Lane Change

Lane change maneuvers are a critical aspect of autonomous driving, with research focusing on accurately predicting driver intentions and generating safe, efficient, and comfortable trajectories. Current studies employ diverse machine learning approaches, including deep reinforcement learning (e.g., DQN, TD3, PPO, SAC), long short-term memory networks (LSTMs), and large language models (LLMs), often incorporating sensor fusion and knowledge graph embeddings to improve prediction accuracy and explainability. This research is crucial for enhancing the safety and efficiency of autonomous vehicles and improving traffic flow, with implications for both the development of advanced driver-assistance systems and the broader understanding of human driving behavior.