Non-linear MPC Motion Planner for Autonomous Vehicles Based on Accelerated Particle Swarm Optimization Algorithm

Automotive field is about to face a deep revolution: autonomous driving. This paper presents an algorithm of motion planning that can be considered as a part of a hierarchical scheme describing the complex task of driving. The algorithm proposed consists in an optimization problem by means of a MPC formulation, numerically solved through Accelerated Particle Swarm Optimization (APSO). It can properly handle with urban environment considering the presence of moving obstacles as well as constraints (e.g. vehicle dynamics, road boundaries,:.). Simulation results confirm the performances of the algorithm.