Application of Basis-Splines for Trajectory Planning in Highway Scenarios

Abstract:
Many approaches plan an automated vehicle’s future motion by solving an optimal control problem (OCP). This work discusses the criteria for the OCP design in automated highway driving. The criteria are applied to extend an existing spline-based motion planning algorithm with an additional target manifold and cost features. The resulting planning algorithm is applied to a merge-in maneuver, where the impact of cost weights on the driven trajectory and OCP complexity is evaluated. It is observed that the positive effects of additional features on the closed-loop trajectory are limited by the trajectory parameterization. However, it provides a regularization if the trajectory consists of multiple polynomial segments.