A nonlinear programing model for collision-free lane-change trajectory planning based on vehicle-to-vehicle communication

This study focuses on the trajectory planning problem for automatic lane-change maneuvers, for which we propose a nonlinear programing model to generate the collision-avoidance lane-change trajectory in a spatiotemporal space. In this model, we regard a lane-change trajectory as the combination of a lane-change path and its velocity profiles. Based on the polynomial curve, the lane-change path is planned first in 2-dimensional Cartesian coordinates to connect the initial position with final position, and then a nonlinear mathematical programing model is used to generate the velocity profiles for maintaining the driving safety and comfort. To solve the nonlinear model efficiently, we design an inverse strategy to set the initial guess that can coordinate with the discrete scheme of the trajectory planning problem. Moreover, as the vehicle shifts from a lane-change maneuver to a car-following maneuver at the end of the lane-change process, the planned final acceleration is constrained from the car-following model. Finally, a series of numerical examples are provided, and the results indicate that the automated vehicle can drive along the planned lane-changing trajectory safely and efficiently.