SCRO RESEARCH ANNUAL REPORT Vol. 3, pp. 1-8, 2015 ISSN 1494-7617 USING LQG/LTR OPTIMAL CONTROL METHOD FOR CAR SUSPENSION SYSTEM Navid Razmjooy 1 *, Mehdi Ramezani 2 , Esmaeil Nazari 3 1* Department of Electrical Engineering, University of Tafresh, Tafresh, Iran, Tel:00989144539067, navid.razmjooy@hotmail.com. 2 Department of Electrical Engineering, University of Tafresh, Tafresh, Iran, ramezani@aut.ac.ir. 3 Department of mathematics, University of Tafresh, Tafresh, Iran, nazari.esmaeil@gmail.com. ABSTRACT Three optimal controllers for a multivariable active suspension system are designed and compared. The main idea of these controllers is to achieve optimal stability margins and good performance in step response of the system. LQG/LTR method is an optimal design technique based on shaping and recovering open-loop singular values while LQR is a technique based on pole placement methods which characterize the optimal pole location. LQG/LTR controller has been applied to the suspension system and the results have been compared by pole placement and LQR methods. Results show that LQG/LTR method is prominent and has the priority of application rather than the other compared techniques. Keywords: Car suspension system; Multi input multi output Control, Optimal Control; LQG/LTR design; Kalman filter; LQR control. INTRODUCTION Suspension systems include all springs, dampers and other linkages that connect the wheels to the body of the vehicle. They develop the passenger ride convenience and handling quality, insure good road holding, decrease road damage, and afford stability (Pedro et al, 2012). The desirable features in a suspension system are:  Body movement adjustment. The suspension system should ideally isolate the body from road disturbances and inertial disturbances associated with cornering (body roll) and braking/acceleration (body pitch).  Suspension movement adjustment. Intensive vertical wheel travel will result in non-optimum attitude of the tire relative to the road. The result makes a poor handling and stick of vehicle and fatigue of driver.  Force distribution. To achieve good handling characteristics, the optimum tire/road contact must be preserved on all four wheels. The model for suspension design with parallel connections of hydraulic actuators, and passive springs and dampers is shown in Figure1. Suspension is located between the sprung mass (vehicle body, m2 in the figure) and the unsprung mass (tires, wheels, brakes, mu in the_gure). The tunable parameters of the suspension are the solidity of the spring k12 and the damping value b12 and additionally the unsprung mass has its own solidity value k10. Z0, Z1 and Z2 are the levels of the sprung mass, unsprung mass and the road, respectively. The sprung mass has a resonant frequency about 1-3 Hz which is identified as the rattle-space frequency (Fijalkowski, 2011). (A) (B) Figure 1. Quarter-car active automotive suspension. a) Standard implementation. b) Implementation in the paper. There are a lot of studies about solving the trade-off problem between the ride comfort and good road holding by using different active suspension based on different control approaches, like LQR control(Hrovat, 1997), (Desai and Rajendra, 2014) LQG controller (Debbarma et al, 2014). Among these methods, LQG/LTR method is a robust and optimal method which is utilized and noticed. In this paper, we will use the LQG/LTR method to optimal control of a considered suspension system.