作者: Eugenio Alcalá , Vicenç Puig , Joseba Quevedo
DOI: 10.1016/J.ROBOT.2019.103392
关键词: Simple (abstract algebra) 、 Track (rail transport) 、 Smoothing 、 State (computer science) 、 Focus (optics) 、 Computer science 、 Mathematical optimization 、 Obstacle avoidance 、 Structure (mathematical logic)
摘要: Abstract In this paper, we present an effective online planning solution for autonomous vehicles that aims at improving the computational load while preserving high levels of performance in racing scenarios. The method follows structure model predictive (MP) optimal strategy where main objective is to maximize velocity smoothing dynamic behavior and fulfilling varying constraints. We focus on reformulating non-linear original problem into a pseudo-linear by convexifying function vehicle equations be expressed Linear Parameter Varying (LPV) form. addition, ability avoiding obstacles introduced simple way with reduced cost. test compare proposed against its approach through simulations. testing trajectory scenario. First, case free track afterwards scenario including static obstacles. Simulation results show effectiveness reducing algorithm elapsed time finding appropriate trajectories under several input/state