OTFS-Assisted Sensing Adaptive Cruise Control for Highways: A Reinforcement Learning Approach

In this paper, we propose a novel channel estimation approach and driving decision method for adaptive cruise control (ACC) systems for vehicular networks, leveraging the properties of deep learning, reinforcement learning, and orthogonal time frequency space (OTFS) modulation. To achieve that, we propose to leverage deep learning (DL) to estimate motion parameters. Subsequently, we develop a reinforcement learning method to process the obtained target motion information to enable adaptive vehicle-following strategies. This ensures robust decision-making and precise control under dynamic and uncertain driving conditions, achieving superior performance in terms of both accuracy and reliability.