Study and programming a state estimator and a model predictive control for an autonomous formula student car
Tutor / director / evaluatorMorcego Seix, Bernardo
Document typeBachelor thesis
Rights accessRestricted access - author's decision
The main aim of this dissertation is to study and program a state estimator and a Model Predictive Control for an autonomous driving application. The problem statement relies on developing an autonomous vehicle to compete in Formula Student Driverless. To achieve that, two of the main pillars are the state estimation, to know the vehicle state for location and mapping the environment; and the vehicle controller, in this case, a Model Predictive Controller. Kinematic and Dynamic vehicle models using the bicycle model are studied and tested to validate the real-world correlation. These will be used for both state estimation and controller. Different approaches to Kalman Filters are studied in Sec. 3 State Estimator to analyse which is the best solution. Kalman Filter needs different sensors installed on the vehicle. For this reason, a study about the sensors has been done to find the most appropriate sensor given the system and algorithm requirements. At the end of the section, the experimental results are discussed. A cascade controller using MPC and PID controller or State Feedback for fast dynamics is discussed in the controller chapter. The results from the fast dynamics controller are exposed, discussing which fits better in the control scheme. Finally, the Model Predictive Control is explained. Both MPC controller and state estimator have been programmed in ROS and simulated using a Lap Time Simulator which simulation-reality error is less than a 1% in lap time. To obtain a full understanding of how both codes are implemented, the simulator structure is presented. Both EKF and MPC results are also exposed in the last chapter.
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder