Adaptive optimal control for suppressing vehicle longitudinal vibrations

Sudden torque change under the tip-in operation often causes driveline low-frequency torsional vibrations, which seriously impacts vehicle drivability. Typical driveline resonance frequency is under 10Hz in the longitudinal direction and it cannot be eliminated through mechanical design optimization. To provide a smooth acceleration with minimal vibrations, an adaptive optimal tracking controller of engine torque is designed in this paper. A nonlinear model, elaborating the driveline and vehicle longitudinal dynamics, is developed. Based on the linearized control-oriented model, a receding horizon linear quadratic tracking (RHLQT) controller is designed along with the Kalman optimal state estimation. The optimal control design parameters (weightings) are tuned under different road conditions. In addition, the road surface contact friction coefficient is estimated using the recursive Least-Squares method. The RHLQT adapts to the estimated road condition (surface friction). The control performance of the adaptive RHLQT is studied under different road conditions, compared with fixed control parameters LQT controllers. The simulation results confirm the effectiveness of the proposed control scheme.