FabianPlum

FabianPlum/CrowdRL

MARL-learned pedestrian navigation policies trained in procedural 2D environments, validated against IAS-7 experiment data. Augments hand-crafted locomotion models in JuPedSim.

0 stars 1 forks 0 watchers Python MIT License
View on GitHub
2 Open Issues Need Help Last updated: Aug 3, 2026

Open Issues Need Help

View All on GitHub
Preferred-speed fallback is 1.34 in the ego block and 1.3 elsewhere 2 days ago
bug good first issue
FabianPlum/CrowdRL
0

MARL-learned pedestrian navigation policies trained in procedural 2D environments, validated against IAS-7 experiment data. Augments hand-crafted locomotion models in JuPedSim.

Python
chest_depth default duplicated between the adapter and the lockstep instrument 2 days ago
enhancement good first issue
FabianPlum/CrowdRL
0

MARL-learned pedestrian navigation policies trained in procedural 2D environments, validated against IAS-7 experiment data. Augments hand-crafted locomotion models in JuPedSim.

Python