Adaptive model predictive controller (MPC) hybrid with braking torque distribution for trajectory tracking on autonomous vehicle: A review

Abstract

The main benefit of using a predictive controller (MPC) model is that it explicitly addresses various constraints. MPC’s ability to handle obstacles so is applied to avoid collision scenarios. On the other hand, MPC’s weakness is the complexity of the computational process. This causes the use of a long time in getting a solution. However, there is a potentially efficient solution for using MPC in real-time applications. This paper aims to analyze the use of adaptive MPC controllers used in autonomous vehicles in several scenarios. Adaptive MPC has been presented to deal with model uncertainty, which updates the plant model in real-time based on vehicle state measurement data. The adaptive MPC combined with the braking torque distribution controller can be used in autonomous vehicle control due to its superior performance compared to other geometric and optimized controllers. The combined advantage of this controller is due to its competence to navigate different road conditions with minimum computational costs.