Towards Safer and Smarter Integration of Conflict Management Algorithms: A Quantitative and Comparative Study

Autonomous vehicles (AVs) face significant challenges in overtaking maneuvers, where balancing safety, efficiency, and adaptability is crucial. This paper provides a quantitative and comparative analysis of conflict management algorithms, including Reinforcement Learning (RL), Model Predictive Control (MPC), and Game Theory, assessing their strengths and limitations in handling decision-making under uncertainty. Through bibliometric evaluation and performance assessment, we highlight gaps in existing methods and propose a hybrid approach that integrates multiple decision-making paradigms to enhance robustness, adaptability, and real-time responsiveness. This study lays the groundwork for more intelligent and reliable overtaking strategies, addressing a critical aspect of AV decision-making.