Abstract: This paper establishes a game theory-based lane-changing behavior model of intelligent vehicles and uses a bi-level programming algorithm to analyze the model in order to solve the decision-making problem of lane-changing behavior of intelligent vehicles in traditional environment and networked environment. The data gathered by the driving simulator and HighD database is used to simulate and analyze the model in this work. In both the traditional and networked environments, the model is able to forecast the best course of action for every vehicle involved in lane changes. The model's prediction results can aid in improving traffic efficiency and safety, serve as a valuable reference for the design and implementation of intelligent transportation systems, and provide recommendations for the intelligent vehicles' lane-changing techniques.

Key words: Lane-changing decision; Game theory; Harsanyi transform; Driving simulator; Bi-level programming model

摘要： 为解决智能车辆在传统环境及联网环境中换道行为的决策问题，文章基于博弈论建立 智能车辆的换道行为模型，并使用双层规划算法对该模型进行分析。使用 HighD 数据库和驾 驶模拟器采集的数据对模型进行了仿真分析。结果表明，该模型可以预测在联网环境和传统 环境下车辆换道行为中各参与者的最优策略，模型的预测结果可以为智能车辆的换道策略选 择提供指导，为智能交通系统的设计和实施提供重要参考，有助于提高交通效率和驾驶安全 性。

关键词: 换道决策；博弈论；海萨尼变换；驾驶模拟器；双层规划模型