Optimal Control Based Trajectory Planning for an Autonomous vehicle on Curvy Roads

Abstract

We proposed a trajectory optimization method for curvy road environments containing irregular embedded obstacles. A novel framework is developed to ensure safety and smoothness of trajectory by combining a Spatial Safety Corridor (SSC) with an Optimal Control Problem (OCP). The Spatial Safety Corridor (SSC) is first constructed to exclude obstacle regions from the drivable area, forming a collision-free corridor that guides feasible motion. Within this corridor, an Optimal Control Problem (OCP) is formulated based on a kinematic bicycle model to generate a trajectory that satisfies both feasibility and safety constraints. Simulation results illustrate that the proposed framework effectively handle diverse obstacle configurations, generating trajectories that are smooth and safe. Furthermore, the proposed framework demonstrates high computational efficiency, making it suitable for real-time trajectory planning.