Is it practicable to train multiple agents simutaneously using RL Toolbox？

Here is a question about applying multiple agents simutaneously when training using RL Toolbox.

Background: I have been trying to apply reinforcement learning method to multiagent control in which scenario communication between agents is required. So I need to train agents simultaneously while they exchange data/ message with each other，like states（observations）and reward.

According to my knowledge，there only can be one single RLagent in one Simulink file while creating environment. Besides，function ‘train’ seems not to support to train several agents simultaneously while using several Simulink files.

So how to solve this problem? Any suggestions would be really appreciated!

[Selva Kumar]

I too got the same question. Help needed !!!

Answer 1

[Ari Biswas]

Multi-agent training is currently not supported, however, it will be soon in a future release.

[Shi yu]

Thanks very much for your answer!

Answer 2

[Emmanouil Tzorakoleftherakis]

As of R2020b release, Reinforcement Learning Toolbox lets you train multiple agents simultaneously in Simulink. Please see following examples for reference:

1. Train Multiple Agents for Path Following Control
2. Train Multiple Agents for Area Coverage
3. Train Multiple Agents to Perform Collaborative Task

Hope that helps

[Shi yu]

Thanks very much for the answer! Really helps!