Cooperative Ramp Merging Strategy at Multi-Lane Area for Automated Vehicles

Traffic congestion is a serious social problem, while the multi-lane merging is a common and typical road structure that easily causes traffic congestion. The advent of connected and automated vehicles (CAVs) enables vehicle cooperation, which can significantly improve traffic conditions. In this paper, a cooperative ramp merging strategy for CAVs at multi-lane area with three lanes on main road and one lane on ramp is proposed based on cooperative game theory with transferable utility (TU) and optimal control. On main road, the lanes are allocated based on the driving payoff of each vehicle via cooperative game, while the first-regularization-then-action (FRTA) method is introduced to achieve the real-time cooperative lane changing based on optimal control. Afterwards, the merging sequence (MS) containing vehicles on inner lane and ramp is also determined by cooperative game and the merging trajectory is computed via optimal control. The combination of game theory involving side payment and optimal control achieves a win-win situation at the decision-making level and the cooperative control at the execution level. The simulation results of different cases verify the feasibility and effectiveness of proposed framework, while the comparison result indicates that the proposed framework can improve the traffic efficiency and economic benefit of transportation system.