RoW-based Parallel Control for Mixed Traffic Scenario: A Case Study on Lane-Changing

Jingru Yu, Yi Yu, Shengyue Yao, Ding Wang, Pinlong Cai, Honghai Li, Li Li, Fei-Yue Wang, Yilun Lin

Abstract: To address the challenge of developing effective control methods for governing the entire driving process of both connected and automated vehicles (CAVs) and human-driven vehicles (HDVs) in mixed traffic scenario, this paper proposes a parallel control framework with calibrating module and pricing control module. Based on parallel learning theory, this framework enables intelligent agents to take over both HDVs and CAVs and engage in right-of-way (RoW) negotiations in the mixed traffic network. Game theory is employed to model RoW transactions and incentivize transactions through utility compensation. Empirical datasets are used to showcase the performance of the proposed framework in a lane-changing scenario. Our findings demonstrate that the introduction of incentives can increase the RoW trading probability of drivers significantly. To the best of our knowledge, this is the first parallel control framework based on pricing mechanism allowing both CAVs and HDVs to negotiate based on intelligent agents.