Does Reinforcement Learning Toolbox support discretization in space, instead of time?
1 view (last 30 days)
Show older comments
Hello,
I have noticed that almost every example in the 'Reinforcement Learning Help Center' uses a fixed sample time Ts, and the Episode Steps equals to Tf divided by Ts (Tf/Ts). But now I have a problem that a vehicle runs on a hilly road of which the total length is 10km. The action of RL agent is the torque from vehicle motor. If I set a fixed sample time, lets say Ts = 1s, and Tf = 600s, I cannot guarantee that the vehicle reaches destination right on the moment of 600s. So I need to discretize the total length of 10km into 1000 segments, each segment ( delta_S ) equals to 10 meters. I was wondering does the Rinforcement Learning Toolbox in Simulink support this operation?
I really appreciate if someone could help me!
Thanks very much !!!!
1 Comment
Emmanouil Tzorakoleftherakis
on 22 Apr 2021
I don't think this is possible but at the same time I am not sure why you would want to do that. If you action is torque, wouldn't the whole point be to apply sufficient torque such that the vehicle reaches its destination within the specified time?
Answers (0)
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!