Genetic Algorithm based Parameters Optimization of Model Predictive Control in Lane Keeping

Abstract

In this paper, a Genetic Algorithm (GA) approach is proposed to find optimal tuning parameters of a Model Predictive Control (MPC). The proposed parameter optimization method using GA has been demonstrated on an MPC for lane keeping. A bicycle model is utilized to track the predefined references. To evaluate said MPC, which was optimized using GA, it has been compared to an MPC for lane keeping, which was tuned by a human according to general guidelines.