Value Iteration on the Cart Pole Domain

Previous slide

Next slide

Back to first slide

View graphic version

Cart pole provide a rather inelegant solution to this problem. The command

to produce this graph is

nrdp cartpole vi statediscmeth simplex statediscres 15 actiondiscmeth simple

actiondiscres 9 x0 1 -0.5 0 0 graphstates