Abstract
Vehicle steering dynamics show resonances, which depend on the longitudinal speed, unstable equilibrium points and limited stability regions depending on the constant steering wheel angle, longitudinal speed and car parameters. The main contribution of this paper it to show that a combined decentralized proportional active front steering control and proportional-integral active rear steering control from the yaw rate tracking error can assign the closed loop eigenvalues, without lateral speed measurements, using a standard single track car model with nonlinear tire characteristics and a nonlinear first order reference model for the yaw rate dynamics driven by the driver steering wheel input. By choosing a suitable non linear reference model it is shown that the responses to driver step inputs tend to zero lateral speed for any value of longitudinal speed: however the resulting static closed loop gain from driver input to yaw rate differs from the open loop gain at higher speed. The closed loop system shows the advantages of both active front and rear steering control: higher controllability, enlarged bandwidth for the yaw rate dynamics, suppressed resonances, new stable cornering manoeuvres, enlarged stability regions, reduced lateral speed and improved manoeuvrability; in addition comfort is improved since the phase lag between lateral acceleration and yaw rate is reduced. For the designed control law a robustness analysis is presented with respect to the critical car parameters such as mass, moment of inertia and front and rear cornering stiffness coefficients. Several simulations are carried out on a higher order experimentally validated nonlinear dynamical model to confirm the analysis and to explore the robustness with respect to unmodelled dynamics.
Keywords:Vehicle Dynamics, Control System, Active Steering