Firefly algorithm-based nonlinear MPC trajectory planner for autonomous driving

The automotive ecosystem is experiencing a technological leap forward. The science and technology progress are converging to a new disruptive scenario, where the autonomous driving is the leading player. An autonomous vehicle can be described as a dynamical system, that manages its own state by performing a sense-plan-act loop. In this paper a hierarchical structure of the planning module is presented, with a specific focusing on the motion planning layer. This latter is based on a Model Predictive Control (MPC) strategy, where the optimization process is carried out by exploiting a Firefly-Algorithm (FA). This approach is able to make purposeful decisions in order to guide the vehicle towards the designed goal in an urban environment, within the imposed constraints (e.g. vehicle dynamics, road boundaries, obstacles avoidance, etc). The optimal control problem formulation is provided as well as the mathematical model of the vehicle. Simulation results are obtained to assess the controller performance and to validate the feasibility and the safety of the generated trajectories.