burlap.oomdp.singleagent.pomdp

Class BeliefMDPGenerator.BeliefAction

java.lang.Object

burlap.oomdp.singleagent.Action

burlap.oomdp.singleagent.pomdp.BeliefMDPGenerator.BeliefAction

All Implemented Interfaces:

FullActionModel

Enclosing class:

BeliefMDPGenerator

public class BeliefMDPGenerator.BeliefAction
extends Action
implements FullActionModel

A Belief MDP action. (transitions between belief states). This requires that the input states on which the action operates be an instance of BeliefState and EnumerableBeliefState. The getTransitions(burlap.oomdp.core.states.State, burlap.oomdp.singleagent.GroundedAction) method (used by planning algorithms that require the full transition dynamics) operates by iterating over all possible observations.

Nested Class Summary

Nested classes/interfaces inherited from interface burlap.oomdp.singleagent.FullActionModel

FullActionModel.FullActionModelHelper

Field Summary

Fields

Modifier and Type	Field and Description
protected PODomain	poDomain The source POMDP domain
protected Action	pomdpAction The source POMDP action this action will turn into a Belief MDP action.

Fields inherited from class burlap.oomdp.singleagent.Action

actionObservers, domain, name

Constructor Summary

Constructors

Constructor and Description
BeliefMDPGenerator.BeliefAction(PODomain poDomain, Action pomdpAction, SADomain domain) Initializes

Method Summary

Methods

Modifier and Type	Method and Description
boolean	applicableInState(State s, GroundedAction ga) Returns true if this action can be applied in this specified state with the parameters specified by the provided GroundedAction Default behavior is that an action can be applied in any state, but this will need be overridden if that is not the case.
protected java.util.List<TransitionProbability>	collapseTransitionProbabilityDuplicates(java.util.List<TransitionProbability> tps) Finds transitions that go to the same state and collapses them into a single TransitionProbability object with the sum of their probabilities.
java.util.List<GroundedAction>	getAllApplicableGroundedActions(State s) Returns all possible groundings of this action that can be applied in the provided State.
GroundedAction	getAssociatedGroundedAction() Returns a GroundedAction instance that points to this Action, but does not have any parameters--if any--set.
java.util.List<TransitionProbability>	getTransitions(State s, GroundedAction ga) Returns the transition probabilities for applying this action in the given state with the given set of parameters.
boolean	isParameterized() Returns true if this action is parameterized; false otherwise.
boolean	isPrimitive() Returns whether this action is a primitive action of the domain or not.
protected TransitionProbability	matchingStateTP(java.util.List<TransitionProbability> tps, State s) Finds a transition in the input list of transitions that matches the input state and returns it.
protected State	performActionHelper(State s, GroundedAction ga) This method determines what happens when an action is applied in the given state with the given parameters.
protected double	probObservation(BeliefState bs, State observation, GroundedAction ga) Computes and returns the probability of observing an observation in a given BeleifState when a specific action is taken.

Methods inherited from class burlap.oomdp.singleagent.Action

addActionObserver, clearAllActionsObservers, deterministicTransition, equals, getAllApplicableGroundedActionsFromActionList, getDomain, getGroundedAction, getName, hashCode, performAction, performInEnvironment

Methods inherited from class java.lang.Object

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

Field Detail

pomdpAction

protected Action pomdpAction

The source POMDP action this action will turn into a Belief MDP action.

poDomain

protected PODomain poDomain

The source POMDP domain

Constructor Detail

BeliefMDPGenerator.BeliefAction

public BeliefMDPGenerator.BeliefAction(PODomain poDomain,
                               Action pomdpAction,
                               SADomain domain)

Initializes

Parameters:

poDomain - the POMDP PODomain

pomdpAction - the POMDP Action that this BeliefMDPGenerator.BeliefAction will wrap.

domain - the Belief MDP Domain to which this BeliefMDPGenerator.BeliefAction will be associated..

Method Detail

applicableInState

public boolean applicableInState(State s,
                        GroundedAction ga)

Description copied from class: Action

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

Specified by:

applicableInState in class Action

Parameters:

s - the state to perform the action on

ga - the GroundedAction specifying the parameters to use

Returns:

whether this action can be performed on the given state with the given parameters

isParameterized

public boolean isParameterized()

Description copied from class: Action

Returns true if this action is parameterized; false otherwise.

Specified by:

isParameterized in class Action

Returns:

true if this Action is parameterized; false if it is not.

getAssociatedGroundedAction

public GroundedAction getAssociatedGroundedAction()

Description copied from class: Action

Returns a GroundedAction instance that points to this Action, but does not have any parameters--if any--set.

Specified by:

getAssociatedGroundedAction in class Action

Returns:

a GroundedAction instance.

getAllApplicableGroundedActions

public java.util.List<GroundedAction> getAllApplicableGroundedActions(State s)

Description copied from class: Action

Returns all possible groundings of this action that can be applied in the provided State. To check if a grounded action is applicable in the state, the Action.applicableInState(State, burlap.oomdp.singleagent.GroundedAction) method is checked. The default behavior of this method is to treat the parameters as possible object bindings, finding all bindings that satisfy the object class typing specified and then checking them against the Action.applicableInState(State, burlap.oomdp.singleagent.GroundedAction) method. However, this class can also be overridden to provide custom grounding behavior or non-object based parametrization.

Specified by:

getAllApplicableGroundedActions in class Action

Parameters:

s - the State in which all applicable grounded actions of this Action object should be returned.

Returns:

a list of all applicable GroundedActions of this Action object in in the given State

performActionHelper

protected State performActionHelper(State s,
                        GroundedAction ga)

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 (Action.performAction(burlap.oomdp.core.states.State, GroundedAction) 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

ga - the GroundedAction specifying the parameters to use

Returns:

the resulting State from performing this action

isPrimitive

public boolean isPrimitive()

Description copied from class: Action

Returns whether this action is a primitive action of the domain or not. A primitive action is defined to be an action that always takes one time step.

Specified by:

isPrimitive in class Action

Returns:

true if the action is primitive; false otherwise.

getTransitions

public java.util.List<TransitionProbability> getTransitions(State s,
                                                   GroundedAction ga)

Description copied from interface: FullActionModel

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.

Specified by:

getTransitions in interface FullActionModel

Parameters:

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

ga - the GroundedAction specifying the parameters to use

Returns:

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

probObservation

protected double probObservation(BeliefState bs,
                     State observation,
                     GroundedAction ga)

Computes and returns the probability of observing an observation in a given BeleifState when a specific action is taken.

Parameters:

bs - the previous belief state

observation - the observation that will be observed

ga - the pomdp action that would be selected in the previous belief state

Returns:

the probability of observing the given observation

collapseTransitionProbabilityDuplicates

protected java.util.List<TransitionProbability> collapseTransitionProbabilityDuplicates(java.util.List<TransitionProbability> tps)

Finds transitions that go to the same state and collapses them into a single TransitionProbability object with the sum of their probabilities.

Parameters:

tps - the List of transitions specified by TransitionProbability objects.

Returns:

the collapsed list of TransitionProbability with any duplicate transitions aggregated into a single TransitionProbability object.

matchingStateTP

protected TransitionProbability matchingStateTP(java.util.List<TransitionProbability> tps,
                                    State s)

Finds a transition in the input list of transitions that matches the input state and returns it. If no match is found then null is returned.

Parameters:

tps - The input List of transitions to search.

s - the query state for which a matching transition is to be found.

Returns:

the TransitionProbability in tps that matches state s or null if one does not exist.