Code developed for the work published in "C. Tiriolo, W. Lucia - On the Design of Control Invariant Regions for Feedback Linearized Car-Like Vehicles - ACC - LCSS 2023"

Full paper at: https://doi.org/10.1109/LCSYS.2022.3224680

Consider a self-driving car whose kinematics are described by bicycle models. Design a tracking controller with associated control invariant region for the tracking error of a feedback linearized car-like model guaranteeing stable full-state tracking while fulfilling the time-varying input constraints differential-drive robot subject to input constraints, i.e., limits on longitudinal and angular steering velocity of the car (see full paper for furthe details).

The code was tested on Matlab 2020a environment and it requires Ellipsoidal Toolbox ET (https://www.mathworks.com/matlabcentral/fileexchange/21936-ellipsoidal-toolbox-et).

• Invariant_set_Trajectory_Tracking_Control_approachACC23.m: It is the main script to run in order to simulate the algorithm developed in the paper.
• DiffDrive.m: It's a Matlab function that implements the bicycle kinematics. It's used by the method ode45 to solve the differential equations describing the car's motion in the plane.
• Simulink_model_parameters.m: It's a script implementing all the parameters necessary to simulate the algorithm on Simulink.
• Car_like_robot_trajectory_tracking.slx: It's a Simulink model that simulates the tracking control algorithm developed in the paper.

• Run "Invariant_set_Trajectory_Tracking_Control_approachACC23.m"

• Run the script "Simulink_model_parameters.m"
• Run the Simulink model Car_like_robot_trajectory_tracking.slx (required Matlab 2020a or later versions)