Abstract: A robust gain-scheduling yaw moment control scheme is proposed based on linear parameter-varying (LPV) modeling method. An error dynamic model for vehicle yaw moment control is built for controller design. Then the error dynamic model is transformed into a LPV model with surface adhesion coefficient, longitudinal velocity and the combination of the two as varying parameters by selecting surface adhesion coefficient and longitudinal velocity as scheduling variables for gain-scheduling control. Considering the LPV model with four varying parameters, the design of the controller is transformed into the problem of linear H controller design for each of the 16 vertices of the polytopic model. 16 vertical H controllers are designed by solving 17 linear matrix inequalities, for each of which the closed-loop system is quadratic stability with quadratic H performance and the controller for a certain working condition is obtained online by weighting the 16 vertical controllers. Test results obtained by numerical simulations on an 8-DOF nonlinear vehicle model demonstrate that the gain-scheduling based H controller is much better in adaptability to the variation of operating conditions than the single H controller.

Key words: Gain-scheduling control, Vehicle stability, Vehicle system dynamics, Yaw moment control