Abstract

Human need safety, comfort, and speed in driving-requirements that can be fulfiled by autonomous vehicles that enable drivers to avoid obstacles and maintain a safe distance from other motorists. These function are executed through lateral vehicle control, which has been the subject of considerable research. The current research was aimed at providing a comprehensive review and description of previous investigations that implemented both conventional and innovative lateral control methods, such as proportional-integral-derivative control, fuzzy logic, artificial intelligence, neural networks, genetic algorithms, and combined approaches. The evaluated studies were also classified into two categories, namely simulation and experimental research that used real-world tools. The paper concludes with a recomendation to use an alternative method called direct inverse control. Which is a modification of neural network-based control. This method is advantageous because it uses output/input feedback, thereby effectively functioning in unpredicable terrain. This feature is highly suitable because autonomous vehicles are non-linear system.