MPC-based Lateral Controller with Look-Ahead Design for Autonomous Multi-vehicle Merging into Platoon

In this paper a Model Predictive Control (MPC)-based approach is implemented to control the lateral maneuver of nonlinear vehicle dynamics. The controller is then applied to a multi-vehicle merging strategy for several vehicles to autonomously merge into a platoon using a variable-gap lateral reference trajectory. The MPC controller features a look-ahead design so that the controller could take actions in advance and thus improve the tracking objective. A combination of constraints, and slack variables-which quantify constraints violation-are chosen to guarantee a safe, efficient, and comfortable ride. The performance of the linear MPC controller is evaluated against a nonlinear plant to demonstrate a more realistic scenario.