A Comparative Analysis of Machine Learning and PID Controllers for Implementing Adaptive Cruise Control in Autonomous Vehicles

Adaptive Cruise Control (ACC) is a widely adopted Advanced Driver Assistance System (ADAS) technology that enhances driving comfort and safety. As part of the broader effort to advance autonomous driving, this paper investigates the imple-mentation of ACC systems using two approaches: a standard PID controller, valued for its simplicity, reliability, and widespread use, and a machine learning-based model, designed to offer similarly straightforward usability. The experiments were carried out within the CARLA simulator, a platform that provides a controlled environment for testing and evaluation. By comparing the performance of these two methods, the study aims to evaluate whether machine learning can achieve improved results while retaining the ease of implementation associated with traditional PID controllers. This comparative study highlights the strengths and limitations of both approaches, offering valuable insights into how machine learning can enhance ACC functionality and contribute to safer autonomous driving systems.