• autonomous driving
• mpc
• predictive control
• trajectory tracking
• vehicle modeling

In this thesis deals with the problem of autonomous driving of a vehicle.
The first objective was to model a vehicle from the kinematic (position and speed) and dynamic (forces and moments) points of view.
On the two models obtained, a control system was designed to allow the vehicle to move independently on a straight road with possible obstacles and constraints.
Subsequently, these models were modified to solve, through two different algorithms, the problem of trajectory tracking. Trajectory tracking was therefore possible through the continuous updating of the state vector to be reached or through the use of two additional equations that analysed the error on position and angle with respect to a reference value. In particular, the physical behaviour of the various systems in relation to their computational cost was analysed.
The main components of a modern autonomous vehicle are location, perception of obstacles and the surrounding environment and control. In this thesis we will discuss acceleration, brake and steering control of the vehicle using Model Predictive Control (MPC).
The predictive control model (MPC) is an advanced method of process control that is used to control a process by satisfying a series of constraints. The main advantage of MPC is that this control has the ability to anticipate future events and can take control actions accordingly. PID controllers for example do not have this predictive capability.
The proposed approach takes advantage of two very simplified models compared to a real vehicle model, these models will therefore be less computationally expensive than the more complex existing methods using vehicle tyre models.