planner.avclPlan.search

Class BestFirstSearchPlanner

java.lang.Object

planner.aStarPlan.Planner

planner.aStarPlan.SelfPruningPlanner

planner.avclPlan.search.BestFirstSearchPlanner

All Implemented Interfaces:

Runnable

public class BestFirstSearchPlanner
extends SelfPruningPlanner

Class that will use the Best First search algorithm to compute a series of waypoints that will search a geographic area with a prescribed probability of detection

Version:

$Revision: 1.4 $

Author:

dtdavis

Contact: Don Brutzman (web.nps.navy.mil/~brutzman) brutzman@nps.navy.mil

Field Summary

Fields

Modifier and Type	Field and Description
protected static double	COS60
protected GeoArea	searchArea The geographic area that is being searched
protected static double	SIN60
static double	SIXTY_DEGREES
protected double[]	startPoint Reference to the search starting point (transit waypoints should end here)
static double	THIRTY_DEGREES
protected double	trackSpacing Space between tracks of the search

Fields inherited from class planner.aStarPlan.Planner

agenda, die, goal, plan, planReady

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	BestFirstSearchPlanner() Creates a new instance of BestFirstSearchPlanner
	BestFirstSearchPlanner(GeoArea searchArea, double S, double PD, double[] startPosit, boolean pointFocus) Creates a new instance of BestFirstSearchPlanner

Method Summary

Methods

Modifier and Type	Method and Description
protected List<double[]>	computeVisitPoints() Computes a set of points that blanket the search area.
double[]	getStartPoint() Returns a reference to the search's starting position
void	reset(BestFirstSearchWorldState start) Resets the planner to start over with a new start state
void	reset(BestFirstSearchWorldState start, BestFirstSearchWorldState goal) Resets the planner to start over with a new start state and a new goal
void	setRequiredPD(double value) Sets the required single-search probability of detection of the search
void	setSweepWidth(double value) Sets the sweep width (defined in NWP 3-22.5 SAR TACAID) of the search
double	setTrackSpacing() Computes and sets the track spacing to achieve the desired 1-search probability of detection The method uses the formula PD = 1 / (1 + 0.25 * (1 / C)) where C is the coverage factor

Methods inherited from class planner.aStarPlan.SelfPruningPlanner

run

Methods inherited from class planner.aStarPlan.Planner

addToAgenda, clearAgenda, getGoal, getPlan, getRunTime, goalAchievable, goalAchieved, isPlanReady, reset, reset, reset, setDie, setRunTime

Methods inherited from class java.lang.Object

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

Field Detail

THIRTY_DEGREES

public static final double THIRTY_DEGREES

See Also:

Constant Field Values

SIXTY_DEGREES

public static final double SIXTY_DEGREES

See Also:

Constant Field Values

SIN60

protected static final double SIN60

COS60

protected static final double COS60

searchArea

protected GeoArea searchArea

The geographic area that is being searched

trackSpacing

protected double trackSpacing

Space between tracks of the search

startPoint

protected double[] startPoint

Reference to the search starting point (transit waypoints should end here)

Constructor Detail

BestFirstSearchPlanner

protected BestFirstSearchPlanner()

Creates a new instance of BestFirstSearchPlanner

BestFirstSearchPlanner

public BestFirstSearchPlanner(GeoArea searchArea,
                      double S,
                      double PD,
                      double[] startPosit,
                      boolean pointFocus)

Creates a new instance of BestFirstSearchPlanner

Parameters:

searchArea - geographic area that is to be searched

S - sensor sweep width as defined in NWP 3-22.5 (SAR TACAID)

PD - required single-search probability of detection of the search

startPosit - double[2] defining the vehicle's position at the beginning of the search

pointFocus - set to true if the search is to focus on the centroid of the area

Method Detail

reset

public void reset(BestFirstSearchWorldState start)

Resets the planner to start over with a new start state

Parameters:

start - world state when planning commences (must completely define world)

reset

public void reset(BestFirstSearchWorldState start,
         BestFirstSearchWorldState goal)

Resets the planner to start over with a new start state and a new goal

Parameters:

start - world state when planning commences (must completely define world)

goal - desired world state following plan execution (not necessarily complete world description)

getStartPoint

public double[] getStartPoint()

Returns a reference to the search's starting position

Returns:

value of startPoint

setTrackSpacing

public double setTrackSpacing()

Computes and sets the track spacing to achieve the desired 1-search probability of detection The method uses the formula PD = 1 / (1 + 0.25 * (1 / C)) where C is the coverage factor

Returns:

value of computed track spacing

setSweepWidth

public final void setSweepWidth(double value)

Sets the sweep width (defined in NWP 3-22.5 SAR TACAID) of the search

Parameters:

value - new sweep width value (meters)

setRequiredPD

public final void setRequiredPD(double value)

Sets the required single-search probability of detection of the search

Parameters:

value - new required probability of detection

computeVisitPoints

protected final List<double[]> computeVisitPoints()

Computes a set of points that blanket the search area. Ultimately, the search will consist of visiting each of these points in the most efficient sequence.

Returns:

ArrayList of double[2] points that define a set of (x, y) coordinates that blanket the area to be searched with sufficient granularity to achieve the required PD