Why does the Reinforcement Learning 'sim' command terminate early?

Why does the reinforcement command 'sim' terminate early? Please refer to the link below for more information on 'sim':

sim - Simulate trained reinforcement learning agents within specified environment - MATLAB

Answer 1

[MathWorks Support Team]

The 'sim' command for a Reinforcement Learning environment terminates early because the default number of 'MaxSteps' in 'sim' is 500. To increase the number of steps taken in 'sim', use:

>> simOptions = rlSimulationOptions('MaxSteps', <number of steps>);

Then simulate using the command:

>> sim(env, agent, simOptions);

More information on 'rlSimulationOptions' can be found by referencing the link below:

rlSimulationOptions - Options for simulating a reinforcement learning agent within an environment - MATLAB

More information on 'sim' can be found by referencing the link below:

sim - Simulate trained reinforcement learning agents within specified environment - MATLAB