search

layssi / Carla_Ray_Rlib

Carla-Simulator environment compatible with Ray/Rllib
GNU General Public License v3.0
35 stars 12 forks source link

Support for MARL algorithms #4

Closed rsuwa closed 4 years ago

rsuwa commented 4 years ago

Does this simulator environment support multi-agent reinforcement learning algorithms such as MADDPG?

layssi commented 4 years ago

The environment can be modified to support multi-agent rl but it would require major modifications where you would need multiple "hero" vehicles. The infrastructure is in the code but more work would be needed.