Is it possible to establish two RL agents?

Question

question

Martin K5 asked Jul 21, '23 Julie Weller commented Jul 31, '23

Is it possible to establish two RL agents?

Hi,

I would like to ask if it is possible to establish two reinforcement agents in FlexSim model? Let's say that I have two resources (forklifts) each working at the same warehouse (loading/unloading packages) and have specific contraints. From my point of view it would be better to have two RL agents one for each forklift.

Has anyone tried this?

Thank you,

Martin.

Software Version:

FlexSim 22.2.0

reinforcement learning resources agents

· 2

Joerg Vogel commented · Jul 21, 2023 at 04:06 PM

As I read throughly some of posted threads, they said you do it on Python program side, because only one program interpreter gets started on a windows machine. FlexSim is just interacting with this single running program, but within you can apply any possible structure to execute different routines.

2 ·

Julie Weller commented · Jul 31, 2023 at 01:53 PM

Hi @Martin K5, was Jeanette F's answer helpful? If so, please click the "Accept" button at the bottom of their answer. Or if you still have questions, add a comment and we'll continue the conversation.

If we haven't heard back from you within 3 business days we'll auto-accept an answer, but you can always comment back to reopen your question.

0 ·

Answer 1 · 2023-07-25T20:45:19Z

Jeanette F answered Jul 25, '23 Jeanette F edited Jul 25, '23

Hello @Martin K5,

I consulted Jordan Johnson. You can create a single AI that "drives" two objects. So the AI will observe for both forklifts and then issue actions for both forklifts. This does not scale well though. If you wanted to do this for more than 10 objects it wouldn't run well.

Joerg is correct. You cannot train two AI's at a time. There is a way to make something work with Python but that is beyond our expertise.

question