Design of a Constrained Model Predictive Control Law to Improve Driving Comfort of a Quarter-Car Model

Abstract

In this study, a controller law is investigated using a quarter-car model to ensure that a performance measure expressing passenger comfort has the lowest value. In this controller law, kinematic and dynamic inequalities associated with the state vector and the control signal are considered. The dynamic and kinematic constraints are defined as: the magnitude of the force applied by the dynamic actuator placed between the vehicle body and the wheel, the magnitude of the displacement variable of the damper, the dynamic force generated by the elastic deformation of the wheel which is a numerical indicator of the vehicle's gripping ability. Abovementioned basic problem is solved with the predictive controller and the design of the model predictive controller working under the constraint conditions is transformed into a semidefinite programming optimization problem then a control algorithm is found to solve this optimization problem.

Keywords

• Model predictive control
• Linear quadratic regulator
• Quarter-car model
• Semi-definite programming