Global parameters / data store memory with RL agent block Simulink

2 visualizaciones (últimos 30 días)
I am using a Simulink model for reinforcement learning (RL). I have created the environmnet using createdIntegratedEnv. I would like to calculate the change in error at each sample time step and feed this as an observation to the RL agent. I have tried using a data store memory block in the model environmnet (i.e. the environment which models the plant considered) to keep track of the previous error value, however I receive an error stating that I cannot use this block with a RL environment.
Any insight on how to update a global parameter in the model within the RL environment would be greatly appreciated.

Respuestas (1)

Emmanouil Tzorakoleftherakis
Emmanouil Tzorakoleftherakis el 21 de Dic. de 2020
Not sure what error you are seeing, but if you only need to use the value of the previous time step, I think the Memory block is what you should use.

Productos


Versión

R2020b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by