7ýAMRL-TR-76-55

SUMMARY REPORT OF AMRL REMOTELYPILOTED VEHICLE (RPV) SYSTEM SIMULATIONSTUDY IV RESULTS

AEROSPACE MEDICAL RESEARCH LABORATORY

JUNE 1976

Approved for public release; distribution unlimited

AEROSPACE MEDICAL RESEARCH LABORATORYAEROSPACE MEDICAL DIVISIONAIR FORCE SYSTEMS COMMAND

WRIGHT-PATTERSON AIR FORCE BASE, OHIO 45433

BEST AVAILABLE COPY

NOTICES

When US Government drawings, specifications, or other data are used for any purpose other than a definitely relatedGovernment procurement operation, the Government thereby incurs no responsibility nor any obligation what-soever, and the fact that the Government may have formulated, furnished, or in any way supplied the said drawings,specifications, or other data, is not to be regarded by implication or otherwise, as in any manner licensing the holderor any other person or corporation, or conveying any rights or permission to manufacture, use, or sell any patentedinvention that may in any way be related thereto.

Please do not request copies of this report from Aerospace Medical Research Laboratory. Additional copies may bepurchased from:

National Technical Information Service5285 Port Royal RoadSpringfield, Virginia 22161

Federal Government agencies and their contractors registered with Defense Documentation Center should directrequests for copies of this report to:

Defense Documentation CenterCameron StationAlexandria, Virginia 22314

TECHNICAL REVIEW AND APPROVAL

AMRL-TR-76-55

This report has been reviewed by the Information Office (01) and is releasable to the National Technical Information

Service (NTIS). At NTIS, it will be available to the general public, including foreign nations.

This technical report has been reviewed and is approved for publication.

FOR THE COMMANDER

CHARLES BATES, JR.ChiefHuman Engineering DivisionAerospace Medical Research Laboratory

AIR FORCE - 13 AUGUST 76 - 100

SECURITY CLASSPFICATION OF THIS OAGE (W-hen Data ýEntered)REPORT DOCUMENTATION PAGE READ INSTRUCTIONSREPORT__DOCUMENTATIONPAGEBEFORE COMPLETING FORM

1. REPORT NUMDER 2. GOVT ACCE'SSION NO. 3. RECIP'ENT'" CATALOG NUMBER

AMRL-TR-76-554. TITLE (and Subtitle) 5, TYFE O-F REPORT & PERIOD COVERED

SUMMARY REPORT OF AMRL REMOTELY PILOTED VEHICLE Summary report(RPV) SYSTEM SIMULATION STUDY IV RESULTS 6. PERFORMING ORG. REPORT NUMBER

7. AUTHOR(@) 8. CONI RACT OR GRANT NUMBER(s)

Nilss M. AumeRobert G. MillsAldo A. Gillio

S. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGR7AM ELEMENT, PROJECT, TASKAREA A WORK UNIT NUMBERS

Aerospace Medical Research Laboratory, AerospaceMedical Division, Air Force Systems Command, 62202F 7184-14-AA

-Wright-Patterson Air Force Base, Ohio 45433

I. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE

Systems Research Branch June 1976Human Engineering Division 13. NUMBER oF PAGES

4214. MONITORING AGENCY NAME & ADDRESS(If different from Controlling Office) 15. SECURITY CLASS. (of this report)

UNCLASSIFIED15a DECLASSIFICATION/DOWNGRADING

SCHEDULE

16. DISTRIBUTION STATEMENT (of this Report)

Approved for public release; distribution unlimited

17. DISTRIBUTION STATEMENT (of the abstract entered in Block 20, If different from Report)

IS. SUPPLEMENTARY NOTES

19. KEY WORDS (Continue on reverse aide if necessary and identify by block number)

Aeronautics Operations, Strategy, TacticsMan-Machine Relations Navigation and GuidanceComputers Combat VehiclesSubsystems

20. ABSTRACT (Continue nn reverse aide If necessary and Identify by block number)

The AMRL RPV System Simulation and Research Program was initiated inresponse to requirements for support of the design of the man-machine/environment interface of AF RPV systems. The major objectives of the AMRLRPV System Simulation and Research Program are as follows: (1) Perform RPVsystem design evaluation studies, i.e., evaluate alternative designconfigurations, assumptions, operating procedures, etc.; (2) Assess RPVsystem effectiveness, i.e. evaluate the expected effectiveness of a given

DD I JAN73 1473 EDITION OF I NOV 65 IS OBSOLETE

SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)

SECURITY CLASSIFICATION OF THIS PAGE(ehan fDats Entered)

system configuration such as its overall probability of achieving a target,etc.; (3) Provide man-machine/environment interface engineering data, i.e.recommend displays etc.; (4) Test bed new technology, e.g. evaluateeffectiveness of contractor designed consoles, video bandwidth compressiontechniques, etc. The results of the fourth simulation study are reportedherein. This study included automatic RPV heading correction and positionreport smoothing functions in the simulation. The study employed scenariosrequiring a limited number of support mission RPVs, via reprogramming,to-provide 'coverage for a set of Strike RPVs (or manned vehicles).The study evaluates RPV system performance under the simultaneous effectsOf RPV Automatic Heading Correction Cross Track Threshold, Total Number ofRPVs Under System Control, Ratio of Strike Set to Support Vehicle Set,and Display (Window) Sizes or Scales.

SECURITY CLASSIFICATION OF THIS PAGEOWbhen Data Entered)

PREFACE

The authors acknowledge the assistance of the following individuals:

Mr. Gene Sebasky and Mr. Dale Wartluft, Federal Systems Division, IBM;

Mr. Robert F. Bachert, Sgt Jon VanDonkelaar, AMRL; Mr. Jeff Baughman,

Miss Jody Kusy, Miss Suzanne Gross, and Mr. Mike Berger, University of

'Dayton. Each of these individuals has contributed substantially to

the success of the RPV System Simulation program being conducted in

this laboratory.

CONTENTS

1.0 INTRODUCTION 1.......................................

2.0 AMRL RPV SYSTEM SIMULATION TEST BED ................. 2

3.0 RPV STUDY IV MISSION SCENARIO PARAMETERS ............. 4

4.0 CONTROL/DISPLAY PARAMETERS ......................... 8

5.0 OPERATOR RESPONSIBILITIES AND PROCEDURES ........... 10

6.0 RPV STUDY IV -- SPECIFIC OBJECTIVES AND SYSTEM

PARAMETERS INVESTIGATED ............... ............ 13

7.0 OPERATOR TEAMS ................... ................ 15

8.0 RPV STUDY IV RESULTS ........................... 16

9.0 MAJOR CONCLUSIONS .................... 22

1Q.0 REV SYSTEM SIMULATION STUDY V - PLANS ............. 22

TABLES AND FIGURES ................................... 24

1.0 INTRODUCTION

1.1.0 The AMRL RPV System Simulation and Research Program was initiated

in April 1973 in response to requirements for support of the design of

the man-machine/environment interface of AF RPV systems. The major

objectives of the AMRL RPV System Simulation and Research Program are

listed here.

1.1.1 Perform RPV system design evaluation studies, i.e.evaluate alternative design configurations, assumptions,operating procedures, etc.

1.1.2 Assess RPV system effectiveness, i.e. evaluate theexpected effectiveness of a given system configuration suchas its overall probability of achieving a target, etc.

1.1.3 Provide man-machine/environment interface engineeringdata, i.e. recommend displays, etc.

1.1.4 Test bed new technology, i.e. evaluate effectivenessof contractor designed consoles, video bandwidth compressiontechniques, etc.

1.2.0 The major thrust of this program has been to bring the state-

of-the-art in system simulation and system research concepts to bear

on the problem of RPV system design and development. System simula-

tion and research concepts assert that an RPV system will be comprised

of a number of sequenced phases and events that are dynamically

interactive when multiple RPVs must be controlled in real-time by

multiple operators. If possible, then, exploratory development efforts

involving these interactive elements should provide much of the

required data for system design. Since large-scale, multiple RPV

systems in the AF are yet to be developed, the most practical approach

to performing RPV systems research and also satisfying the system

dynamics problem is via computer simulation of a "postulated" real-

world system.1

1.3.0 This technical report summarizes the data of the fourth RPV

systems simulation study (RPV Study IV) employing the AMRL RPV System

Simulation Test Bed.

2.0 AMRL RPV SYSTEM SIMULATION TEST BED.

2.1.0 RPV Study IV employed the AMRL RPV System Simulation Test Bed.

The simulation incorporates a large number of the parameters of a

postulated real-world RPV system. The simulation employs four Enroute!

Return Phase operators and one Terminal Phase Pilot operator (henceforth

referred to as pilot). Each Enroute/Return operator monitors and

operates a computer graphics terminal (IBM 2250) comprised of a graphic

display (cathode ray tube, CRT), alphanumeric keyboard, lightpen, a

programmable function keyboard, and a small panel for hand-over actions.

The pilot operates a pilot station comprised of basic flight instruments,

a joystick for controlling a camera over a terrain model, and a closed

circuit TV monitor. The simulation is executed in real-time and permits

simultaneous control over up to 35 simulated RPVs.

2.2.0 Figure 1 is a schematic layout of the simulation facility as it

is presently configured. Note that there is closed loop manual or

automatic control over the camera of the terrain model facility. Figure

2 is a photograph of a team of 5 operators performing a simulated RPV

mission in the simulation facility. Figure 3 provides a closer view of

one Enroute/Return operator's terminal and the pilot's station with

video from the terrain model.

2.3.0 The AMRL RPV system simulation can be briefly characterized in

terms of the major submodels that are dynamically interfaced in the

simulation. These submodels are listed below.

2

2.3.1 Simulated RPV heading, altitude, and velocity flightparameters with automatic or manual, digital updatecapability while in the automatic flight mode or manual,continuous update capability while in the continuous controlflight mode (Terminal Phase).

2.3.2 Three data links (Command, Position Reporting, andVideo) for each simulated RPV and with interference para-meters.

2.3.3 Simulated RPV fuel load and rate of usage as afunction of velocity and altitude.

2.3.4 Simulated RPV attrition probability parameters basedon altitude and on the extent of (lateral) cross track devia-tion from the programmed flight plan.

2.3.5 Simulated RPV subsystem reliability operating in real"operational" time in conjunction with a simulated RPVinventory..

2.3.6 Simulated RPV navigation system parameters (RPV StudyIV employed parameters for Inertial, Doppler, and Basic DeadReckoning systems).

2.4.0 Specific displays and controls available in the simulation,

and operating procedures are described in "Remotely Piloted Vehicle

(RPV) Simulation Program Instruction Manual". This manual is

intended for operator instruction and generally interested persons.

It is being continually expanded and updated, and therefore is in

draft form. A copy can be obtained from AMRL/HEB, Wright-Patterson

AFB, Ohio 45433.

2.5.0 Enroute/Return operators are provided with displays (updated

in real-time) showing, for each RPV, a flight plan, a track signature

and vector displayed according to reported position, expected times

of arrivals (ETAs) to waypoints, status of Command data link, command

velocity and altitude, fuel remaining, lateral distance of the RPV from

flight plan, various alarm conditions, etc. The CRT display in Figure 3

3

shows four flight plans, a listing of RPV tail numbers and associated

information down the right side of the display, a status information

block in the lower left corner, and two other message blocks in the

lower center and the lower right areas. Operators also have the

capability to window (zoom) around an RPV at two display scaling

levels. Each operator can make use of all control devices (i.e.

light pen, alphanumeric keyboard, programmable function keyboard,

and the hand-over switch panel).

3.0 RPV STUDY IV MISSION SCENARIO PARAMETERS

3.1.0 RPV Study II* employed a "generalized" mission scenario,

intended to establish base-line data, and represented a cross-

section of specific scenarios in that it contained system task

elements considered to be present in most real-world RPV missions.

RPV Study III** employed a slightly more specialized mission scenario

assuming that a limited set of support RPVs (Electronic Warfare and

Low Altitude Reconnaissance) is available for coverage of a large

number of strike RPVs or manned aircraft. RPV Study IV continued to

use this type of mission scenario.

* Mills, R. G.; Bachert, R. B. and Aume, N. M.; Summary Report ofAMRL Remotely Piloted Vehicle (RPV) System Simulation Study IIResults, AMRL-TR-75-13, Aerospace Medical Research Laboratory,Wright-Patterson AFB, Ohio, Feb. 1975, AD 006 142.

**Mills, R. G.; Aume, N. M. and Bachert, R. F.; Summary Report ofAMRL Remotely Piloted Vehicle (RPV) System Simulation Study IIIResults, AMRL-TR-75-126, Aerospace Medical Research Laboratory,Wright-Patterson AFB Ohio, December 1975, AD A020064.

4

3.2.0 Most of the major parameters of the RPV mission scenario

for RPV Study IV are the same as for RPV Study III. All are

listed below. It is indicate d where parameters differ from those

of the RFV Study III scenario.

3.2.1 RPV Launch and Recovery Phases are assumed tooccur "1outside't the simulation.

3.2.2 Each RPV has an in dependent programmed flightplan that is assumed to be stored in the DCF (DroneControl Facility) and RPV computers.

3.2.3 'Each flight plan is assumed to be optimal withrespect to terrain and defenses. Thus, the MissionPlanning subsystem is also assumed to be outside thesimulation.

3.2.4 A round-trip of approximately 400 NM per RPV issimulated, with the center of the target area located200 NM from the launch insertion and recovery coordinates.Strike RPVs held quite closely to their path lengths. Eand L RPVs, due to repeated passes over targets, travelledconsiderably farther than the programmed path length.

3.2.5 Each RPV has an initial command velocity of 400knots and a command altitude of 200 feet.

3.2.6 Each RPV is designated one of three mission types;Strike (Weapons Delivery), EW (Electronic Warfare), andLow Recce (Low Altitude Reconnaissance).

3.2.7 In RPV Study IV scenario RPVs are launched accordingto type. The group of EW EPVs are launched first.-on 15second intervals. These are follo~wed by the Low RecceRPVs also on 15 second intervals. The Strike group islaunched last on 120 second intervals. This interval was90 seconds in RPV Study III. The number of vehicles ineach group is parametric and is determined prior to eachsimulated mission (see 6.2.1).

3.2.8 EPVs are launched from one of three launch areas.

3.2.9 For the RPV Study IV scenario each Strike, EW, andLow Recce group must be time-phased (coordinated arrivals)to target. Time-phasing is such that a single EW must be

5

within a 7 NM radius of the target assigned to the ccor-dinated Strike and a single Low Recce follows the sameStrike vehicle of 2 minutes (simulating BDA). Thesecoordination requirments will be discussed later (see5.6.0).

3.2.10 Each Strike flight plan has a designated Hand-offcoordinate (H) for Hand-off to Terminal Phase (Weaponsdelivery). Each strike flight plan has a designated way-point S, prior to H, for cueing the start of Hand-offprocedures. In addition to the S and H waypoints eachStrike flight plan has designated Target (T, one of threetargets), Hand-back (B), and Recovery (R) coordinates.

3.2.11 In RPV Study IV scenarios EW and Low Recce flightplans are programmed through all three targets (labeled 1,2, and 3). No other waypoints are designated on theseflight plans (except Recovery coordinate).

3.2.12 Prior to hand-off, an RPV is given a command toclimb to an altitude of 3000 ft. (in the case of Strikethe small area on terrain model also required a change incommand velocity to 250 knots).

3.2.13 For Strike flight plans the distance from S to His 7.77 NM, from H to T the average distance is 1.5 NM.

3.2.14 Each RPV is given just enough fuel to completethe roundtrip mission. Strike RPVs are assigned a fuelload of 2000 pounds. EW and Low Recce RPVs are assigneda load of 4000 pounds.

3.2.15 Only Strike RPVs are handed-off to Pilot controlon a first-come first-served basis.

3.2.16 It is assumed that EW and Low Recce RPVs must behanded-off but that the recipient operator is outside thesimulation (e.g., EWO, photo interpreter, etc.).

3.2.17 All EW, Low Recce, and overload Strike RPVs (Strikesnot handed-off to the Terminal Pilot operator) are handed-off to other Enroute/Return operators in the system but therecipient is passive (i.e., the only action on the operator'spart is throwing the "accept" switches; the operator does notmonitor the RPV, etc.).

6

3.2.18 Video imagery from the terrain model and continuouscontrol are available to the Pilot only during the actualTerminal Phase. Terminal Phase control will be achievedonly if the Strike RPV is within an approximate + 1500 footwide corridor to target.

3.2.19 The simulation employed for RPV Study IV included afunction to smooth raw RPV position report data. Thesmoothing function essentially fits a statistical, best-fitflight path to position reports. (See the Instruction Manualnoted in 2.4.0 for more detail.)

3.2.20 The simulation employed, for RPV Study IV included afunction to perform automatic RPV heading correction basedon smoothed position report data. The automatic correctionis ordered for an RPV when the lateral deviation or crosstrack error is in excess of a given threshold value. Thelateral deviation of an RPV is measured relative (perpen-dicular distance) to its stored flight plan. (See theInstruction Manual noted in 2.4.0 for more detail).

3.2.21 Both of the above functions (3.2.19, 3.2.20) areassumed to occur at the DCF level and not onboard the RPV.This "smarter" RPV system was initiated in RPV Study IIIand was continued in RPV Study IV. The impact of thesetwo functions was reported in an RPV Study II supplement.*

3.3.0 Figure 4 is a basic profile of scenario requirements for RPV

Study IV. In this figure one group consisting of Strike, EW, and Low

Recce flight plans are shown. The circles along flight plans and at

waypoints indicate that there will be an expected RPV position error.

In other words one can expect that given any desired point on an RPV

flight plan, true RPV position will vary inside an ellipse (shown here

as circles).

* Mills, R. G.; Bachert, R. B. and Aume, N. M. Supplementary Reportof RPV System Simulation Study II: Evaluation of RPV Position ReportSmoothing and Automatic Heading Correction. AMRL-TR-75-87, AerospaceMedical Research Laboratory, Wright-Patterson AFB, Ohio, Sept 1975,AD A-017 334.

7

3.4.0 Figure 4 also indicates that the EW and Low Recce shown in

the figure are to rendezvous with the Strike RPV at Target TI.

Since there are less EW and Low Recce RPVs than Strikes, the EW

RPV shown in the figure has previously covered a Strike at Target

T 2. Also, the Low Recce has previously provided BDA for a Strike

at Target T3 . In order to provide the necessary coverage at T1

operators are required to reroute (reprogram) both support RPVs

to T such that the rendezvous with the incoming Strike will occur

according to time-phasing requirements.

4.0 CONTROL/DISPLAY PARAMETERS

4.1.0 Each Enroute/Return operator station consists of an IBM 2250

Graphics CRT Terminals (there are four such terminals). These

terminals are equipped with a 12 inch CRT, light pen, alphanumeric

keyboard, and a programmable function keyboard. A small panel of

switches and lights has been added to each terminal for operator

control during hand-offs.

4.2.0 The Pilot station is equipped with a joystick, basic flight

instruments, and a TV monitor for displaying the terrain during the

Strikes or the information on one of the CRT terminals during the

rest of the mission.

4.3.0 Each CRT has the capability to display flight plans, track

signatures, geographical reference, etc., for up to 10 RPVs

simultaneously. RPV status parameters such as velocity, fuel

remaining, RPV mission type, etc., can be displayed for an individual

RPV. Other displayed parameters are ETA to the next designated way-

point, flight mode for each RPV in the system, and elapsed mission

time.

8

4.4.0 Operators can "call" displays and can make changes to RPV

flight parameters using the various control devices on their terminals.

In the case of heading changes the operator can employ any one of the

three window sizes (a 10 x 10 NM to 30 x 30 NM, a 60 x 60 NM to 140 x

140 NM, (see 6.2.4 for complete details) which are centered on the RPV

to be changed, and a 220 NM x 220 NM window which is centered on the

entire geographical area). The operator then introduces a set of

points (not to exceed 10) on the CRT face using the light pen. This

is called a "patch", and it always starts from the current RPV

position and must end on a point on the original flight path. If the

operator's prescribed points call for an impossibly tight turn, the

computer rejects the patch, in which case the operator must introduce

a new patch. Valid patches are transmitted over the command data

link to the on board computer and the RPV proceeds to fly through the

points prescribed in the patch.

4.5.0 Each attempt to communicate an instruction to an RPV is assumed

to employ the command data link. The possible commands are altitude

changes, velocity changes, navigation system changes, destruct, deploy

chutes, and heading patch (change heading). After a command to an

RPV is entered, it remains in the "outstanding commands" block (during

this time it is displayed) until it has been transmitted.

4.6.0 The displayed position of each RPV is in the form of a track

signature consisting of a heading velocity vector and an ID number.

The displayed position is computed by adding, vectorially, the position

reporting system error, navigation system error, as compounded by

operator error, to the true RPV position. The simulation frame time

(display update) is five seconds.

9

5.0 OPERATOR RESPONSIBILITIES AND PROCEDURES

5.1.0 Enroute/Return operators are required to perform the following

general tasks.

5.1.1 Monitor and update RPV position based on minimizingoverall cross track (lateral deviation) error.

5.1.2 Coordinate all RPV arrivals to the target and recoveryareas.

5.1.3 Time-phase each Strike RPV such that it achieves its"original" ETA (assigned during flight plan generation) toeach designated waypoint (S, H, T, B, R). To achieve amaximum number of hand-overs to pilot, some deviations arepermissible (also see 5.6.1).

5.1.4 Time-phase RPV recoveries such that EW and LowRecce RPVs arrive at R in any order, Strike achievesoriginal ETA to R, and the arrival interval of all RPVsis as near to, but not less than, 15 seconds as possible.

5.1.5 Perform hand-offs to other operators when required.

5.1.6 Accept RPVs (on a passive basis) handed off by otheroperators.

5.1.7 Hand-back RPVs upon request from another operator.

5.1.8 Respond to RPV failures, e.g. by Destruct, DeployChutes, Replace Navigation System, etc.

5.1.9 reprogram (recycle) RPVs to replace RPVs that arelost due to malfunction, attrition, etc.

5.1.10 Manage RPV fuel.

5.2.0 ETA adjustment for an RPV is accomplished by the operator

altering RPV velocity and/or RPV heading (i.e., increasing or

decreasing RPV flight path distance). Reprogramming an RPV is

accomplished by causing the replacement or support RPV to go to the

desired target area after it has completed its assigned mission. A

replacement RPV must be of the same type as the RPV lost and must

be time-phased with the remaining RPVs of the group. Additionally,

10

a replacement strike RPV must be assigned to the same target as the

lost RPV.

5.3.0 The Pilot is required to perform the following general tasks.

5.3.1 Direct coordination of RPV hand-offs and arrivals totarget and recovery areas during "dead-time" of Enroute andReturn Phases of the mission.

5.3.2 Accept Strike RPVs from Enroute/Return operators.

5.3.3 Switch into Video and Continuous Control mode whena Strike RPV is successfully handed-off.

5.3.4 Perform target acquisition and simulate line-of-sighttarget lock-on or weapon release (activate Trigger switch).

5.3.5 Perform hand-back of a Strike RPV following thetarget phase.

5.4.0 Target detection and acquisition requirements were minimal

during the Terminal Phase. This is due to repeated runs to the same

targets and the small area of the terrain model. At the present time,

the lack of significant detection and acquisition problem is not

viewed as a serious deficiency. The simulation study is concerned

with the dynamic interaction of the major elements of mission phase

integration, multiple RPVs, near simultaneous hand-offs, multiple

operator interactions, etc.

5.5.0 A successful hand-off could occur (a Strike RPV achieves

continuous control) only if the RPV was within an approximate + 1500

foot wide corridor on both sides of the flight path to the target.

5.6.0 There were a number of performance requirements which the RPV

system was expected to achieve. These were prioritized for the

operators. Operators were instructed that in order to achieve the

criterion of highest priority, some accuracy might have to be

11

sacrificed on lower priority items. Furthermore, each team (see

5.7.0) was allowed to employ its own strategy for accomplishing the

criteria. The priorities are listed here.

5.6.1 Priority 1 -- Maximize the number of Strike RPVshanded-off successfully to the Pilot with the EW andLow Recce rendezvous criteria (see 3.2.9). This impliesoccasional adjustments to the prescribed ETA's,remembering to keep the squared error(difference betweenprescribed and actual ETA) to a minimum.

5.6.2 Priority 2 -- Maximize the number of Strike RPVshanded-off successfully to the Pilot but fail to meet theEW and Low Recce rendezvous criteria, (e.g., a Strike RPVproperly handed-off but not provided with BDA coveragebecause it was not possible or BDA coverage provided viaa Low Recce RPV 4 minutes after Strike instead of therequired 2 minutes).

5.6.3 Priority 3 -- Maximize the number of Strikesachieving Priority 1 or 2 and original ETA accuracy forStrikes.

5.6.4 Priority 4 -- Maximize the number of Strikesachieving above priorities and minimal position erroralong flight plan.

5.7.0 Operator teams were required to do their own planning and

scheduling of RPVs. The planning was done at the start of a

mission. Operators were provided with computer print-outs listing

original ETAs to all waypoints for each RPV flight plan. Operators

were also permitted to use electronic calculators or other devices

of their-own choosing.

5.8.0 The three targets for a given mission were chosen randomly

from a set of twenty-seven targets located on the terrain model.

12

The targets were in the form of small white disks easily identifiable

on the Pilot's video monitor. Target assignment was made at the

start of each mission. The Pilot was provided with maps and photo-

graphs of--the terrain model. It was the Pilot's task to locate the

three targets prior to performing the first strike and without the

benefit of the terrain model. (When not in the continuous-control/

video-on modes of a Terminal Phase for an RPV the Pilot has access

to closed circuit video from one of the Enroute/Return CRTs. The

display is on the same video monitor that provides the terrain video.

This display provided the Pilot with ongoing mission information, e.g.,

progress and ETA of a given RPV.)

6.0 RPV STUDY IV -- SPECIFIC OBJECTIVES AND SYSTEM PARAMETERS

INVESTIGATED

6.1.0 The objective of RPV Study IV was to evaluate RPV system

capabilities to perform the support tasks described above and to

investigate the simultaneous effects of a set of independent vari-

ables on RPV system performance. The RPV system referred to here

is strictly that system postulated by the RPV systems simulation.

6.2.0 The effects of five independent variables on RPV system

performance were investigated. These variables are listed here.

6.2.1 Number of RPVs under simultaneous control by thesystem (18, 21, 24, 27, or 30 RPVs). This parameter canbe converted to an RPV/operator Ratio assuming fourEnroute/Return operators. It can be used interchangeablywith the number of RPVs parameter'. However, RPV/OperatorRatio can be employed across systems of varying expectednumber of operators.

13

6.2.2 RPV Lateral Deviation Alarm Threshold Value (100,325, 550, 775, or 1000 ft.). This threshold determinesthe value upon which the automatic heading correctionfunction is triggered. It also triggers a counter (0-9)that displays the number of consecutive frames an RPV hasexceeded the threshold value thus, indicating a lateralerror trend. The automatic heading correction functionwould trigger on the fourth count. However, operatorscan override the function, disable it, and it is auto-matically disabled during an RPV turn and under certainother conditions (see the Instruction manual noted in2.4.0).

6.2.3 Ratio of Strikes to EW and Low Recce RPVs (1.0,2.5, 4.0, 5.5, or 7.0). This parameter is calculatedby computing the ratio of the total number of StrikeRPVs to the average of the total numbers of EW andLow Recce RPVs (i.e., (EW + Low Recce)/2). The averageis required to account for unequal numbers of EW andLow Recce RPVs. Thus, a ratio of 7.0 indicates there isonly i EW RPV and there is only 1 Low Recce RPV availableto provide coverage for 7 Strike vehicles as they arrivesequentially in the target area.

6.2.4 The callable window displays' scaling was avariable in this study. The full size, 220 x 220 NMdisplay remained unchanged, but the smaller scale windowscould take on different values. The smallest (Mod 1)could be 10, 15, 20, 25, or 30 NM square. The intermediate(Mod 2) could be 60, 80, 100, 120, or 140 NM square.During a mission, both were held at the preselected values;these values were also printed on the ETA sheets that wereavailable to the operators at the start of each mission.

6.2.5 RPV Position Reporting error was held constant at525 ft. range error and 0.6 miliradian azimuth error.

6.3.0 The five variables were varied in combination with each

other according to a Central Composite/Fractional Factorial

experimental design.* This type of design allows one to investigate

a large, multivariate, experimental space by taking a minimum number

of observations. An observation constitutes a single execution of

the RPV system simulation (i.e., a simulated mission).

*Experimental Designs, Cochran, W. G. and Cos, G. M., New York:Wiley, 1957. 14

6.4.0 In RPV Study IV 32 observations (5 of them were replications)

were obtained on each of four operator teams, yielding a total of

128 observations. Each observation required approximately 1 3/4 to

2 1/4 hours of execution time.

7.0 OPERATOR TEAMS

7.1.0 Operators were obtained from universities in the Dayton, Ohio

area. They were required to be undergraduates for program longevity

purposes and to have at least a "B" average. Howevei, because many

had been incoming freshmen at the start of the research program and

there were occasional immediate needs, the grade requirement was

sometimes relaxed. Teams were comprised of five operators. Four

primary teams were formed. A fifth secondary team served as a

back-up pool of trained operators. Data were collected on all teams.

7.2.0 Individual operators acquired a minimum of 5 months training.

Most operators had completed 6 months initial training, RPV Studies

I, II and III, or some combination thereof. No member of the prin-

cipal teams had less than the 5 months training. This training had

been acquired durinig the initial training in the program or as a

member of the back-up team. Pilots had been given additional training

(2 weeks) in instrument flight simulators, as well as controlling the

camera over the terrain model. In addition, each team executed a

number of practice missions. Data collection was started after a

unanimous vote of the primary teams that they were ready to start.

15

8.0 RPV STUDY IV RESULTS

8.1.0 In order to allow readers to gain the information most relevant

to their needs, interpretation of results will be kept to a minimum and

will be general in nature.

8.2.0 Approximately 100 dependent variable measures of RPV system

performance were taken in RPV Study IV. The mean and standard deviation

of each of these measures are presented in Table I. This table lists

each dependent variable in terms of the following categories.

8.2.1 RPV Position error Variables--relate the deviationbetween "True" RPV position known only to the simulationto "Ideal" RPV position (where it should have been) tovarious mission phases and major waypoints.

8.2.2 Enroute/Terminal/Return Transition Related Variables--are primarily concerned with phase transition and TerminalPhase performance.

8.2.3 Command Data Link Usage Variables--attempt to quantifysystem transmission rates, etc.

8.2.4 RPV Scheduling Variables--quantify system capabilitiesto achieve the various time-phasing requirements of RPVStudy II.

8.2.5 Miscellaneous Variables--are simply those that do notfit conveniently into a large enough category.

8.2.6 Coordination Variables Involving Timing and RangeCriteria--express capabilities to meet the support ofStrike vehicles with EW and Low Recce RPV coverage require-ments.

8.3.0 Table I presents means and standard deviations of data that have

been collapsed over all teams and missions which implies that some

missions yielded better and some worse results than those shown. By

collapsing over missions the assumption has been made that the variables

of Table I are system non-parametric. In other words the five experimental

independent variables (see 6.2.0) are assumed to have had no practical

16

effect on system performance. Under this assumption, it is possible

to get a summary perspective of system performance despite the large

number of measures and the complexity of the system problem. The

examination of Table I can provide insight into the objectives of the

missions as well as lead to additional questions. (Since EW and Low

Recce RPVs underwent considerable reprogramming during a mission,

position error data and scheduling data relative to the flight plans

for these RPVs are not relevant; thus, these data are not presented

in Table I.)

8.3.1 A review of the data in Table I shows that ingeneral, ETA management and RPV position control aresatisfactory. Cross track error under the variousmission phases showed a general decrease as comparedto RPV Study III. This can be explained by the factthat in the previous study, subjects relied entirelyon the automatic heading correction to keep the RPVson course. In RPV Study IV, they were instructed tointroduce manual heading corrections if and when thesituation demanded it (see variables 1, 3, 5, 7, 9,11, and 13, Table 1). The added manual headingcorrections are also reflected in the slight increasein heading commands per vehicle (variables 19, 20,and 23).

8.3.2 The major new variable investigated in RPVStudy IV was the effect of various window sizes.No major overall influence was expected, and nonewas observed. The effect of the various sizes ismainly on visual resolution along the outboundand return legs. The size can become a limitingfactor while rerouting EW and Low Recce RPVs aroundthe targets. Very small sizes may not have enoughspace to make a good heading change, and at verylarge sizes the displayed situation may become toocluttered and cramped.

8.3.3 A note of explanation on variables 25 and 26:variable 25 pertains to the time between commandsfor a single RPV, while 26 pertains to the entirefleet consisting of a large number of RPVs.

8.3.4 One might note that for EW and Low Recce RPVs,the recovery rate is of the order of 2 percent, while

17

about 16 percent are lost due to malfunctions andattrition. The remainder were lost because theyran out of fuel, which may have to be consideredan experimental artifact.

8.4.0 In order to assess the effects of the five experimental variables

a regression analysis of variance was performed on each variable listed

in Table 1. This analysis yielded second order regression equations,

confidence interval estimates, analysis of variance, and a multiple

regression coefficient (NRC) for each variable. The MRC is an indication

of the extent to which the experimental variables have an influence on

the measured variable. Nineteen variables with the highest MRCs were

selected and are presented in Table II according to the rank order of

their MRC. (The same analysis can be provided for any other variable in

Table I and any other measurable variable on request.) Thus, the first

variable is Number of Heading Commands Transmitted per Vehicle for'the

Enroute Phase (Strike only) with the highest MRC, R = 0.9711. It is note-

worthy that the first eleven variables are concerned with various facets

of heading commands. Table II lists the MRCs for the variables as well

as the regression equations.

These equations can be employed by engineers etc., to assist them

in the specification of system parameters or making up "expected" data

profiles given selected values of the equation parameters. However, low

correlation coefficient (e.g., less than R = 0.85) in combination with

an experimental hyperspace (i.e., five dimensions) may cause the

equations to produce unexpected or illogical results, especially if

used with values outside of the ranges that were employed in the

experimentation. Therefore, one should exercise caution and good

reason in performing analyses with these equations.

18

8.4.1 The analyses performed did not reveal the existence ofof a dominant system parameter in a practical sense. Althoughthere are a number of statistically significant relationships,with the exception of the Lateral Deviation Threshold parameter,none could be even remotely practically significant. The para-meter of Ratio of Strike RPVs to EW and Low Recce RPVs had onlya minimal effect on any of the performance variables, a findingthat is similar to the one arrived at in RPV Study III. Aftera review of all performance variables, it was concluded thatthis Ratio could account for less than 1.0% of the total vari-ability in most of the variables. The ratio accounted for alarger percentage (5 to 15%) of the variability in the numberof altitude and velocity change commands for EW and Low RecceRPVs, and in the coverage provided for Strikes by the supportRPVs. While the behavior of the above variables disallows thepinpointing of an optimum Ratio, all these variables distinctlyworsen in the range of 5.0 to 7.0. Some of the variables alsoworsen as the Ratio approaches 1.0, which suggests that theRatio of Strikes to EW and Low Recce RPVs should be somewherein the range of 2.0 to 4.0. It is possible that the absolutenumber of support RPVs is a significant parameter, in that withvery few support RPVs the loss of one can result in the loss ofcoverage, and with a large number of support RPVs the manage-ment of the fleet becomes confusing and difficult.

8.4.2 The only system parameter of any significance is theLateral Deviation Threshold. It is the parameter solely res-ponsible for the high correlations obtained for RPV Cross Trackand Command Data Link usage variables, In RPV Study III, theenroute/return operators relied entirely on the automaticheading correction to keep RPVs on course. In RPV Study IV,they were instructed to introduce manual corrections when thethreshold value was large and when an RPV was observed to bedeviating from its flight-path by some decision value deter-mined by each team. Such a situation then influences twogroups of variables: one relating to cross track errors andthe other relating to command data link usage. It should beobvious that the basic driving function is RPV navigationerror, or the tendency to drift off course. As long as thistendency is constant, attempts to keep RPVs on course will re-quire a large number of heading commands; and an attempt tokeep the number of heading commands low will result in an in-creased cross track error. A review of the analyses and theregression equations discloses that a large number of systemperformance variables (in the "number of heading commands"and in the "cross track error" categories) are determined toa great extent by the lateral deviation threshold. A review

19

of cross track error data revealed that the magnitude of crosstrack error is approximately one half of the threshold value.The implication here is that the operators did not uniformlyand consistently introduce manual heading corrections, but stillrelied largely on automatic heading corrections, although therewas some manual heading correction. If a real-world RPV systemis to have an automatic course correction capability, then itshould be so designed as to make manual corrections unnecessary.In the design process, one should consider the desired accuracyor the maximum allowable cross track error to determine theautomatic threshold value. However, the inherent error of theposition reporting system also has to be considered, and thethreshold should not be set so low that it is triggered fre-quently by the random errors produced by the position report-ing system. Manual heading changes, such as for recyclingsupport RPVs around a target, of course, are in a differentcategory; the system must have a manual heading change capabi-lity in addition to the automatic course correction.

8.4.3 While it is possible to make a general statement aboutthe magnitude of the cross track error as a function of auto-matic threshold, the number of heading commands cannot be gene-ralized as easily, as the latter depends not only on thethreshold value but also on the tendency of the RPVs to driftoff course and on mission duration.

8.4.4 The largest recorded influence of the combination ofdisplay window sizes was on the Ratio of Distance Travelledto Length of Flight Path (variable 15 in Table I and variable17 in Table II). According to the data, optimum performancecan be expected with the windows approximately 20 and 70 NMsquare. These numbers, though, may be specific to the scena-rios of RPV Study IV, and other scenarios may require differentsizes. Furthermore, it was the interaction term which wasstatistically significant, but it accounted for only 18.09%of the variability in performance, which makes it a weak in-fluence. It also must be pointed out that the Ratio of Dis-

-tance Travelled to Length of Flight Path is not a very goodvariable to base judgements on. As a general rule, it can bestated that a single window size is insufficient because itlacks flexibility. There is the possibility of allowingoperators to select the window size they want as they dealwith real time events - and this is completely within thestate-of-art of computer graphics. With such a feature, theoperator would type in a number which determines the windowsize until such time when the operator decides to change it.

20

Otherwise, window sizes can be decided upon by consideringhow large an area has to be available (e.g., for makinggood heading changes for recycyling RPVs around a target)and how fine a resolution is needed (e.g., to visuallydetect deviations from flight path). There still should bethe capability to change the window sizes if a radicallydifferent situation demands it.

8.4.5 Table III provides a list of confidence intervalestimates for the 19 variables based on 0.95 probability.In other words 0.95 expresses the fact that one is toexpect the variable to be within the tolerance valueslisted 95% of the time as long as the system parametervalues are within the parameter continuum quantified andinvestigated. The intervals presented in Table IV can bequite useful in that they provide estimates on the limitsof system performance. For example, for the first vari-able, Number of Heading Commands Transmitted per Vehiclefor the Enroute Phase only (Strike Only), one can expectthe number to be greater than 15.13 but less than 95% ofthe time. This is not to say that some specific missionswill not yield numbers outside these limits because thelimits are based on expectation. It is also noted that thelimits, while valid for present purposes, are approximate,This is due to the experimental design procedure selectedfor the study. However, the approximation is minor and thedetails of how it comes about need not be discussed here.

8.5.0 The lack of practically significant parametric effects had

three major implications. First, the data in Table I and of the type

in Table III are adequate for estimating expected system performance.

Second, the result is consistent with all previous RPV system simula-

tion studies performed under the present program. This observation

may be very important. It suggests that the RPV system as postulated

herein is compensatory, i.e. it can make real-time adjustments to

parametric and dynamic change such that specific effects of outside

influences, i.e., IVs "wash out." It also suggests that the "acceptable"

21

range of a parameter may be quite wide and only extreme or catastrophic

changes will in fact perturb the system. Third, it is not possible at

present to optimize the parameters in the strict mathematical sense.

The effects are too small and the trade-offs are too complex. The only

recourse then, is to apply equations to a specific parameter set.

Unfortunately, systems technology does not as yet, possess the tools

to deal with these latter two implications.

9.0 MAJOR CONCLUSIONS

9.1.0 The system as postulated by the AMRL RPV System Simulation Test

Bed is capable of effective, multiple control and time-phasing of

principal (Strike in this case) RPVs. The system is only partially

successful at coordinating support RPVs (EW and Low Recce in this

case) with principal RPVs. The capability of the system to perform

such tasks, given selected parametric values, can be estimated by

solving the regression equations presented herein or made available

upon request.

9.2.0 There are, in general, no practical effects of the system

independent parameters investigated. The implications of this conclusion

were discussed earlier (section 8.5.0).

9.3.0 All other conclusions are consistent with those arrived at in the

previous RPV Simulation Studies.

10.0 RPV SYSTEM SIMUIATION STUDY V - PLANS

10.1.0 The fidelity of the simulation will be increased by introducing

more realistic dynamic changes in RPV altitude and speed in flight plan

22

follow mode instead of the instantaneous changes used up to now.

To investigate the effects of a limited data processing capability,

various delays will be introduced in the system's reactions to

operator commands (such as speed and altitude changes, and heading

commands). These modifications have been made possible by an increase

in computer capability (from IBM 360/40 to IBM 370/155), allowing the

parametric study of data transmission rate requirements and other

refinements and additions to the simulation of future RPV systems.

23

4- U5 * * * *

;e ;IIII~:*i * f f

cc* f t * * * * t f t

oL a: * * f t * ft * : :* f

t> * t f t f t t * f01 1:W * t f t * f t f t * * fCOu l ft M t * * ** * * t f t f

i In t t * * *OLt Z 0 * ftfA N * * fCC0 4- *xt f t f t f f t f

z- 0 4-5 ft ** ft * * * ft f£ 30. 0 t f t * f t * * fWWW Was V)t f t * * * * t f t f

<05 NU 0 '0 ml w ' A '0 q Al N 0 -

L a-z >ý An N A A 0 44 co Nl 0- '0 -4 N tCc~ o)0 A ' 0 a '.4 N vt Ao 1- 0

Z0 '.4 V Nv -N N3 0ý 1+ a 0 - 0 'A0 0<O) 0l A 4 41 - 'A z0 4 54 N;

- OG 4- N 0 N A f 0 a- 00 0 A N -4 N Im In (3 N 0 04 '5 0 f N 0 '

4 i A1 > ;" & tA N4 0 4 N 0 -iz 010 N N m < Li N1 nc 0 4 N 5

4 u t4-iL LL"4Uu i U4- L5a c Ll x-0 1 ..

crow0 u 4 - - -= c XZLU 00.z < 0 cI-3 c ) i-

"-j 0 i.4-0 4x. In -4 4t >j! La0> .1 U a ,c : u>0D a.rt >5 >- " 0. .4 0j -u 0-t S

W" In WU...Z"'Z 0 x-j ý t-WW I.- : L""4w owll. 0. < 0 4- 0. Z0CS I- Z u ) 0- <= -

o0. - 1 A ý-c I. x - - 4- 3~f 0.4 4-.-0 0 0: D4-SU - )5.Xa.< -u U.4 4- 5 1&U.0 :r4-4- 04 4-)- 0r - z 41-

In CC 01 W. O1CWL0 OW x 06 t. OW -U1 ) W4 W) = fl. $0 J 0 3zý iL-N U 4044- UU)04 Xu I.- - 0> 0e m UL I . X4IS: 0 . -4 4- Ut) o

AX w4am I451'4 .- cc WUo0l.4 <s Z'-." 0 0. ItW >05 uL 0 0 0.-#-w- 1S i.4 >x 9 11 0.4- OWa D 4U 0i m 4-L 4-40000 0w & - 0-s I- '-44 #.- 0 0 SL I.- -< Of 4-U Z 4L- LL5 4 l

w CC -_j '-W-CU <1)~ 4- 14 3 LL --- Z - = W 4 " 4- 0 0- 400-cc4 LU 04-40:0(3-m - - Ke u: rz 't- of w U ce 0.4- c 0 . 04--< 0c 4 cc04. U) WO U) 1 03 C30 T5-a w 0 0.3c 503 4- 4-

o-400 > U) 0 5 = uu So. IL010> c .nal c - re3 X0 4-'-.4 CC 0.1 5 ý 54- LU) 0<> cc 0.4m- o W4>0It 4-> zU Il -50. ft. 4-> 50 i5 f0l-s ý- L- wa w 440a.-

4-4 W tA- 4- a--4 4 0.01 U. 44 4- 0 L WO 0 0- U. 0 50. -0 U)"L 0. -. 4 .- iit l U. 0 j- Q5 5l- j 4 U. 0 4w U. 0 . tW

;.a 4- C <) UZ -0.00 <-u 0. x4 - a.4- <-L 0. < <LJ 10. w i 0. o z -4>4 1-0 w) 0. 1-swcr acc OW> W c ) OW cz x ow > . 00.'-'0. 0w In 4 0 CC4 )a - 4-

e.5V) In 0 0.09L(z 0cc5--OW. " :.-3 00 0 w 0 I- .0. 0. 0tS- 5 0$.01 01-I^- 0 000 Wa: 0 0.4->-. 0.0 0> D.4- 0 < I0.-40- 0.- m- z.0 -L I L DW

4 0.0000 04 us -. 0. 0- 04 03 0 0003 00 04 00 00 0>

01.0 - "0 Q 040 0 04 - 0Q 4 C ) Q 0 0Q U) 4

41~I fn InJU0400 03 0.4 03-mE U 4>350 0 . 3. 10* 0 O~usus 0 U)4 05 W5 U W4 403000- '-4 L Li 4U

540 0.45-404-4U 0.0.-. 0.. 4 04 W40U)< . 41 .5 44

w * 1 *' % '

0I-. 41 * *4. >~l* 4 1 4

M :U :1 t3 41

41 41 41- -

In >:

44z > N1 41 4 1 N N -4 4 4 1 4C)W x1 a1 4 i -4 1 1 412W !Z 4 ý419 0 * S * 4 4141 L 41 -44 441 * 44'.. V I--4 1 *4

0 1 cc 41 41 41-~

en o1 N1. O -4 s 1 41W 0 n Z*4e1 ~ % 4 f 41 c 1 41ZU 04 * <1 41 0' 4 1 4wu1 a 0w *U *x1 ~ r ' 41 1 4c O- Wn 41 * -4~"L cL41 4

0.Ci.* 4

>uW j/ 41 4.0z 0 0'0 0 3 o w 0n Ll ý N '0>4Z 0 4' 3% N -4 L P. Lru 0j U I 0L ixot oU 0f N 0 1 4- tn <-4<L C0 M I- " 9U CO -q < " = X E "C i o (, LII0 < . 0 < 2 0 0. V) NL X UJ.4V 0(A ccu V) D .CC4 .J-V - ( L LZ 0 U<- uj U- WI- :0: z z z < z . -

Z U.0 U CL4 I.- Ni 0 0 CA 0 En 0 0 N (A >I. Zza c 0o- N r 1 0 N) ac LO 0- D " ý- I. t% zx Q < =)0 z L 0 e3 C32 a 0 ui xO oU "' '4l 0"0 (n e > S < .ý (A ZSzqt < zc 0 U a = c 0

W J -4Q We 0 4 3% N 0 Q% 0 4 M 4 L '0. LuC-J ti) Ei z d *e CL N u-) Q e c c QU'2 lZo _')1 -i c i )-u mLU Q ) I -4U -4l-= T0.10 1 I- > 0 >0W -N z ý :c Uo cz - L Z 0 - 0 : 3 zU : . <U iu

z LU 0 z = oI j- 3a 3L0fl z - Lu NWe N(/ ac' Nx/ 0 > i> 0:,- 0> C -- L -ui> -Dl )I .4: <tf I4 - 4NW -)0 xUi uzXc

M~U .0 0 0 / 4 Co CL a)0. ::o a. CO)0 CL Lu 0. CO w Oc-w i(

Z L/).4 N co M N ~ - N mZ 0 L/ 400-4 4Nu- ~ N -4l ) U ~ ~ W

1-4 0 0 I- - W 0 U) L/) 7 /)U 0 4 u .25)

*j 0 ~ t

* * * * fI *co*4 A *N * 0N-

.* *

* *

L- I* * 0 * * 0 -04 *C * 0

CIOW *J *9 0

0 1. * co . ** *f -a *t

Z0 *- * . 0 0 P

*to* * in *o LAiw~~ M 4U N 00MN N - a c

Wj ca En %n * J 13A. I- S I.

L a- 4L I. A m * * 0 0 0. M

Z-O Un * in *4 fn f* *4!

W ~ W -0A U km ON w 0* - 0 * 4 Q'- 0 f^ *3 X N *1 N c I o

-o 0^ 1 - * .40 * IA 0 -

CC 04 * * *3*WWW WW ix 0 * ~

L3 0n 0'L Z ý 0 4. to .4 t"kn m N O -Z IA 0. 5e 0j N

in .4 < 1 0 U4 N- " . 0~ 0l f- 003 L1- 0 w ly n i M: U CL m

m~ Jc) -1 0 o .0 4. 01-- 0 cc c0izý

Z W0 < < N L n - C zoz -. L3 >WA - a 40 w u C - .

u m < " w- 0 IA fA 0 Z -L l 0 0 x 4 .- 0> ff - 0~ 0 U. 0- < . 4 - 7- 1.-

-4 i Wj < I--C ý- w - '0 kn 0i j W LW -~0 Wj IA >' < 0' inV 4i 0 -% -4 0 o1

cc -tf - N3 z NMz Z 1 0 X Z

0. 1-- 4 u LIa jCf L - L x- )D= U

XZL a 00 -1wC 4 w .4w0 U. - >. m ~ i AL U i z _u o z

OIAW 040 L 0~ 0 c- l IJ L. 1L W. 0 0 DI W%10 0> Z 4 )-- IA I--Z-0 0ý- L I - 04

IA ýq I--W L-u WI 0. u w i U)" 0 t0 3 30x )6-LXN4 WUA I-- >L ccc )-- z< cc< o= t r LUJ 4 WZ cc W We ( c

a- iC Z 40- ý- u - Uj 0 W Sof w-w 4 40. 0 OZ0004 LL O 4 -. WI WW U. 4Zu > D W (L A - C C 0.

W L L4W 1-. LLI 40 QA 0 x 0A1W l

4 . .z 4 t- ý-o Ix cz1 1 4 4te>31 ' -0 La 0 1.- 4 ac 0-

W< e 4~r- Q 4 I A- 3 <In A.0WW~~~ ~ n M-. iw iU iI 4.. c W 3 0 W

I-4 4>--IAI 4W 4- ~ 4 4 45 0 Z26

>. * * * *

*t *

wi W 3 :L3cc : : :U. W

f^ LL. *oU *c a

"Oz u *(A* * *

W u.W W W WZ~U 0 (f *- w

~-z t LA * l * - N * M * M

I.- 0 I.-Z

0.1 *ý *0 * 0 %n * * *0z %0 ým co - 00

Q~3 0 C- * 0 04 0 0Oz 0 < .U '. .4 0 N 0

-1 0 0U x

o o CI-)-cc 04 ri

Le L) x' '0 W Q z rx Li >t UL'4 Z 0 a 00 w~ W ' 0. L M0 W

Li V) W U) z z -9 W 9c 9r vý l L jm- 4" " < X U. W "4 -. 0 - wt 0 X N ) 0 0 -2

r" > ZM W, O I00 x 1-

0Z0 0 1-0 I-= I-O Z w O W.-4 UJ'.4 0.W

>04 ~ o WOZ WUOUW Ezi-0z

W Z W "..W U. I. wr 0001to000qm0-4IX .-. I O -I O L-WI- .W-> >r W >, Wcc~L -Z I.

> X:co I- F-ý- > 0 0>xw0 .t eC u...-4Z. woW 3"- 4 <0 UZ > 00 o~ nf -. ZU. Xfl 7)U

SLu W >.-Z>.W o Zo WOC00 3 00z Z40 0 C4o 0 <'A -43 <0 1-- 0 .- - ue he

ýe* > w O W1 om = = W OLDZ - -Z

C3 4z Z cz lW W ý4W0 IZ Z a -- Z W 4 0. CC C m 04 0 WI- 0 <WI-~4- U 44 -1. ~I..4% 1 Cfý eý Sý(A L _40 0 <A <, 0'0~ 0 u*-

<w z- >cZi 0u mt 00 LI) L

w'~ -q .0X L "Z U.P- 01 1 4x1W O xl-wxl--w W- 4 W 0~ wI w (d Wi le x>L 0-1 a 0-4

00-00-m0 (L I- ----- M -Q.0

1 40 0 c 0 WO W

P- Z 0ý 0 c4 N ~ 0

27

WAW --- -- - -- --- - - --- - -- - -----cccc C4 a 4 -4 N N -4 0 40aN -4 00 0.4,4N .4 00 0 .4.4 N 4 a00 01 4m 4ý

=ZU 0000000 0000000 0000000 0000000 0000000in6-2 11 L

w O0 NMN N N 4040 ItWW -4-0-4 *NNI-W NN 40 .- ' 4 40 1(4OOO0 NOO 470a2ý %0 N 4 ,%4 Q0 c. UA0 N r-44I N A-4 at 0'0'

LU .6. o 04 04-t 4 k 4 nýa .- 4 N o0%o400rn IIN400104 0 (A 40CF. O40.f- V 4 .4?- N in -1 UNN-clNr 4 0 00G 4 N 1- ON0NNM~N 0 mI-MNIN . N0 > L 44fl4kgl 'Q-4-4- 4, 0 4 .N4 t 4AQ9ý4 4-4 -4 N-4-41fLn Q

FACECf ;004;*400000 000000 000000000 0000000

>LU916i -. ------- ---- - - -4 --SUW 44 4 4 444 44 44 4 4 +4 + 4 +44+++44+.*+44 ...LUofcc N A UN LM AW4W NA 4 ItA N~14tm W4

2 2x 'C KK 2'CKKK KKKKK KKKKK KKK)Cw m t .4 -0 N m 4 -4.0 4 in 44-4,0 0 N It -44-4-' Nrft ý.4 4~4N~006- KKxKKKx KK)CKx K XXP KKKKKX~ K K KKK KCK KKK '2-a ( -- - - - --- ----- - --- --- -- - ---- -- -----LU )1- -4OdIFNO all -4 0.-4 c1mNON 0 0-4 -4 m NN 00no4ft .4 0N14 4 m eft0.00A 0000000 0000000 0000000 0000000O 0000000Wz $6il6 16 6 11666 6llt 1 66666

a0- W LULWLwLW WLUL I IJLJ IId, WWWWWUtLUW WWWW tuwwLu WWWWWWWx .4 0fer0 -41-f0Ot' 0 0lrmtIOOOOWI 04$M DItP r40% ON ftn -a0x1I- 444 4 ý44OPf-r- Q o 4 in N - mr- Nakmfl0.n-q4v00N'ww A 000W4) )W0 %U O~ltIo m A U4 1 4N CcN 10 rn c lf 4I0 a 3-Ofl N AtN WI1 a m e

.I. 2 ý4v0 ) mý 4 a.4 00NS90 N .4 0 m PN0 4, 4 ?6-4 9 # 0$m e1Il40ar-. 4InLi0-. 0 0 m~n 1 nC 40110 V - %P. 46-0o -p..n WI-f1*1W l- o Of $0 a- Nw 4 1- 4 m 1-0 In l4It4QN -. 4 40'Or-4 01 0 m m in o m4N4j V 4ý 0ýý .46- 4 .4..4 0 ON..1f.44Inq O 1N -4 UN.4 N. O4.1-4.4mm N

w t. .0 ** 0* 0 . .0 .0 * * 00 .0 **?* ??.#? 0 0 00

IL09.; a4f1~ T T T T T :;TT: T;T ;:::V,

x I- to -4-40N~t -1-401t -q~4 f-40ýtl .4 -4t 4; 0

-. 1L45- a 4.4 m m 0 flI 00-4Q~01-40m 0 4.4m 4 0 m ý4 04m 4 0 m ¶-.4 m 0 Itf60 1fUJfNOC AN- Qcft 0-4cirOQO N000000t> 00q49 -4.

LU. 6-4- 4Na-0MN=-% infl- M0 OkC47 iNttN- 0 -4I 4N Oct1.0 C) It 0.OtNtOa0% ý4Ntr- 410 in4 1.-4 t N r-0 12.O- 4ý ý004 8 r NIAm 10CN4 m CN 4Q04fNII NONlPMP- i'4lNt0tO LU cc N ýiq* r- A 0 -4 40tt.4N Nn0 ?- 44.4f1f 1 1flP.WINMIEA?-(PcNi O t

%M-. Li mt -4 ýI4NN N If fNLn - mcfm m-4 NN0 -4NN f-4 ' NMf4-4 -4-4 N't4-tcM

ILK 000000 0000 0000 0000 0 0000IL'S z6 I I

0N NO0 4

ý e.4 0 0

LU I.-2 0 6-0 0 0 cc 0 1.-0cz). f-0( 1- 40 W-A 6-ez xz I - H W H l- LU H c04-J- zW ZomtnC LU42U CL0 0a 6At- C4 4c x24 a .- 6- j46L4 24 2 25 2 111 x 0 2u =2 2

WU- LW 4c Wj 40. Wj A-. LU (AW LU 490 LU40 x cc - -- or - cc-

t. I.i -2 42 Li 2- Li 6-LU.111i u 0 0 - 40 44

V) I. 0- IL t 0- IL L. L 2X IL 04 ILU-jN 1- 0:) IL 0a 4LA 00 AL x20 IL oS It=x 4 20 Ms 2U 0 LU 2 0 LU. 0~ 0s r.1 as .

x4~z ce 0 4-%. 0 4<.. 0 Li 02W L z z Li w2 x Q waj0Q 1z

w Lx t acc2- 0 2 04 2j 04 2 W4 2 - 003 2

4 UW; 0 L40 L40 6i0 Lim- 0m 4- 6 LUJ '

2.40 tm6 6- 1.- - (36 95- cc 60K x 2c 4 2c 4e 2c 44 2W 4 2 4.ý0W '- CC -w 4 .-.W 4i . 0 > 4 -4 4C-- 0(3 0 0 gi (a> as 0 > as 2 LU 00 Ili42x(t 41.9 . 4- cc -t -40 g - CIE 4tL cc 4i&.- at

0 - LU ). 1z LU ). cc LU ). cc lEO0 cc LU >. cc(3>Wcc4U- 0 x IW4 0 c 2LU4 0 6- 0 5W4- 0

-C I- 42 Li 42 1 42 Li.z 0 Li 42 Li4j LLiVC) U.LQ0QILLi Q 3.11. uLLUQC

20.6 - 0w- LU. 0.4 LU. 0'-" s 00m LU. 0' W00 5W 4. 50 4l 50 =4 >. 4. -0

cccc uo.d( ccwIn> IL a LIS lhJ . Ix t- 0. cc ii, 0.006 aI. W>1 .. W0 o" W>0 -4 LU >-. . W>0M4 2 0c cc 6,- 0D cc 6- a c 6.- 04 6--.1 aWL UW 2lE6u c - 4. xc Ix El 4 24a 11 2.l cne =0I- OW) 0U (AW 0 0)1110 01411Db(3 -44j 20. 2L 20. z 0-2-2 2.W. K 4 ILLcc ->o -0t t

28

N In In tn m N in Ln In m N tm tm $A C4 tf% in IA 4" N tn In inxxxxx xxxxx xxxxx xxxxx xxxxxIn ý4 N N f" In N tn .4 N N m n N In -4 N N m I^ N in ý4 N N m In N In .4 N N m A(^Lq vxxxxxx xxxxxxx X*XXXXX xxxxxxx xxxxxxxwtu ------- ------ ------- -

ce ac cc 0 -4 ro N pq ýo ý4 ý* rq N m M ý4 N 0 -4 0 -ýp P4 0 .4 0 -.4 0 m -4 0 0 -4.4-4 m N 0Mmuj 0000000 OQOOOOO 0000000 0000000 0000000tA)-,z I I tL I I . I I I I I I L I I I I I I r I4 0 LLI Lu tu Lu LU LU LU U.1 w Uj LU W LLI to tu LU W LU Ui LU Lu w uj uj wuiwwwwwLU 11 CL tA l'- W ol (41 ON w It ON P. co (V 04 0 0% Ot 0 0 pw P. M -* a .4 N 434 -* 3 co 1ý- GO w f.

In 00 Cc tA 0 (" N f",Q -4 fý .4 4 0 ew N 4) kn co Qý It m 00 N It In ap 1-1 0 f" t-It 44 o 1- N -t 1-,Ot 0 4 -N * m g a* m ýd r. $ m 'eve -W co 43 0 4 0 f. P. at I-. V1X x LU 0% 'Of,-(*-* NM (40 ImIrloo -0(7.knp- 0 C% (14 -t 0 ft 0 .4 4.4 t-4 QwUJ - ý- 't 't (" ý4 ý4 0,0 1" r- ý4 a N 0 N CC N tM t W ý4 M 4 N ý41 .4 a m m -4 0 klý 4 .4 A .01-0 41 oomootno 0 ý4 -4 m 0 N 1,ý 0 N 4 in t- 0 w N -4 ý4 N 0 4 .4,0 4 A N d m

> Q 0 LM m ý4 m m m -4 ý410 -4 0% N N z 4.4 0 NCL "

orga i8880-80'#> LU

M 0de L.) Ui + + + ++ + + + 4 4,w

I- ce N tm It It LM N Ln ýt It LIN N tm 't 't im N im kn Ln It It LMz xxxxx Xxxxx xxxxx xxxxx xxxxxLu M LU .4 't -4.4 N M 4 .4 It -4 -4 N m 0 -4,r ý ý4 N m 4 ý4 ý .4 .4 N Q -* -d 4 ý4 ý4 N m ýtc3oý- xxxxxxx xxxxxxx XXX*XXX x xxx x xxxxxxxZ-Jn ------- ---- ------- -------LIJ ý1- c a -414 m 14 m -4 N LIN m llq 0 '-1 N ý4 0 C,$ ýn .41-414 ý4 0 m N0.00 0000000 0000000 0000000 000000a 0000000LU z I I i I i 11 1 1 1 1 1 1 1 1 1 10 4 LU W.WLU LU Lu ui uj uj wwwwwww .. ii w1ii wiiiiitulm op-,tolmm 0ý 0 MI. cpý m MO(AMMZO at (" 0 * 0 m " M0404404x 1- 0 (1 (A km 0 IT ýr m a Qý 10 It r" '0 N r. 0 .4 N 4 .4100 IrNtA m ý4-4 Ift fý -0 mui Lu OtAONOOM ýr 4 0 0 N A tA 0 0 m N %M 0 N 0 0 N -4 m rý r. w ý4 r- 0ý w N 4I- z 1- m m P. 0 ý4 m p N It 4 ý4 10 0 UIN r- -0 tf% 4 LIN tm ý! -4 0 A N 0 0 -4 ý4 ý4 v 000.4 0 't N kn M 0 0% to r- 4m w (a N A 0 04 0 00 of-. NONT N Iý v a- -4 0 Lnw I-. ONOOMO-4 Ln Q Q ()ý a 0 N ýo m m 0 0 4 .4 0% Q w -4 w m fý N a N N 4rý4 ýo ý4 N .4 tn ý4 ý4 P. N 04 .4- .4 N .4 m m N N ý4 In N Pý- m M ý4 UN .4 N m tm NW>&

.4 U. ------- ------- ------- ------- -------

LA. LU 0 . . . . . . . . . . . . * + + . . . + + * 4 . . . . . . + + + + + +!! . -4-41 (n cl It ý4 It rn m It ý4 rq m It ý4-rffllm-r It (VI (n It0 ac w xxxxx xxxxx x XX x x x x x x x xxxxx

1- 51- = Q -4 ý4 ýl Q It M ý4-4 14 Q 't S2 ý4 14 -4 In Alt 141.4 ýQ It ý414 NQ 1+z xx_,x xxxx x xxxx xxx x Qxxx x- - - - - - - - - - - - - - - - - - - - - - - -LU U. Z m -414 0.4 0 0 N tn M .4 LIN a N -. 4 m ý4 0 3 4 NN-0-0-t ZIMN-400 0 C2 0 cp 0(ý U000

ac I I I I I I I I I00.4 wwwwuiww wwwwwww wwwwwww wwwwwwwU. I- N m N 0 0 1-4 0 0 P- N 0 kn 4 fý 0 -* N .4 oto fý 0 4 r- -4 -r N 0 m 4 mf-MMNWMN m M P- A Ln .4 m r a N,* tn M (7. Fý LA 'A w 0 ý4 -V .4 m 0 Iý in -tu 0- 4 4 -4 N 4 fl- tn 0 10 LA 4 It w 4 M 0 M M Qý p.. 4 N 0 M 4,N 44 r- 4 -r m m w f--i -- ac CA t 1ý ý4 0 M M 0 Ln ý4 N r v 1" 4 r -4 Qý Ift v m m U% w M -4. W -t ýn 0 0 pý;pco LU 11 M 0 0 ý4 r-4 .4 tn 0 ý4 %A Q 0 r- Pl. ý4 a, 10 Pý Ln .4 m 0 .4 0 w ?-. -ýr 4 -4 Cý fý ý^ Ln 0ý 11 .4mmmomm 0 0 m 1-4 0ý 0 4 a 4 0 10 le .4 14. Ml4mcomlt -4 0 Iý 0 N .4 40 tr% UN N m P- 0 m .4 N -4 M til Ln M r- p- .4 M,4 ývm X

LL X ol ol 1; 4840, 0, 1;1; 8 4 4 4 1; 8 1; 8444448 8 8 4 (ý 8 8 8ILLU 1;ONU ------ ------ ------ ------ ------

+++ .... ...... ..... ......

Q N LU 0 0ý tox cc Ln 0

.j x 0 w 4 Lu Wj C; tu -iLu uj 4

C.5 I- I- z t- zCL ý- 0w LLI

LA Z x 0 z LU x LUu CL a 0 z; l'- uj 'A.1 1.- 0 0 1.- (A CA0 .9 z z 0 z ZZ z z .9 z z .9 zUJ - UJ 4 x Ill LLI Lu .4 0. Lli -C = U.1 -C x LUj T 0 _j ce - cc 0. (9 iLCL x u z 0 0 Z 0 0 l'-It Q 0 0- - 0 LU U.1

0 (A LL z X U. U. 0 ti. (A I.- U.N I- Lli tL -d Lu 1.1. 00 %L M U. 93 :) LLx -d cl -i U.1 cc > U) z Ul Ill z C) ILI z C3 U.1N:zz Lli U 0 M 4 0 4c ce C) < ae a0 1.- Q U. 0 T Q m z 0 w z Qfm N I- I- z 0 z z ui Z uiZ " = ;ý uj z Lu z a -i z 0 z 0 z<(n 4 > 0 tu Ln - 0 (1 ui Q ILI 0LA - :m LU (1) x -Z ý4 Ln z U. I.- I- *. tz F- 0 1.- Q C5C3 x oc 40 < LU >. z < z oe .4 x z0 UJ (9 -1 CO ý- 0 -1 - UJ -i " Cc -A ý cr

1-011.0 1- f^ ui u tu 0>- LU 00- LLI 0 0 - LU< 2: 0 w C9 -i -j LU Cie -C 0 >. ac 4c LL. )". cc -q LL >. cc;) " Z Q. cc < -.10 cc LU -i 9z w j cc LU -i cca ;-- Of tu >. 0 > g - 0 X LLI Z 0 M U.1 z X U.1 Z 0Lu 4 Lu I- Q C4 - 0 -j tj i Q i:x - Uj - LL Q U. Q LL Vý- -1 0 LU ý- o a Ill C) ý V) w 0."0 U.1 0 - (A ujC) cm Uj -i a -i z C3 L" -i X 11. -1 X >co z CL Z Z. IL w Uj CL CL (z U.1 C). e6 0:

u c - LU > W LU > .1c ILI > cz(A z uj 0 1- uj z co cc co coLU 11 LU w 3c > LU x X x -j x cc :3 -1 x ac 31 -4 z CC 09Cc M I.- CL tA 0 0 :3 :) LU LU m :2 U.1 tu :D LU -jCD -4 j C9 Q x X 0. - a. -LU X -4 LL

> co at 0

29

N i N .4 N N M A r NU-4 N rmtfi NISCYrMNNIry m W NMt-4NNMftA NWI-4NNMflU

USU ----- ------ a-a- -- a---- - ------- acc cc at t-4-40- *0 4 0 NO0NOa-to .0 IN14 0'.4 0 .OaNNO40N(

=xWj 0000000 0000000 0000000 0000000 0000000o

4 0 ILI~ zdwd usdidiuw w w uý wIIwII I I u u US di w WU di u i ts 1111 wu%us 4 950 5-Qt CV M0 N It .0 4 m f0 at0 FftItI-atN t4ý 4 .- f0 -4 00 InlUNOV. NIn WN

8 t .4 0 A t00 W%ý4I)Q N 0 0ený %nocq ýnan 00.4q aot# COI m N a $ O4.Ut 04-4hVC *S -V N0 I W 4.4U(F.0~4 -0 A NMInN44$ýv M04kW Ný ,*0 n ýNnk0 4In C0N4 44w tme No0' .4 7NN pý In4. 0 4 t

W.-9- £N01l N MC0O.- 4 a - t 00ow0 4N- 4-ACAN -,ý 01¶ O I - f"'4%n(A 04 N0f-4N -4sif 4 M tm.4* NNý400O904 *o .4% 44 4rotN 4

0A0 -LLD-- - 4 - - --- - --- - -roWt E +44*++ 44+4++ +4+4 44+4++4 +44+4+

e NIA -r in NW~in NinIt X0 LA NIt 4It NIAw+ýt A

us x W .4 4.4 -4 N;2 -4t- 4-4fNrn M t'44 ttM -4 # 4.4 N M t .4t -#4dN MtCl0t- )C)OXx x )Cfflxxnx XXn"4X4ý xnxxx xxx

US 31. M0-ee N-40I 44 '94 -4 NItN N --4 -4-4 -4 M.4 N ~0 0 N-4 ý40.0h 0000 0000000 0000000 0000000 0000

x 0 g't ZS P.WI mg 0 8M0 0Q A MsoI' CenýtN0M1-N tml 019 Mtas(30 4 tngomofOm0 x5- MUNN 'DoomCQ N4O'a .4Q'4 4 ut M P.N'.4 tUN MfIN r-u 4ns N *V0rCOMO00Q

ww ý4W 0.0 4 .0 0 0 r-40-4 -40(tr Y ~f0% t N N %ItMI~nnnM to0 0MMq t4I - 2 o COOCO%0-4 -* 0o0tO04It#oI N -4 9NNr2 NOO'1N 4 JANP , 4- N at NO'Nt

WUI- tt0Nfin V-t4 * o M n MN kn a to IwnrP-r fl. m r w M-4 00. iinw0 IAACON4j 04 r-- Noo 4 -4 N .4 - N tfl-4 tn Q-4 N. N Em . -4N -4 -4.-N'N 'JNNW ^n> 9 99 # 999 0t 9* 99. 0s * t 0 ~ #0 >0 0000000 0000 0000 0000 0000

-4IL -fU.O 4 4 4 4 4 4 4 4 4 4+ 4 4 4 4 4 4 4

X- I.- w tfl4M44 -4 At M944fl .4 -4 4 -4a at

4a 4mZ~ AL tn#,1xl -cxxxxX XW1LX. eN -4 M.4 0* NON.0 4 014 N d4 n4Ot4 N -4'*- 0t ; 4N~f

-O 0 0 0000 0000 0 0000000 000 00 0 00 (a00

004 wwww USUSSUSUUSU USUSSUSUUSU USSSSSSS wwwvuww wwUwwUww- .IL51- 0 -4r- 1ON-N *t ?.N N M 40 tm ?- t N a t .4 4 0 UNý4 zN -4 fM.O 0tfcm

o- ( ýtr-oQsP -4 N m an ,nN M M l N 4 tAh' 4 -4f4'r Z 12 -4 -Fl, %n m o04tUS I. 5-4I- M I9rM-4,th.M'c Nt -IN 0-4 m tfN M 0 10 I-4EtI I.flt 4 4 0 %n N M-4 4 ?4j =~ cc *%A 2 0 M- -4 1r.0 Ma GO 't 13IAOM Olt N Q-10r )%-4 0 tLifl M OIV tN0IrC us 55 4QNM 'Oa O 4 a 4 0 m &AO 14 N N = MN CR 00a$Mw An NNtnNOC.44 '- It $Aa5 tm N M CýtfQ. Mtn O4 m NMf -4 m '4 m 4 N N Q0 00-4 0 -fUNM N

5190 '

4 V

5.- U0 l (v NNIt 0'O4tA 10 P.O 0-4 - .tt4n N4V-* 4 -4 - ý-. -4 4%-10 -4t .. 4.1fl4 0 4..4N'

ILK 0000000 0000000 0000000 0000000 oooooroo

0.0---------------- - -- - --- - -- - -

ON- -U Nw -M- 2Ne Nm

4 00 -C 0o 0450 x I.-2 9 9 US9 5US 0 InC 0 0 1- 0 Urc).-0 Wj n us- W(n 5-I 0e4z t- US t5 05 US I.-U if - i 55If

0.7 a t CL14In 50.0 CC 5- CLtna 5- XC5 04 5CI- 2 5

I44. US 2 9 40C 2e Wz 2j 25 2 4LUSUS I- CL US 40(L.- US I.-0 US 40. US US-inr X- a 0 f - 154 ae- U0. 42 0 Q >- o t 4, z 0 Z- 5

5It 0 a a0. a - 0-aL4 - to.- aL cc5- IL Ina L In4A-. IL 0 I

4 5 Len IL. z US IL 0a IL D.an ILt WImDt-t z4Av Zj n US USU US 20 US V) US 2 US

42<-. 0 0.IL m 0 c - 0 g -. 0 US 0LUSC 0 CieCC- Q X x XX 0w

4 ý-US51- 3: XZ Z 04 21 0. US 2 04 21 04 21 05t- 240 0j 0) -0 0 04 0 0 0 ILU U

4 t- 5-4 1 - US2-0 0 cc CC cc 5- 5 0 5- CCI 5-zc . -A'-sOWu '- US 4i ccC 4 u -w 4 'IU ac 0C 4t-00 ( 0>- USJ ixCjUS US 0 >- US O>a US (ZU US

4-9 03- CC tUSCLI- cc -40>- cc <403)- cc US zccLu aj US 4 cc CC US 4 C: US 4 CIE - cc

Q>CY ac = l 0 itý- 0 0 mZUSIz 0 Lai 0 14 0aLUSCI1- 40 0 0 C Q 40 cc 40 j Q5 0iQQI.

.j I.QLC0 402. QL ILO 40Q20.5L - Ca0 US 1^''U L wUS 0-In US 0 -I US) tu cc US1

004 5> 4i 5-4i 4 > > i 4 5 -'- i M~O 0. 1.D0 CL CatS0. 0. cc US0 . 0. In0.

f0- 9 U>C ac In 2 L- U>CC a U>C af 1^W aIn 2 Ca 5- x In~ - am 5'- ca 5- tinWIUU IfC tuu x c: 0 0 cc x XCCac x SCie % 4 0.44cc m I- =USU CCCC w cWc W =W D: = j:IU im cCIm0 -4 4 20a.- z Qtt XL 2 . - 5 20.' X 00.US1 X 4C ILI

-4 4 '4 -4 4

30

.2N Inb-4 N!(14M U N A .~N N M A f4 t ~4 N M L Ntm .4NNflMtn

Ww LU-- - - - -a a a - a ý a a.% - - - a---a-ac c c dom mm M 04nW4 N . 00 .4 a M 0. 4

=c 0000000 0000000 0000000 0000000

At 02 LuiwW LUWt iUjL WWJJJLAI IWLuuu WW w WWH110 NN P .N ~N O% 0. 76c or. t 'ON cn k t. ~0 CO P-N

1#04 olt No 4 ON a0 M o6m1 4 tM -%P W .40" I O 0

it x tat C* ItI atU 2sAIn14.4 MdN * AO tLA' gurpmo

ocn 0000000 0000000 0000

>LU"

Z 4. q - ------ a-- a-a----- -- a- -- a-----deU ¶). 4 .. A + + ++ +4 & + + ++ +

z -1-0 *g- -%~ g - %ý -- ---- - ---

ui - ". .- 4NmN It 0 r4 U4MA - 400 M p 0. dý-4N -4P.ý4ý4N 4

mm~o^ 0000000 0000~009 0? 00000 0000000uz iiii i i I i

Qa 4 ýd a a4 -- n a% a a al a a 0 a a a aN $ A a t a -40

04( OOQ0OOLJ Q00'a000 0700000 0000000

.4 U..'1 vO~~ N

5 .- coM 1- 4 4Q t- -- 1I -4- ý 4 ;z~4~ ýq0 1Q411-4 MI

u 0 U. II ".4rIt'CMN 1 0 N.n44 .0-t m Nar-UI-N m 4 N-

Q 0000000Q 000000i0 0Q 0 00000 00,00000,ce jI I I I I I fI

0 N tU -ýI Q-A Nt -r A m - -Q o o M --- Qt -- w-M

Z - I. TO OPN 'oP. 4 It 00 N(.I ,ONW z ui '0 -4U mQI . 49sý . D"a00wm M0w ý

co 0 '01 4M M N 0 It Q04r' 4tm00 NI-"0-400 0 .r 0 0 LU '0 kAMM4 a 0 '0NaQLn1 7

MN0 e4ý4 4 m -4I-r * w4N04LnUN0MLU9#9 0 WI 0 Z40 0 3> 0

U'-.

X 00 5-4 5- 0-. 1-W0-

LU-lW LU a- LU 31% - LU 4-ZL

U: 4 0 U -Z- z I- 00 04 4j a) .4O .4 2c

ac >.a U. 0Wr> ". Woe1 U. 0ad1 ox U. I-i- Lo 0 LU U. 11

0j -2 0Z 4U. 0L ZýW. n W. 0 "2 U. 0WU UD>

0U NI- LU w4 01)-4 9 4I-Q

a. > U Uj Z Q~02 1.U 0 ZW4 q U.z 0 -

u 1. :D - U. 1- c.- U. 00 1- L-jNLU U. 1.- U 0 LA = u4W LL XI.4IU

mx-0C -z LU U0 14 LU ~I'-- LU WO4 LU

z .m 04 cc 0L-UW 0 x 044-L2 0 I.0 0 tu= Z 0 0 W4 00

LU1- 2 U U 1-U U 011 U _2j UC4 a 0 '.'. 4.L a.4

.4..L z0 0 0. W4 cc 'a . 0.1/501-e ~ ~ ~ ~ ý 3:. 0 4 s51 4WIILUW 0I. - W cc 0 - 4 oaWO

o0 - ~LU < 4).4 0 <1- ac jI 0a-4 0.- > - oa j 044 3i 011 j 1ý

x U > c L ZU.ccc

1- >. czz0 Xic 0b 'aw--1 0-4 04%-4 a.-4 1

z M ) w LL XX u a: - > ui n Xz31

TABLE III

95% CONFIDENCE INTERVALS FOR EACH OF 19 SELECTED DEPENDENT

RPV SYSTEM VARIABLES

Measure Minimum MaximumLimit Limit

1) Number of heading commands transmitted 15.13 20.11per vehicle for the enroute phase(strikes only)

2) Number of heading commands transmitted 33.71 46.95per vehicle over all mission phases(strikes only)

3) Number of heading commands attempted 34.99 48.31per vehicle over all mission phases(strikes only)

4) Number of transmitted commands per 46,29 57.45RPV, all types of vehicles

5) Number of heading commands transmitted 17,67 24,95per vehicle for the return phase(strikes only)

6) Time between transmitted commands per 82.88 110.32RPV, all types of vehicles (seconds)

7) Time interval between transmitted commands, 4.04 5.28all types of vehicles (seconds)

8) Number of heading commands transmitted 18.98 30,88per vehicle over all mission phases(EW only)

9) Number of heading commands transmitted 10.47 22.31per vehicle for enroute phase(EW only)

10) Number of heading commands transmitted 10.47 22.27per vehicle fdr enroute phase(Low Recce only)

ll)Number of heading commands attempted per 22.81 36.31vehicle over all mission phases(EW only)

32

TABLE III CONTINUED

Measure Minimum MaximumLimit Limit

12) Cross track error during enroute phase 218.03 288.37(strikes only)

13) Number of heading commands attempted per 22.02 36.90vehicle over all mission phases(Low Recce only)

14) Number of heading commands transmitted 18.16 32.84per. vehicle over all mission phases(Low Recce only)

15) Cross track error for enroute and return 249.37 367.53phases (strikes only)

16) Proportion of vehicles recovered (EW only) 0.000 0.075

17) Ratio of distance actually traveled to the 1.0012 1.0028length of flight path

18) Cross track error from H waypoint when 243.16 612.22pilot took control

19) Ground speed error from H waypoint when 907.86 2652.90pilot took control

33

C--

<2!

<2 _ _ _ _ __D

2W =<-D=I-C 2CDl CD -J C

C,/

Z~Li-

0 M C0

-j-

CL:)

L<LU

w C10jC1

CD-

0: L0

Si4 L34

FIGURE 2 -Teami of Five Operators Performing in the ANRL RFY System Simulation

35

a:w

0<0

9wwcrg

01 >

w ~

axw

Z

~3

o- W-A

'16U4"

0~

LS.'

(I,,a:o~02ILZU,-

w3US OENETPITN OFC:17 5-3/5