Design, Validation and Implementation of an Adaptive Model Predictive Control for an Autonomous Racing Vehicle

In recent years, automotive control has become a significant factor in automotive innovation. Over the years, it has been necessary to use automotive control to meet lower fuel consumption and lower missions demands. Besides, automotive control ensures greater driving safety and comfort. In any technology area, control design is a fusion of reality, physics, modeling, and design methods. This also happens in automotive control, where extensive research and development has led to numerous descriptions, models, and design methodologies suitable for control. An autonomous vehicle's control consists mainly of three separate modules: environment perception, planning and decision-making, and vehicle control. This master thesis aims to design, validate, and implement a vehicle control strategy to automatically manage the lateral and the longitudinal dynamics of a full electric all-wheel drive racing vehicle participating in Formula SAE. The vehicle control strategy is based on Model Predictive Control, using a combined model that manages both the lateral and longitudinal dynamics, or using two separated MPC, one for the lateral dynamics and the other one for the longitudinal dynamics. In the first part of the thesis work, vehicle dynamics is modeled using a three degrees of freedom vehicle model, and simulations are performed in virtual scenarios created using Automated Driving Scenario Toolbox on MATLAB and Simulink. In the second part, there is an integration of the Path Planning and the Vehicle Control Strategy based on acquisitions performed in real-time. The third part of the thesis work is performed on Simscape Vehicle Template with the vehicle dynamics modeled using a complete vehicle model. The last part of the work consists of the implementation of the developed controller into Real-Time Target Hardware Platforms.