Answers are grounded in this paper and cite their sources.
With the development of autonomous driving technology, the human–machine mixed driving environment has become the predominant form of future road traffic. This paper presents a systematic review of lane change decision-making in mixed driving scenarios. First, the behavioral characteristics of discretionary and mandatory lane changes are analyzed, and the lane change process is divided into decision-making and execution stages. From the perspective of driving safety, the importance of behavior prediction and risk assessment in ensuring the safety of decision-making is emphasized. It comprehensively reviews existing lane change risk evaluation methods, including probabilistic models and traffic conflict indicators, aiming to reduce traffic accidents caused by hazardous lane change behaviors through accurate risk evaluation. Then, through the analysis of existing lane change decision models, they are categorized into three major types: rule-based, data-driven, and game theory-based models. From the perspectives of input features and applied algorithms, the advantages, limitations, and applicable scenarios of models are compared and analyzed. Finally, current shortcomings and challenges are discussed. Key issues include insufficient consideration of human–machine interactions, low efficiency in multi-vehicle coordination, and high dependency on data. Future research directions are proposed to address these challenges. This study provides theoretical support for constructing safe and efficient lane change decision models in mixed traffic environments.
No takes yet. Share an insight, caveat, or question.
Huang et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: