Trajectory optimization for Car-Like Vehicles in Structured and Semi-Structured Environments
Clemens Nietzschmann, Sebastian Klaudt, Christoph Klas, Devid Will, Lutz Eckstein · 2018
In this paper we propose a local trajectory planner for front steered car-like vehicles based on a combined direct optimization of the lateral and longitudinal vehicle guidance. The planner is designed for continuously optimizing a local trajectory based on a provided reference path in a structured or semi-structured driving environment. The planner respects constraints of the driving dynamics as well as actuator lim- itations and avoids static and dynamic obstacles. It is not restricted to a limited set of maneuvers. The implementation of the planner allows an online adaptation of the resulting driving behavior to satisfy different comfort or driving style demands. After Software-in-the-Loop simulations the algorithm was tested in two different real-world driving scenarios in ika’s automated vehicle which provides interfaces for full lateral and longitudinal control.