Fault-Tolerant Trajectory Tracking Control for Autonomous Vehicle Based on Camera and GPS

The aim of this paper is to design a fault-tolerant trajectory tracking control for an autonomous vehicle based on a camera and Global Positioning System (GPS) unit. The trajectory tracking control has been founded on the Linear Quadratic control framework. The camera and GPS units calculate the lateral error of the vehicle. A standard webcam is used for lane detection and the lateral error of the vehicle is calculated by using the Hough Transform. A GPS sensor, which is the TruckSim model is used to calculate lateral error by comparing it with a reference database. In the case of lateral error calculated from the camera system is unstable and increased abruptly, the camera unit is faulty and lateral error that is provided by the GPS unit is reliable. Then, a reliable sensor signal is used to calculate the lateral error of the vehicle in order to perform trajectory tracking. The TruckSim dynamic vehicle simulation software is used with a real-time Simulink/MATLAB model for validation of this study. During the simulation, extra distortions have been integrated to perform camera fault and the GPS sensor has been considered as a reliable sensor. Results show that the system has performed successful trajectory tracking with the fault scenario.