Robust longitudinal motion planning using vehicle model inversion

Abstract Trajectory planning for autonomous vehicles has significantly increased as more and more ADAS are included in modern cars. It A vehicle definitely needs to globally plan a trajectory, taking all the driving factors into account. For an autonomous terrestrial vehicle, this article proposes using an optimal trajectory on a specific range as a reference in a tracking loop. In previous studies, the optimization has been made using a Genetic Algorithm (GA), and the obtained trajectory has been injected into a Potential Field (PF) so as to be reactive to unforeseen events. Here, the previously developed GA-PF method is inserted in a new global planning and tracking method for longitudinal feedback control. An optimal trajectory is used as an input reference and a tracking schema is developed using an inverted bicycle model as an efficient feedforward control, and a robust controller to take the vehicle parameter variations into account. Autonomous car simulation results are given.