Flexible Unit A-star Trajectory Planning for Autonomous Vehicles on Structured Road Maps

Zahra Boroujeni1, Daniel Goehring2, Fritz Ulbrich2, Daniel Neumann2, Raúl Rojas3

  • 1Freie university
  • 2Freie Universität Berlin
  • 3Berlin University

Details

Category

Regular Paper

Sessions

10:48 - 12:36 | Tue 27 Jun | | TuBPl

Trajectory Planning for Autonomous Vehicles

Full Text

Abstract

In this paper we propose a trajectory planning approach for autonomous vehicles on structured road maps. It is using the well-known A* optimal path planning algorithm. We generate a safe optimal trajectory through a three-dimensional graph, considering the two-dimensional position and time. (1) The graph is generated dynamically with flexible distances between nodes, based on the vehicles velocity and a structured road map. (2) Furthermore the position of dynamic obstacles is predicted over time along the road lanes. The proposed Flexible Unit A* (FU-A*) algorithm was tested for real-time applications with execution times less than 50 ms. The feasibility and reliability of FU-A* is validated by implementing on simulated autonomous car of Freie university "MadeInGermany" using the roadmap of Tempelhof, Berlin.

Additional Information

No information added

Video

No videos found