Comparison between Model Predictive (MPC) and Model Reference Adaptive Controllers (MRAC) for Electrohydraulic Steering System Implemented as Real-Time Simulink® Program

The purpose of the article is to investigate tracking performance of steering cylinder when controlled with two nonlinear controllers. The system under study is a laboratory electrohydraulic steering test rig. Two controls strategies are examined - model reference adaptive control (MRAC) and model predictive control (MPC). The MPC controller design is based on an identified model from experimental data in open-loop. The model is represented in state-space form and the future values of the state trajectory and output variables are used as optimization objective with respect to the future values of the control signal. Since this optimization procedure have to be performed every sampling interval and involve iterative calculation the implementation of MPC requires considerable computational resources. Hence, we have decided to develop a distributed computational platform, where the controller is executed in real-time as a Simulink model. The control and measurement signals are processed by the PLC and are synchronized with the real-time model over high-speed CAN channel. The process model for the MRAC is assume to be a simple integrator with unknown gain, possibly slowly varying with time due to physical parameter drift. The MRAC is not so computationally intensive as MPC, however for the sake of comparison it is implemented in the same architecture. The structure of the MRAC is selected with respect of the Lyapunov stability criteria to guarantee asymptotic convergence of the cylinder position error.