The Department of Automation and Applied Informatics has been participant in the Eurobot robotics contest for years. In the competition, the robots have to solve different missions every year, but navigation is always one of the most important task.
In this thesis we will discuss a path following algorithm for mobile robots. The main part of the path following algorithm is the model based predictive controller. The tuning of the predictive controller's parameters are also explained in this paper.
The algorithm will be simulated in Webots robot simulator software, by using robot with differential drive. We will define a robot model, based on the kinematic equations of the the differential drive, and we will also design a predictive controller for a robot, of which behaviour is modelled by using system identication.
In the first part of the thesis, we will explain a Spline based path planning process. The path is used as a reference trajectory for the robot in the path following algorithm. In the trajectory generation we will use Quintic Bézier Spline. A velocity profil, based on the physical constraints of the robot, is also generated for the trajectory.