Cooperative Merging Trajectory Optimization of Connected and Automated Vehicles in the Mixed Traffic: a Receding Horizon Control Approach

Abstract

The Cooperative on-ramp merging control of the connected and automated vehicle (CAV) can effectively address problems of traffic congestion, excessive energy consumption, and even traffic accident in the on-ramp merging areas. However, the uncertain maneuver of human-driven vehicles (HDVs) in mixed traffic scenarios brings trouble in making merging control for CAVs. In this paper, we proposed a receding horizon on-ramp merging control strategy for CAVs to address the mixed traffic merging control problem. First, control constrained optimal control problems for CAVs corresponding to two control modes are formulated. The flexible merging control mode optimizes merging trajectory with flexible merging time and position, while the mandatory merging control mode gives a fixed merging position to prevent the on-ramp CAV from driving beyond the longitudinal boundary. All the formulated optimal control problems are solved by Pontryagin’s minimum principle. Then, a receding horizon switching control framework is employed, in which the uncertain maneuver of the HDV is regarded as disturbance, thus CAVs collect HDV states, choose proper control modes, and replan their own trajectories repeatedly. Simulation results show that the proposed on-ramp merging control strategy can make CAVs merge flexibly in different scenarios and has the potential in improving safety, optimization, and robustness for the mixed traffic on-ramp merging control.