Active Front Steering Using Stable Model Predictive Control Approach via LMI

Abstract

In this paper, a modelpredictive control (MPC) with guaranteed stability is used for controlling anactive front steering system in an autonomous vehicle. At each controlinterval, it is assumed that by using information obtained from the sensors onthe vehicle, a trajectory is known over a finite horizon, and then it is thepredictive controller duty to compute the front steering angle as the controlinput until the vehicle follows the trajectory. In addition, for reducingcomputational load of implementing the controller we convert the MPCoptimization problem into linear matrix inequality (LMI) form so the problemcan be solved more efficiently. The effectiveness of the proposed MPCformulation is demonstrated by simulation. A comparison is made between thetime consumed for solving LMI form of the problem and another approach ofsolving.