burlap.behavior.singleagent.learning.modellearning

Class ModeledDomainGenerator.ModeledAction

java.lang.Object

burlap.oomdp.singleagent.Action

burlap.behavior.singleagent.learning.modellearning.ModeledDomainGenerator.ModeledAction

Enclosing class:

ModeledDomainGenerator

public class ModeledDomainGenerator.ModeledAction
extends Action

A class for creating an action model for some source action. This class copies source action's name, parameters, and parameter order groups. This action's preconditions are satisfied whenever the source action's preconditions are satisfied. The class's execution and transition dynamics methods are determined using the specified source model. Optionally, the action may be set to use RMax in which transitions that are "unknown" by the model take the the agent to a special RMax state that can only transition to itself.

Author:

James MacGlashan

Field Summary

Fields

Modifier and Type	Field and Description
protected Model	model The model of the transition dynamics that specify the outcomes of this action
protected State	RMaxState The fictitious RMax state this action will transition to for unknown transitions.
protected Action	sourceAction The source action this action models
protected boolean	useRMax Whether this action follows the RMax paradigm of transition to a fictious RMax state when the model is not known

Fields inherited from class burlap.oomdp.singleagent.Action

actionObservers, domain, name, parameterClasses, parameterOrderGroup

Constructor Summary

Constructors

Constructor and Description
ModeledDomainGenerator.ModeledAction(Domain modelDomain, Action sourceAction, Model model, boolean useRMax) Initializes.

Method Summary

Methods

Modifier and Type	Method and Description
boolean	applicableInState(State s, java.lang.String[] params) Returns true if this action can be applied in this specified state with the specified parameters.
java.util.List<TransitionProbability>	getTransitions(State s, java.lang.String[] params) Returns the transition probabilities for applying this action in the given state with the given set of parameters.
protected State	performActionHelper(State s, java.lang.String[] params) This method determines what happens when an action is applied in the given state with the given parameters.

Methods inherited from class burlap.oomdp.singleagent.Action

addActionObserver, applicableInState, clearAllActionsObservers, deterministicTransition, equals, getAllApplicableGroundedActions, getAllApplicableGroundedActionsFromActionList, getDomain, getName, getParameterClasses, getParameterOrderGroups, getTransitions, hashCode, init, isPrimitive, parametersAreObjects, performAction, performAction

Methods inherited from class java.lang.Object

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

Field Detail

sourceAction

protected Action sourceAction

The source action this action models

model

protected Model model

The model of the transition dynamics that specify the outcomes of this action

useRMax

protected boolean useRMax

Whether this action follows the RMax paradigm of transition to a fictious RMax state when the model is not known

RMaxState

protected State RMaxState

The fictitious RMax state this action will transition to for unknown transitions. Only used in RMax paradigms

Constructor Detail

ModeledDomainGenerator.ModeledAction

public ModeledDomainGenerator.ModeledAction(Domain modelDomain,
                                    Action sourceAction,
                                    Model model,
                                    boolean useRMax)

Initializes.

Parameters:

modelDomain - the model of the domain with which this action is associated

sourceAction - the source action this action models

model - the model specifying transition dynamics

useRMax - whether the RMax paradigm of transfering to fictious states for unkown transitions is followed.

Method Detail

applicableInState

public boolean applicableInState(State s,
                        java.lang.String[] params)

Description copied from class: Action

Returns true if this action can be applied in this specified state with the specified parameters. Default behavior is that an action can be applied in any state, but this will need be overridden if that is not the case.

Overrides:

applicableInState in class Action

Parameters:

s - the state to perform the action on

params - a String array specifying the action object parameters

Returns:

whether the action can be performed on the given state

performActionHelper

protected State performActionHelper(State s,
                        java.lang.String[] params)

Description copied from class: Action

This method determines what happens when an action is applied in the given state with the given parameters. The State object s may be directly modified in this method since the parent method first copies the input state to pass to this helper method. The resulting state (which may be s) should then be returned.

Specified by:

performActionHelper in class Action

Parameters:

s - the state to perform the action on

params - a String array specifying the action object parameters

Returns:

the resulting State from performing this action

getTransitions

public java.util.List<TransitionProbability> getTransitions(State s,
                                                   java.lang.String[] params)

Description copied from class: Action

Returns the transition probabilities for applying this action in the given state with the given set of parameters. Transition probabilities are specified as list of TransitionProbability objects. The list is only required to contain transitions with non-zero probability. Since not all planning algorithms require the full transition dynamics (and since it's impossible to enumerate them in some infinite state space domains), this method is not requried to be implemented. However, it will throw an UnsupportedOperationException if it is not overriden by the Action subclass if it is called by an algorithm that requires it.

Overrides:

getTransitions in class Action

Parameters:

s - the state from which the transition probabilities when applying this action will be returned.

params - a String array specifying the action object parameters

Returns:

a List of transition probabilities for applying this action in the given state with the given set of parameters