burlap.mdp.singleagent.pomdp

Class SimulatedPOEnvironment

java.lang.Object

burlap.mdp.singleagent.environment.SimulatedEnvironment

burlap.mdp.singleagent.pomdp.SimulatedPOEnvironment

All Implemented Interfaces:

Environment, EnvironmentServerInterface, StateSettableEnvironment

public class SimulatedPOEnvironment
extends SimulatedEnvironment

An Environment specifically for simulating interaction with a POMDP environments (PODomain). In this case, the currentObservation() returns the last observation made from the Environment, not the hidden state, and the executeAction(Action) method does not return EnvironmentOutcome objects that contain the full state of the environment, but an observation drawn from the POMDP ObservationFunction following the execution of the action. If you would like to access the true hidden state of the environment, use the getCurrentHiddenState() method.

Field Summary

Fields

Modifier and Type	Field and Description
protected State	curObservation The current observation from the POMDP environment
protected PODomain	poDomain

Fields inherited from class burlap.mdp.singleagent.environment.SimulatedEnvironment

allowActionFromTerminalStates, curState, lastReward, model, observers, stateGenerator, terminated

Constructor Summary

Constructors

Constructor and Description
SimulatedPOEnvironment(PODomain domain)
SimulatedPOEnvironment(PODomain domain, State initialHiddenState)
SimulatedPOEnvironment(PODomain domain, StateGenerator hiddenStateGenerator)

Method Summary

Modifier and Type	Method and Description
State	currentObservation() Returns the current observation of the environment as a State.
EnvironmentOutcome	executeAction(Action a) Executes the specified action in this environment
State	getCurrentHiddenState() Returns the current hidden state of this Environment.
void	resetEnvironment() Resets this environment to some initial state, if the functionality exists.
void	setCurObservationTo(State observation) Overrides the current observation of this environment to the specified value

Methods inherited from class burlap.mdp.singleagent.environment.SimulatedEnvironment

addObservers, clearAllObservers, getStateGenerator, isInTerminalState, lastReward, observers, removeObservers, setAllowActionFromTerminalStates, setCurStateTo, setStateGenerator

Methods inherited from class java.lang.Object

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

Field Detail

curObservation

protected State curObservation

The current observation from the POMDP environment

poDomain

protected PODomain poDomain

Constructor Detail

SimulatedPOEnvironment

public SimulatedPOEnvironment(PODomain domain)

SimulatedPOEnvironment

public SimulatedPOEnvironment(PODomain domain,
                              State initialHiddenState)

SimulatedPOEnvironment

public SimulatedPOEnvironment(PODomain domain,
                              StateGenerator hiddenStateGenerator)

Method Detail

setCurObservationTo

public void setCurObservationTo(State observation)

Overrides the current observation of this environment to the specified value

Parameters:

observation - the current observation of this environment to the specified value

currentObservation

public State currentObservation()

Description copied from interface: Environment

Returns the current observation of the environment as a State.

Specified by:

currentObservation in interface Environment

Overrides:

currentObservation in class SimulatedEnvironment

Returns:

the current observation of the environment as a State.

getCurrentHiddenState

public State getCurrentHiddenState()

Returns the current hidden state of this Environment.

Returns:

a State representing the current hidden state of the environment.

executeAction

public EnvironmentOutcome executeAction(Action a)

Description copied from interface: Environment

Executes the specified action in this environment

Specified by:

executeAction in interface Environment

Overrides:

executeAction in class SimulatedEnvironment

Parameters:

a - the Action that is to be performed in this environment.

Returns:

the resulting observation and reward transition from applying the given GroundedAction in this environment.

resetEnvironment

public void resetEnvironment()

Description copied from interface: Environment

Resets this environment to some initial state, if the functionality exists.

Specified by:

resetEnvironment in interface Environment

Overrides:

resetEnvironment in class SimulatedEnvironment