Dual-Loop Tube-based Robust Model Predictive Control for Active Suspension System with Parameter Uncertainty

Abstract

Active vehicle suspension is a promising technology for achieving better ride quality and road handling in vehicles. Model predictive control (MPC) has been widely used in active suspension system (ASS) as it can explicitly introduce system constraints. In addition to the MPC, the robust controller design is also essential in dealing with parameter uncertainties, which are inevitable in the ASS. Together, the offline calculation and real-time constraint simplification make it possible for the robust model predictive control of the ASS. To better control the ASS subject to a random road profile excitation with model parameter uncertainties, in this paper we presented a novel dual-loop tube-based robust model predictive control (DTRMPC) structure, which incorporated a linear quadratic regulator (LQR) controller in the inner loop and a MPC controller in the outer loop. Noticeably, this dual-loop structure utilized the combination of offline and online calculation and had better robustness. Then, we’ll be able to achieve better active suspension performance by solving constrained optimization problems through quadratic programming (QP). The numerical simulations indicated that the ASS with a DTRMPC algorithm is superior to those with a traditional MPC controller in the low-frequency range to which human beings are most sensitive, and it has improved robustness under uncertain model parameters.