Connected Autonomous Vehicle Control Considering Human Driving Aggressive Lane Change From a Cyber-Physical Perspective

The aggressive lane-changing behavior of human-driven vehicles (HVs) under mixed traffic has a negative impact on connected autonomous vehicles (CAVs). This behavior is frequently observed in the vicinity of signalized intersection. From the perspective of cyber-physical systems (CPS), the negative impact on driving safety can be eliminated through the control algorithm at the CPS micro-scale level; and the negative impact on traffic efficiency can be suppressed through the CAV decision-making strategy at the CPS meso-scale level. This paper introduces a two-stage method aimed at safeguarding the right of way for CAVs and optimizing traffic efficiency in the scenario. Additionally, A CAV safety controller is developed to effectively handle the challenges posed by HVs violating game rules through continued aggressive lane changes. Simulation experiments demonstrate that compared to conventional game algorithms, the proposed method enhances traffic flow, increases average speeds, and diminishes traffic congestion in this scenario. The designed controller can ensure the safety of CAV driving in the target lane.