Graph-Based Local Trajectory Planner
The graph-based local trajectory planner is python-based and comes with open interfaces as well as debug, visualization and development tools. The local planner is designed in a way to return an action set (e.g. keep straight, pass left, pass right), where each action is the globally cost optimal solution for that task. If any of the action primitives is not feasible, it is not returned in the set. That way, one can either select available actions based on a priority list (e.g. try to pass if possible) or use an own dedicated behaviour planner.
The planner was used on a real race vehicle during the Roborace Season Alpha and achieved speeds above 200kph. A video of the performance at the Monteblanco track can be found here.
Disclaimer
This software is provided as-is and has not been subject to a certified safety validation. Autonomous Driving is a highly complex and dangerous task. In case you plan to use this software on a vehicle, it is by all means required that you assess the overall safety of your project as a whole. By no means is this software a replacement for a valid safety-concept. See the license for more details.
Documentation
The documentation of the project can be found here.
Contributions
[1] T. Stahl, A. Wischnewski, J. Betz, and M. Lienkamp, “Multilayer Graph-Based Trajectory Planning for Race Vehicles in Dynamic Scenarios,” in 2019 IEEE Intelligent Transportation Systems Conference (ITSC), Oct. 2019, pp. 3149–3154.
(view pre-print)
Contact: Tim Stahl.
If you find our work useful in your research, please consider citing:
@inproceedings{stahl2019,
title = {Multilayer Graph-Based Trajectory Planning for Race Vehicles in Dynamic Scenarios},
booktitle = {2019 IEEE Intelligent Transportation Systems Conference (ITSC)},
author = {Stahl, Tim and Wischnewski, Alexander and Betz, Johannes and Lienkamp, Markus},
year = {2019},
pages = {3149--3154}
}
6 Feb 28, 2022
24 Dec 9, 2022
8 May 22, 2022
6 Oct 27, 2022
117 Dec 28, 2022
6 Jul 18, 2022
25 Sep 26, 2022
129 Jan 4, 2023
37 Dec 16, 2022
31 Oct 28, 2022
0 Oct 21, 2022
4 Dec 15, 2022
44 Jun 27, 2022
114 Dec 29, 2022
13.8k Jan 5, 2023
2 Feb 10, 2022
328 Dec 26, 2022
0 Nov 1, 2022
49 Nov 28, 2022