Real-time dynamic trajectory planning for highly automated driving in highways

Abstract

This paper presents the implementation of two methods for real-time trajectory planning in a dynamic environment applied to highly automated driving in a highway scenario. Both methods have been implemented for the HAVEit European project. The first method follows the Partial Motion Planning approach, and the second method uses 5th degree (quintic) polynomials to generate a detailed spatio-temporal description of a trajectory to be performed. Both implementations are integrated in a simulation environment and in an experimental research vehicle within HAVEit. Results and evaluations of the trajectory planning algorithms are presented.