ABSTRACT
In this chapter, the authors proposed a new adaptive and robust MPC based on classical MPC and min-max MPC. The proposed MPC is called adaptive min-max MPC (AMM-MPC). This controller can deal with inherent slip through adaptation. In AMM-MPC, first a cost function is defined and then, the min-max technique is applied based on the bounds of the disturbances. The control law is adapted based on the curvature of the reference path and path offset feedback as the adaptive part. First, the derivation and stability of AMM-MPC are presented. Then, the performance of the proposed MPC is compared with the classical MPC in kinematic simulations and dynamic simulations. The dynamic simulation platform is developed to create a more realistic environment incorporating slip phenomena. Then, AMM-MPC and classical MPC are implemented on a tractor in the field, and results are compared and discussed. To compare the performance of the proposed controller with other types of controllers, the experiments were extended to include comparison of AMM-MPC with a SMC.
