burlap.domain.singleagent.cartpole

Class InvertedPendulum.InvertedPendulumRewardFunction

java.lang.Object

burlap.domain.singleagent.cartpole.InvertedPendulum.InvertedPendulumRewardFunction

All Implemented Interfaces:

RewardFunction

Enclosing class:

InvertedPendulum

public static class InvertedPendulum.InvertedPendulumRewardFunction
extends java.lang.Object
implements RewardFunction

A default reward function for this domain. Returns 0 everywhere except at fail conditions, which return -1 and are defined by the pole being grater than some threshold (default PI/2 radians.

Author:

James MacGlashan

Constructor Summary

Constructors

Constructor and Description
InvertedPendulum.InvertedPendulumRewardFunction()
InvertedPendulum.InvertedPendulumRewardFunction(double maxAbsoluteAngle) Initializes with a max pole angle as specified in radians

Method Summary

Methods

Modifier and Type	Method and Description
double	reward(State s, GroundedAction a, State sprime) Returns the reward received when action a is executed in state s and the agent transitions to state sprime.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

InvertedPendulum.InvertedPendulumRewardFunction

public InvertedPendulum.InvertedPendulumRewardFunction()

InvertedPendulum.InvertedPendulumRewardFunction

public InvertedPendulum.InvertedPendulumRewardFunction(double maxAbsoluteAngle)

Initializes with a max pole angle as specified in radians

Parameters:

maxAbsoluteAngle - the maximum pole angle in radians that causes task termination/failure.

Method Detail

reward

public double reward(State s,
            GroundedAction a,
            State sprime)

Description copied from interface: RewardFunction

Returns the reward received when action a is executed in state s and the agent transitions to state sprime.

Specified by:

reward in interface RewardFunction

Parameters:

s - the state in which the action was executed

a - the action executed

sprime - the state to which the agent transitioned

Returns:

the reward received when action a is executed in state s and the agent transitions to state sprime.