In this paper, the output-feedback control problem of a vehicle active seat-suspension system is investigated. A novel optimal design approach for an output-feedback H ∞ controller is proposed. The main objective of the controller is to minimize the seat vertical acceleration to improve vehicle ride comfort. First, the human body and the seat are considered in the modeling of a vehicle active suspension system, which makes the model more precise. Other constraints, such as tire deflection, suspension deflection and actuator saturation, are also considered. Then the output-feedback control strategy is adopted since some state variables, such as body acceleration and body deflection, are unavailable. A concise and effective approach for an output-feedback H ∞ optimal control is presented. The desired controller is obtained by solving the corresponding linear matrix inequalities (LMIs) and by the calculation of equations proposed in this paper. Finally, a numerical example is presented to show the effectiveness and advantages of the proposed controller design approach.

In this paper, the static output-feedback control problem of 5 degrees of freedom (DOF) vehicle active suspension systems is investigated. A novel H ! optimal controller is designed for this system to improve simultaneously ride comfort and handling ability. The actuator saturation, suspension deflection and tire deflection are considered in the controller synthesis. To minimize the seat acceleration, taking into account the vehicle body vertical acceleration and pitch acceleration, a new method for designing and solving static output-feedback H ! optimal controllers is proposed. First of all, a 5 DOF halfvehicle active suspension including active seat system model is presented. Then a direct, easily solved and effective method for static output-feedback H ! optimal control is presented. The controller is obtained by solving linear matrix inequality optimization problem and direct computation of related matrices. Finally, a numerical example is presented. The simulation results show that the controller proposed can achieve better performance compared with state feedback H ! optimal controller and the validity of the design method is verified.