dk.deepthought.sidious.planner
Class GreenhouseHeuristic

java.lang.Object
  dk.deepthought.sidious.planner.GreenhouseHeuristic

All Implemented Interfaces:

Heuristic

public final class GreenhouseHeuristic
extends java.lang.Object
implements Heuristic

The class encapsulates the heuristic function of the A*-algorithm.

The heuristic is calculated by evaluating all rules associated with the requester of the plan. In this way the estimated cost of getting to the goal is very close to the actual distance.

Author:

Deepthought

Field Summary

private double	baseEuclidianDistance The base euclidian distance.
private ClimaticState	goal The goal of the search.
private SuperLinkID	id The id of the requester.
private static org.apache.commons.logging.Log	logger Logger for this class
private ClimaticState	source The source of the search.

Constructor Summary

GreenhouseHeuristic(SuperLinkID id, State source, State goal)
Constructs a new GreenhouseHeuristic with the specified id.

Method Summary

(package private) double	euclidianDistanceToGoal(ClimaticState state) Calculates the euclidian distance to the goal state.
double	h(State state) This heuristic is based upon a normalized Euclidian distance.
private double	normalizedEuclidian(ClimaticState state) Calculates the normalized euclidian cost of getting form any state to the goal state.

Methods inherited from class java.lang.Object

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

Field Detail

logger

private static final org.apache.commons.logging.Log logger

Logger for this class

id

private final SuperLinkID id

The id of the requester.

goal

private ClimaticState goal

The goal of the search.

source

private ClimaticState source

The source of the search.

baseEuclidianDistance

private double baseEuclidianDistance

The base euclidian distance. distance(source, goal).

Constructor Detail

GreenhouseHeuristic

public GreenhouseHeuristic(SuperLinkID id,
                           State source,
                           State goal)

Constructs a new GreenhouseHeuristic with the specified id.

Parameters:

id - the id to the requester.

source - the source state of any path in the graph

goal - the goal state of any path in the graph

Method Detail

h

public double h(State state)

This heuristic is based upon a normalized Euclidian distance.

This distance is the distance from the current vertex to the goal divided by the distance from source to goal.

Specified by:

h in interface Heuristic

Parameters:

state - the input state for the calculation

Returns:

an approximate cost

See Also:

Heuristic.h(dk.deepthought.sidious.supportsystem.State)

normalizedEuclidian

private double normalizedEuclidian(ClimaticState state)

Calculates the normalized euclidian cost of getting form any state to the goal state.

 distance(any, goal) / distance(source, goal)
 

Parameters:

state - the state to calculate the distance from

Returns:

the normalized euclidian cost

euclidianDistanceToGoal

double euclidianDistanceToGoal(ClimaticState state)

Calculates the euclidian distance to the goal state.

Parameters:

state - the source state

Returns:

the euclidian distance