy. The key idea of this approach is to find the best mass distribution in the mechanism to facilitate the control strategy. In order to facilitate a con- troller design and improve tracking performance for high-speed system, the mechatronic approach suggests a 0957-4158/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.mechatronics.2007.06.003 * Corresponding author. E-mail addresses: zouhaier.affi@enim.rnu.tn (Z. Affi), kribi_badreddine @yahoo.fr (B. EL-Kribi), lotfi.romdhane@enim.rnu.tn (L. Romdhane). Mechatronics 17 (2007) 489–500