In my current understanding, we can use On Observation in Reinforcement Learning Tools to allow my Agent to observe the current system status.
The official example is as follows:
I now want to observe the values in the GlobalTable through On Observation as the States for my current reinforcement learning, but I don’t know why my code cannot make the values of Parameters change together with the values in my GlobalTable. I want to know how to use FlexSim Does China allow such an operation?
Here is my example: