LunarLanderRF

java.lang.Object
- burlap.domain.singleagent.lunarlander.LunarLanderRF

All Implemented Interfaces:

RewardFunction
```
public class LunarLanderRF
extends java.lang.Object
implements RewardFunction
```
A reward function for the LunarLanderDomain. By default, this reward function returns -1 except when the agent hits the ground, hits an obstacle, hits the side of the landing pad, or lands on the landing pad. When the agent hits the ground, obstacle, or side of the landing pad it receives -100. When it lands on the landing pad, it receives +1000. The rewards for these conditions may also be modified.

Author:

James MacGlashan.

Field Summary

Fields
Modifier and Type	Field and Description
`double`	`collisionReward` The reward for hitting the ground or an obstacle
`double`	`defaultReward` The default reward received for moving through the air
`double`	`goalReward` The reward for landing on the landing pad
`protected PropositionalFunction`	`onGround`
`protected PropositionalFunction`	`onPad`
`protected PropositionalFunction`	`touchingPad`
`protected PropositionalFunction`	`touchingSurface`

Constructor Summary

Constructors
Constructor and Description
`LunarLanderRF(OODomain domain)` Initializes with default reward values (move through air = -1; collision = -100; land on pad = +1000)
`LunarLanderRF(OODomain domain, double goalReward, double collisionReward, double defaultReward)` Initializes with custom reward condition values.

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`double`	`getCollisionReward()`
`double`	`getDefaultReward()`
`double`	`getGoalReward()`
`double`	`reward(State s, Action a, State sprime)` Returns the reward received when action a is executed in state s and the agent transitions to state sprime.
`void`	`setCollisionReward(double collisionReward)`
`void`	`setDefaultReward(double defaultReward)`
`void`	`setGoalReward(double goalReward)`

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - goalReward
```
public double goalReward
```
    The reward for landing on the landing pad
  - collisionReward
```
public double collisionReward
```
    The reward for hitting the ground or an obstacle
  - defaultReward
```
public double defaultReward
```
    The default reward received for moving through the air
  - onGround
```
protected PropositionalFunction onGround
```
  - touchingSurface
```
protected PropositionalFunction touchingSurface
```
  - touchingPad
```
protected PropositionalFunction touchingPad
```
  - onPad
```
protected PropositionalFunction onPad
```
- Constructor Detail
  - LunarLanderRF
```
public LunarLanderRF(OODomain domain)
```
    Initializes with default reward values (move through air = -1; collision = -100; land on pad = +1000)
    
    Parameters:
    
    domain - a LunarLanderDomain generated Domain.
  - LunarLanderRF
```
public LunarLanderRF(OODomain domain,
                     double goalReward,
                     double collisionReward,
                     double defaultReward)
```
    Initializes with custom reward condition values.
    
    Parameters:
    
    domain - a LunarLanderDomain generated Domain.
    
    goalReward - the reward for landing on the landing pad.
    
    collisionReward - the reward for a collision.
    
    defaultReward - the default reward for all other states (i.e., moving through the air)
- Method Detail
  - reward
```
public double reward(State s,
                     Action 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.
  - getGoalReward
```
public double getGoalReward()
```
  - setGoalReward
```
public void setGoalReward(double goalReward)
```
  - getCollisionReward
```
public double getCollisionReward()
```
  - setCollisionReward
```
public void setCollisionReward(double collisionReward)
```
  - getDefaultReward
```
public double getDefaultReward()
```
  - setDefaultReward
```
public void setDefaultReward(double defaultReward)
```

Class LunarLanderRF

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

goalReward

collisionReward

defaultReward

onGround

touchingSurface

touchingPad

onPad

Constructor Detail

LunarLanderRF

LunarLanderRF

Method Detail

reward

getGoalReward

setGoalReward

getCollisionReward

setCollisionReward

getDefaultReward

setDefaultReward