AUTHOR: Francesco Conversano
RELATOR: Stefania Santini
TUTOR: Aniello Mungiello
ABSTRACT: One of the main objectives in control theory is to design a controller that can meet the design specifications and that can provide a control input appropriate to the type of actuator used in the system. For this reason, the control output is sometimes seen as a cost function to be minimised, without losing performance. It is with this in mind that the problem of 'Optimum Control' is posed, which aims to stabilise a dynamic system against a desired reference input by minimising one or more parameters involving the control scheme. Model Predictive Control, MPC for short, is an advanced control method that satisfies the assumption of optimal control over a predefined time horizon. This is widely used in the automotive domain, specifically with self-driving or assisted cars, where the control action for vehicle acceleration and steering is often subject to dynamic constraints due to the physical limitations of the vehicle, environmental conditions and surrounding obstacles. This thesis proposes the design and application on an AgileX 'Hunter SE', of an adaptive cruise control, developed with a Model Predictive Control subject to constraints on state and control output.
