输入状态稳定的鲁棒预测控制

以有界干扰非线性系统为研究对象,设计一种基于近似可达集的鲁棒预测控制方法。该方法以鲁棒控制不变集作为终端约束集,采用一种简单的三次多项式逼近预测控制的待优化控制律,通过在线优化求解三次多项式的各项系数,并从理论上证明了所设计的鲁棒预测控制律可以使系统输入状态稳定。最后通过仿真实例验证了所提出的鲁棒预测控制方法的可行性和有效性。

Robust model predictive control with input-to-state stability

Considering the nonlinear system with bounded disturbance, a robust model predictive control(MPC) method
based on approximated reachable sets is designed. A robust control invariant set is taken as the terminal constraint set. A
kind of cubic polynomial with simple structure is adopted to approximate the controller to be optimized. Then, the coefficients
of cubic polynomial are computed through online optimization. It is proved in theory that the system is input-to-state stable
by using the proposed robust MPC controller. Finally, simulation results show the feasibility and effectiveness of the method.