1

Dr. Bernard Ferrier, Dr. John Duncan, Mr. John Nelson, Mr. Dean Carico, Mr. David Ludwig

Further Validation of Simulated Dynamic Interface Testing

Techniques as a Tool in the Forecasting of Air Vehicle Deck

Limits

ABSTRACT

Validation results are discussed and compared inconfirming the tendency of certain parameters beingwell represented by simulation with the actual at-sea result. The primary objective of this field ofstudy is to determine the feasibility of applying fullmotion simulators and plug and play simulations insupport of dynamic interface at-sea testing andexperimentation. Several years of simulated flighttest programs using the Merlin CAE Trainer Systemat RNAS Culdrose (UK) and the Manned FlightSimulator at Naval Air System Command, AircraftDivision at Patuxent River, Maryland have beenconducted. A typical simulation consists of modular“plug and play” components contributed by theproject participants assembled in a High LevelArchitecture (HLA) with the combined componentsreflecting the ship environment, the long and shortterm prediction methodologies and the ship’sresponse mechanism. The use of 6 degree-of-freedom motion flight simulator to forecast physicaldeck motion and deck motion limits, is discussed. In the manned version, the simulated flight testhas several objectives including assess thecapabilities of the Cockpit Dynamic Simulator(CDS) to support Ship-Helicopter Operational Limit(SHOL) / Naval Air Training and OperatingProcedures Standardization (NATOPS) limits;demonstrate High Level Architecture (HLA) withselected modules; and determine feasibility ofapplying these simulators in support of dynamicinterface at-sea testing. The unmanned andmanned systems studied focused on specificsimulated aircraft–ship interface responses. Anapplication of this simulation is the forecast of decklimits computed by the motion characterization of aplatform in terms of, and as a function of, thedeflection of landing gear configured for verticallanding aircraft or rolling vertical landing. At-seavalidation study results are discussed andcompared with simulated scenarios. Thiscomputational method employs sufficientperformance criteria and correlates well withforecasted quiescent windows of deck motion. Results are presented in relation to the deckstability problems normally confronted by ahelicopter during recovery in progressively difficultconditions. A brief synopsis of several of theintegrated HLA modules representing variousaspects of the maritime environment is presented.

ABBREVIATIONS

AGL Above Ground Level

AO FNC Autonomous Operations FutureNaval Capabilities

ASIST Aircraft/Ship Integrated Secure and Traverse System also RSDBRC Base Recovery Course

CBT Computer Based TrainingCD Clear DeckCDS Cockpit Dynamic SimulatorDDG Guided Missile DestroyerDI Dynamic Interface StudyDIPES Deck Interface Pilot Effort ScaleDLQ Deck Landing QualificationDOF Degree-of-freedom

EI Energy Index (Value calculated byLPD)

FD Free Deck Recovery (RAST trap usedonly)

FDO Flight Deck OfficerGCA Ground Controlled ApproachGUI Graphical User InterfaceHARPOON helicopter handling Sys (UK,USCG)HCO Helicopter Control OfficerHLA High Level ArchitectureHSL Helicopter (Attack) Squadron LightKIAS Knots Indicated AirspeedLCAC Landing Craft Air CushionLPD Landing Period DesignatorLPDLOOP Landing Period Designator softwareLSE Landing Signal EnlistedLSO Landing Signal Officer

MFS Manned Flight SimulatorMRU Motion Reference Unit

MST Mechanical Systems TrainerNATOPS Naval Air Training and Operating

Procedures Standardization

NAV11 Landing Period Designator software

NIREUS NATO Interoperability andRE-Use Study

NVG Night Vision GogglesONR Office of Naval Research

PCCS Portable Computer ControlStation

PTT Part Task TrainerRA Recovery AssistRAO Response Amplitude Operator

Report Documentation Page Form ApprovedOMB No. 0704-0188

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering andmaintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, ArlingtonVA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if itdoes not display a currently valid OMB control number.

1. REPORT DATE 2010 2. REPORT TYPE

3. DATES COVERED 00-00-2010 to 00-00-2010

4. TITLE AND SUBTITLE Further Validation of Simulated Dynamic Interface Testing Techniquesas a Tool in the Forecasting of Air Vehicle Deck Limits

5a. CONTRACT NUMBER

5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) Office of Naval Research,Washington,DC,20375

8. PERFORMING ORGANIZATIONREPORT NUMBER

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)

11. SPONSOR/MONITOR’S REPORT NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited

13. SUPPLEMENTARY NOTES American Society of Naval Engineers Launch & Recovery Symposium 2010, "Launch, Recovery &Operations of Manned and Unmanned Vehicles from Marine Platforms", December 8-9, 2010 Arlington, VA

14. ABSTRACT Validation results are discussed and compared in confirming the tendency of certain parameters being wellrepresented by simulation with the actual atsea result. The primary objective of this field of study is todetermine the feasibility of applying full motion simulators and plug and play simulations in support ofdynamic interface at-sea testing and experimentation. Several years of simulated flight test programs usingthe Merlin CAE Trainer System at RNAS Culdrose (UK) and the Manned Flight Simulator at Naval AirSystem Command, Aircraft Division at Patuxent River, Maryland have been conducted. A typicalsimulation consists of modular ?plug and play? components contributed by the project participantsassembled in a High Level Architecture (HLA) with the combined components reflecting the shipenvironment, the long and short term prediction methodologies and the ship?s response mechanism. Theuse of 6 degree-offreedom motion flight simulator to forecast physical deck motion and deck motion limits,is discussed. In the manned version, the simulated flight test has several objectives including assess thecapabilities of the Cockpit Dynamic Simulator (CDS) to support Ship-Helicopter Operational Limit(SHOL) / Naval Air Training and Operating Procedures Standardization (NATOPS) limits demonstrateHigh Level Architecture (HLA) with selected modules; and determine feasibility of applying thesesimulators in support of dynamic interface at-sea testing. The unmanned and manned systems studiedfocused on specific simulated aircraft?ship interface responses. An application of this simulation is theforecast of deck limits computed by the motion characterization of a platform in terms of, and as afunction of, the deflection of landing gear configured for vertical landing aircraft or rolling verticallanding. At-sea validation study results are discussed and compared with simulated scenarios. Thiscomputational method employs sufficient performance criteria and correlates well with forecastedquiescent windows of deck motion. Results are presented in relation to the deck stability problemsnormally confronted by a helicopter during recovery in progressively difficult conditions. A brief synopsisof several of the integrated HLA modules representing various aspects of the maritime environment is presented.

15. SUBJECT TERMS

16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT Same as

Report (SAR)

18. NUMBEROF PAGES

19

19a. NAME OFRESPONSIBLE PERSON

a. REPORT unclassified

b. ABSTRACT unclassified

c. THIS PAGE unclassified

Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18

2

RAST Recovery, Assist, Securing and Traversing

RCT Rear Crew TrainersRN Royal Navy (UK)RNAS Royal Navy Air Station (UK)RSD Rapid Securing Device (also ASIST)

RTI Run Time InfrastructureSAIF Ship/Air Interface FrameworkSAMAHE Helicopter Handling Sys (France)SHOL Ship-Helicopter Operational LimitSMP Ship Motion ProgramSMS Ship Motion Simulation composed of

routines identified as NAV

SSD Sea Systems DirectorateTCS Tactical Control StationTD Test DirectorTP Test PilotUAV Unmanned Aerial VehicleUGV Unmanned Ground VehicleUUV Unmanned Undersea VehicleVLA Visual Landing AidVMC Visual Meteorological ConditionsVTOL Vertical Takeoff and LandingVTUAV Vertical takeoff and Landing

Unmanned Air VehicleWOD Wind-over-deck

WST Weapon Systems Trainer

INTRODUCTION

Office of Naval Research (ONR)’s LandingPeriod Designator (LPD) project is part of itsAutonomous Operations Future NavalCapabilities (AO FNC) Unmanned AerialVehicle (UAV) Technology program.Autonomous Operations (AO), one of twelveFuture Naval Capabilities programs, is thecapability of performing missions usingunmanned vehicles in dynamic andunstructured environments with greatlyreduced need for human intervention.Autonomous Operations supports three majorcommunities: UAVs, Unmanned UnderseaVehicles (UUVs) and Unmanned GroundVehicles (UGVs). This project is part of theUAV Autonomy program which includesintelligent reasoning for autonomy,technologies to enhance see and avoidcapabilities, object identification, vehicleawareness, vehicle and mission management,and shipboard landing capability. Its primarytransition target is the Vertical Take-Off andLanding Tactical UAV (VTUAV) and theTactical Control System (TCS), with an eye tothe envisioned future programs on the NavalUAV Long Range Plan. The LPD system will

support the effort to increase shipboardlanding capability of vertical takeoff UAVs.Technologies were targeted which wouldaddress the Enabling Capabilities (ECs) thatwere specific to Naval AutonomousUnmanned Vehicles Mission Needs. LPD wasselected because this product directlyaddressed an Objective OperationalRequirement. Additionally, it was anticipatedthat the LPD would be useful across manyUnmanned and Manned air vehicle types andit would be able to interface with all shipclasses.

U.K. DEFENCE EQUIPMENT & SUPPORTOrganisation (DE&S)In the United Kingdom, the LPD tool is apromising technology improvement projectwithin the Directorate Safety and Engineering,Sea Systems Group (SSG), Simulation BasedAcquisition program.

DE&S equips and supports the UK’s armedforces for current and future operations. Itemploys about 20, 000 people with an annualbudget of about £11 billion, its Headquartersis located in Bristol with orther sitesstrategically located across the UK andoverseas. DE&S acquires and supportsequipment and services for ships, aircraft,vehicles and weapons, along with informationsystems and satellite communications. Inaddition, DE&S acquires sustaining andongoing requirements of food, clothing,medica l supp l i es a n d temporaryaccomodations. It is also responsible for HMNaval Bases, the joint supply chain and BritishForces Post Office (BFPO). DE&S worksclosely with industry through partneringagreements and private finance initiatives inaccordance with the Defence IndustrialStrategy (DIS) to seek and deliver effectivesolutions for defence.

The Simulation Based Acquisition program is acomponent of the Sea Systems Group whosemission is to provide whole ship andsubmarine safety and engineering services to awide range of customers within the MoD andthe naval construction industry of the UK.

3

Simulation technologies continue to evolveand can now offer a cost effective means ofsupporting many stages of the DefenceEquipment Acquisition Lifecycle. Manycomplex interactions between systems,operators and phenomena such as ship airwake and ship hydrodynamic! flow fields canbe accurately predicted, allowing manyscenarios to be examined in a ‘virtualenvironment’ that could never becontemplated using traditional methods.

Simulation investment in the earlier stages ofa project can ensure that the project issufficiently de-risked before the majorinvestment decisions are made.! In addition tosystem performance prediction and riskreduction in the early stages, further benefitsare realised if the same simulation technologycan be pulled through to support training andin-service activities.

The sharing and re-use of simulationcomponents and resources between nationsmakes the technology much more affordable.NATO Sub Group 61 is studying how thisconcept could be fostered by the introductionof new simulation standards.

A number of DE&S naval projects have beenworking closely with Sea Systems Group toidentify suitable applications and are leadingthe world in re-using key simulationcomponents such as air wake, ship motion andLPD.

-Experimental / Simulated DynamicInterfaceIt is important to develop options to reducethe cost and cycle time associated with the testand evaluation process, to enhance theproductivity of flight test team members, andto improve the safety of flight test operations,especially in adverse environments. Flighttesting is required in both land based and seabased environments with a variety of testaircraft and ships. Simulation is often listed asone option with the potential to help reducethe cost associated with flight testing [1]. Thesimulation of helicopter operations from navalvessels provides a unique set of challenges.

As described in earlier articles [2] simulateddynamic interface strategies have beendeveloped over a number of simulationprograms. The earliest High LevelArchitecture (HLA) simulation was calledNIREUS (NATO Interoperability and RE-UseStudy) which was followed by the SAIF(Ship/Air Interface Framework) programs. The purpose of these complex programs is touse the HLA standards to integrate air vehiclesimulations, ship simulations and environmentmodels to aid assessment of the dynamicinterface for a range of helicopter / ship andUAV / ship combinations. The initial phase ofthe SAIF program has been focusing uponSHOL prediction where operations mayinvolve recovery in, amongst otherenvironmental factors, high levels ofturbulence to new naval vessels. Primaryfuture SAIF objective is to use the simulationto minimise the time and cost required for firstof class sea trials for ships operating Merlinand Wildcat air vehicles.

This report updates the comparative testsmade in the Manned Flight Simulator (MFS) inthe US and Merlin Trainer Simulator in theUK designed to evaluate simulator uses tosupport Ship Helicopter Operating Limits(SHOL) analysis for manned and unmannedrotorcraft. The purpose of conducting thesetests independently is two-fold. First, dynamicinterface activities are defined as they apply toSHOL tests and aircraft/ship dynamic interfaceexpertise and analysis. Second, to provide aplatform to test devices like the LPD softwareto demonstrate the aid; for example, to signalthe initiation of helicopter launch andrecovery. The objective is to recover theaircraft aboard a moving vessel withinreasonable safety margins regardless of theseaway. The report details the technicalresults of simulated launch and recoveryevents using full motion simulators which iscompared to the same events at sea. Thisreport assesses aircraft and deck availabilityimprovements by using the Energy Index (EI)to signal the top of recovery. Percent ofimprovement for operational availability isdemonstrated. Preliminary discussions on howthe results were validated at-sea complete the

4

article.

HELICOPTER-SHIP LAUNCH AND RECOVERY

DYNAMIC INTERFACE

Dynamic Interface is defined as the study ofthe relationship between an air vehicle and amoving platform. It is performed to reducerisks and maximize operational flexibility[Healey, 1982]. Globally, DI is concerned withthe effects that one free body has in respect toanother. Historically, this means the effectsthat a ship may have on a recovering orlaunching air vehicle. However, recent studieshave concluded that the same principles applyto other motion related activities, such as, theboarding of Landing Craft vessels or LCACsinto the wells of Amphibious Warfare Ships,the docking of submarines or the launching ofunsophisticated missiles.

Dynamic Interface is divided into two broadcategories: experimental or at-seameasurement and analysis, and analyticalwhich is concerned with mathematicalanalysis and solution [Ferrier, B. &Semenza, J., 1990]. The methods are notmutually exclusive. Neither method alonecan produce a comprehensive and timelysolution of the DI problem.

The traditional approach is experimentalDI. It investigates operational launch andrecovery of vehicles, engaging anddisengaging o f rotors , vert ica lreplenishment and helicopter in-flightrefueling envelopes. "Shipboard suitabilitytesting" assesses the adequacy,effectiveness, and safety of shipboardaviation. Testing methodologies andprocedures have been standardized bylaboratories, such as NAWCAD (PatuxentRiver) assisted by NSWC (Carderock), andQinetiq (Boscombe Down). Whileexperimental testing has numerousobjectives, the primary focus is on launchand recovery envelope development andexpansion.

United States NavyThe procedures for launching and recovering

various helicopter types aboard a ship can bedifferentiated on whether the helicopterhandling system is present and used (RapidSecuring Device (RSD) and whether thelaunch or recovery will occur in day VisualMeteorological Conditions (VMC), night VMC,or in Instrument Meteorological Conditions(IMC) with Night Vision Device (NVD) launchand recoveries being a subset of VMC launchand recoveries. Ships containing helicopterrecovery systems have three deckconfigurations for recovering a helicopteraboard a small deck: clear deck, free deck; andRecovery Assist (RA).

A clear deck landing means the RSD is stowedand the helicopter will be landing without theuse of the RSD or RA systems. A free decklanding means the RSD is positioned in thelanding area, the pilot flies the helicopter to aposition above the RSD, and lands thehelicopter, placing the RA probe into the RSDwhich is then used to secure the helicopter tothe deck of the ship (Figure 1). Lastly, toexecute an RA landing, the pilot flies to andestablishes a hover over the flight deck. Fromthe aircraft, the RA probe is then lowered tothe flight deck via a messenger cable. Flightdeck personnel attach the RA cable to the RAprobe, which is then reeled back up to theaircraft and secured. Once the RA cable issecured to the aircraft, the ship establisheshover tension (850 to 2000 lbs) onto the RAsystem which stabilizes the aircraft over theRSD. When the pilot is ready to land, the shipselects maximum tension (4000 lbs), whichhauls the aircraft down into the RSD. TheRSD then clamps around the RA probe,securing the aircraft to the flight deck.

Figure 1 – SH-60 and RSD Trap

The distinguishing characteristic between an

5

approach flown to a small deck in day VMC,night VMC, or day and night IMC is theapproach profile flown by the aircraft. A dayVMC approach (Figure 2) is commenced fromat least 1.5 nm behind the ship at not less than200 ft above ground level (AGL) atapproximately 80 Knots Indicated Airspeed(KIAS) heading along the ship’s base recoverycourse (BRC).

The pilot then flies toward the stern of theship, aligning his approach path with theship’s lineup line meeting a series of altitudeand range gates that terminates with theaircraft at one-quarter nm and 125 ft AGLastern the ship with a closure rate suitable forthe given conditions. The closure rate (thedifference between aircraft ground speed andthe ship’s speed of advance) depends onseveral factors such as sea state, ship motionand visibility. During a night VMC approach,the pilot flies to a point at least 1.2 nm behindthe ship at 400 ft AGL at 80 KIAS aligned withthe ship’s BRC (figure 2).

The pilot then flies to intercept the stabilizedglideslope indicator (SGSI) green/amberinterface and maintains glideslope to arrivebehind the stern of the ship at a suitableclosure rate (figure 3). An IMC approach isflown much like a Ground ControlledApproach (GCA) at any suitably equippedairfield. A controller aboard the ship willguide the pilot to a point astern of the ship byproviding heading and altitude commands. Unlike an approach to an airfield, however,the pilot slows his rate of closure as theaircraft approaches the stern of the ship.

Figure 2 – Small Deck Normal NightApproach Profile

Figure 3 – Stabilized Glideslope Beam

Regardless of the initial approach profileflown, the final phase of the approach tolanding aboard ship is flown purely usingvisual cues.

Royal Navy into wind port approachThe Royal Navy conducts two distinctapproach profiles for landing aboard a vesselunderway; forward facing (from both port andstarboard) and into-wind. As the into-windapproach has many similarities to the standardUSN flight profile, this paragraph willconcentrate on the forward facing profile, inparticular port approach. The aim of theforward facing port approach is to fly aconstant angle approach to a point slightlybehind and above the flight deck (figure 4).Particularly at night, the profile is commencedfrom a ‘gate’ position _ nm astern the vessel onthe Red-165 (Port side 165° from ship’s head)at 125 ft and 60 kts groundspeed (min 40 KIASto max 80 KIAS).

Figure 4 – Port Approach ForwardFacing Landing

From the gate position a controlled approachis conducted with a progressive descent anddeceleration to arrive at a point slightlybehind and above the flight deck in a slowhover taxi; at night the approach angle issupplemented by a 3° Glide Path Indicator.The aircraft is then taxied forward the last 10yards, on the same heading as the ship’sheading, while ensuring there isapproximately one rotor span lateralclearance between the rotor disc and the flight

165°

_ nm, 125ft,

60kts G/S

Bum-line

6

deck. The aircraft is thus brought to the hoverwith the pilot sitting abeam the bum-line, onerotor span laterally and 10-15 ft verticallydisplaced from the flight deck (Figure!5). Thebum-line is used as a reference to prevent anyfore and aft drift with respect to the flightdeck. Once in the hover, and if possible duringthe approach, the deck motion is assessed todetermine the frequency and severity ofmotion. This allows the pilot the opportunityof predicting a suitable quiescent period forlanding. If a suitable quiescent period can notbe forecast, the ships course and speed may bealtered to provide more favorable flightoperation conditions.

Figure 5 – Merlin on T45 Destroyer

Once it is assessed that a quiescent period isapproaching, the aircraft is moved laterallyalong the bum-line until in a hover over thecenter of the flight deck at 10-15 ft. Theaircraft is then descended vertically, with nodrift, aiming for a firm, but not heavy, landing.

Energy Index Measurement/Metric

In order for the helicopter to operate to thedeck without a deck officer, it needs to knowwhat the ship is doing now and what it will bedoing during the final descent to touchdown.The Ship Motion Forward Prediction Federate,based on the LPD EI algorithm, is designed toidentify quiescent periods of ship motionsuitable for the recovery of the helicopter. Thealgorithm uses real-time information on whatthe ship is doing, permitting a computation onwhat it may be doing in the very near future. The LPD essentially performs the function ofan experienced LSO, but without the guess-work.

The EI algorithm was integrated into the HLA

using a software wrapper. This wrapperenables the LPD unit to exchange data withthe other modeling components. The aircraftlimits, which form part of the initializationdata used during HLA start up, are expressedas the ship's EI. The EI value is correlated tothe level of kinetic and potential energycontained in the ship. The ship can onlydisplace from a very low energy state to anaircraft out-of-limit condition by theintroduction of a certain quantity of energyfrom the sea. When the index is low the shipis stable and the ship motion is small. Whenthe index value is below the high-riskthreshold, the landing deck motion isacceptable for aircraft recovery.

The thresholds of the various energy levelsare directly based on the combination of shipcharacteristics (measured) and aircraftlimitations (defined). A limit is defined by theimpact that a certain ship motion conditionmay impose on the structural integrity ordynamic response of a given helicopter. Thesum of these limits produces a red line that isdrawn on the EI scale for a given ship (seeFigure 6).

Figure 6 – Deck Status and Risetime

The time required for the deck motion to risefrom minimal motion (or very safe deck) tounacceptable motion is called the risetime. Interms of the EI scale, the risetime is defined asthe period of time that is measured from theend of a green signal to the positive side of thered line. This is given as (T3 – T1) as shown onFigure 6. The risetime is a thumb printcharacteristic of the ship’s response andremains fairly constant for each ship class.

7

The very safe deck is a special condition inwhich there is insufficient energy in theaircraft-ship system to raise the deck out oflimit for some defined time period or risetime,and it is this concept that was used in thesimulators to indicate that the deck was safe toinitiate landing.

By employing deck quiescence as the metricfor aircraft recovery, deck limits expressed asa static value become redundant. The EI LPDsoftware (LPDLOOP or NAV11) is used toassess deck energies as a function of themechanical and dynamic limits of an airvehicle. Quiescent periods are identified bywhich an operator or computer may signal thepilot or UAV AFCS to descend. The recordedor computed motion time histories containingthe corresponding EI results, were evaluated. Each green deck point was analyzed. Theoretically, any green deck point couldserve for the initiation of recovery. Assumingthat the rise time for a given vessel remainsconstant, the aircraft descends within the risetime value and the aircraft is assured arecovery within the deck motion limits. Essentially, the methodology summarizedabove is a formulation to quantify operatingbeyond the static deck limits as defined inNATOPS or SHOL. Figure 7 displaysgraphically beyond static limit operations. The base envelope is taken from 10 knots ofship speed. As before, any points within thehour-glass structure are conditions withinlimits and contain no appreciable probabilityof out-of-limit deck motions. Outside of thestructure contains motions which areconsidered by static reference as out-of-limit.

Figure 7 – Safe Motion Operations

A schematic representation of how the LPD isused for UAV autorecovery to signal the onsetof the descent (and of deck quiescence) isdisplayed in Figure 8. Throughout theapproach and initial hover (M1), LPD ismonitoring the deck. It is only at the M0

position (low hover) that the UAV autolandsystem would accept a Green Deck signalfrom the LPD. The recovery can occur at anypoint within the green zone as indicated by the“signal to the top” arrow on the EI trace. In afully functional autoland program, should thevehicle be in a descent at an unsafe deckpoint, a signal would be sent to the LandingAlgorithm Federate to stop or abort therecovery[7].

Figure 8- LPD Application in NIREUS

The position of M0 is not fixed but input as thepart of the initialization data for the HLAfederation.

Simulated Dynamic Interface System

To house the dynamic interface programs,existing flight simulators are used withexternal federate models. These areintroduced to provide ship and environmentfunctionality such as real time representationof ship motion and the air wake flow field.Each external federate function can then beintroduced and run on a remote computer,separate from the core flight simulator. A keyobjective is to provide a system capable ofconducting SHOL assessments during shipdevelopment and prior to sea trials. It is

8

envisaged that a cost-effective combination ofsimulation and first-of-class flight trials at seawill maximize the operating envelope for thevarious new ship platforms from which amanned or unmanned rotorcraft is intended tooperate. Real-life flight trials are expensiveoperations and are also limited by theprevailing weather conditions available for theduration of the test period.

Whether a simulation represents anunmanned or manned system, the systemmust be capable of accurately responding in avariety of environmental conditions. Theeasiest is to evaluate the device in a closed andcontrolled environment. The primarydifference between a manned and unmannedsystem revolves around pilot drivencommands and controls. The pilot isrepresented in the simulated UAV system by aseries of flight laws and mission commands.The primary elements of the imagined UAVsystem are generalized: UAV, Data Link,Tactical Command Station (TCS), PortableComputer Control Station (PCCS), andtraverser and landing grid, and an automaticrecovery system. Each of these systems arefederates along with the simulatedenvironment which were also composed offederates.

A typical HLA design defines 6 separatefederates (Figure 9), connected via the HLARun-time Infrastructure (RTI) software. Thestructure is applicable to either a manned orunmanned scenario.

Figure 9 – Federation Architecture

The initial software package was designed touse a full motion simulator (in the mannedcase) and a TCS (in the UAV case), to estimatesystem effectiveness as a function of simulatedship motion, visual environment and syntheticoperational systems, and to compare theresults to related analytic data [2].

By discipline the Federation is reduced tofigure 10.

Figure 10 – HLA Federation by Discipline

As displayed in Figure 11, the UAV TacticalControl Station federate is integrated to thefederation though gateways at the InertialNavigation, Tracking Sensors and theoperator control station. When the airvehicle is hovering in the appropriateposition for recovery, the EI signals theonset of quiescence and through the uplinksets the air vehicle on its descent to thedeck.

9

Figure 11 – Autoland Concept

Figure 12 displays the basic TCS Monitorgraphical user interface (GUI) which displaysthe EI and colour land command.

Figure 12 – UAV Deck Monitor GUI

In the manned version, the test bed at RNASCuldrose (which mirrors that of the US MFS)is essentially an HLA simulator. The MerlinSimulator Facility is located in a purpose built28,000 m3 building at Royal Naval Air StationCuldrose. The facility comprises a CockpitDynamic Simulator (CDS), 3 Rear CrewTrainers (RCT), 6 Part Task Trainers (PTT),Computer Based Training (CBT) classrooms, aMechanical Systems Trainer (MST) and aWeapon Systems Trainer (WST).

The CDS offers a full motion simulator, whichis an exact copy of the cockpit of the aircraft.Its state of the art graphics allow a veryrealistic training environment for aircrews. Figure 13 displays the external view of thesimulator.

Figure 13 – MERLIN Simulator

The Pilot’s view from within the simulator isshown in Figure 14.

Figure 14 - Pilot’s View

TEST OBJECTIVES

Focusing on the ship motion characterizationaspect of aircraft/deck interface study using acommon measured metric, several tests areconducted in the simulator which are laterrepeated by the actual devices at-sea.

The indicator for success was the pilot’s abilityto safely and repeatedly recover the aircraft inthe range of desired conditions, such that thedeck lock could be engaged. Pilot/operatorflight evolutions were consistent with currentflight patterns. Evolutions were programmedfor day and night and under progressivelydifficult deck conditions. In addition to the

10

objectives indicated earlier, particularattention was made to recovery times and deckmotion envelope limits.

Primary Testing Objectives and Conditions

a - Day and night and underprogressively diff icult deckconditions.

b- Programmed deck SHOL by aircraft

c- Standard Circuit: First Circuit LPDoff. Thereafter: LPD ON, LPD OFFfirst day then night, same order. The pilot rated workload anddescribed task cue.

In the manned helicopter scenario, the LPDpilot display is attached to the upper starboardside of the hangar (figure 15). It is in plainview over the flight deck and in full view fromhover. From this location on the starboard sideof the ship, the indicator is visible duringeither stern approach (USA) or the port-approach (UK) and hover. If the SH-60 issimulated in its positive pitch-up attitude, theindicator light visual might be at the limit ofthe field of view.

Figure 15 – USN MFS Ship with LPD Display

The LPD calculates the EI and broadcasts itover the simulator visualization system. Depending on the value of the EI, theappropriate symbol is illuminated on the LPDindicator visual box. Figure 16 displays thelight on the actual ship.

Figure 16 – The Light Indicator on DDG 88

In the UAV scenario, such as the NIREUSprogram (Figure 17), no external indicator isrequired. External view images wereprogrammed to visually demonstrate theinitiation of recovery which was correctly andrepeatedly identified by the LPD federate inthe autorecovery system.

Figure 17 – Simulated UAV Auto Landing

Figure 18 displays several of the manyplatforms programmed to receive the NATOgeneric VTOL UAV created by simulation.

11

Figure 18 – NATO Fleet Programmed

SIMULATED TEST SUMMARY

-UAV

Figure 19 shows an example of a simulation.

0

2

4

6

8

10

12

0 50 100 150 200 250 300

time seconds

UAV

Figure 19- Example UAV simulation

The UAV status trace on Figure 20 representsthe position of the UAV; where 1 is the UAVapproaching M1, 2 is hover at position M1, 3 isgiven as the transition to M0, 4 is the hover atM0, 5 is the final descent and 6 is touchdown. The aborts are given by 6 (back to M1) or 8(back to M0).

.

0

2

4

6

8

10

12

0 10 20 30 40 50

time seconds

red deck

hold

UAVlanding

descent

Figure 20- Sample UAV Recovery

In the NIREUS implementation of LPD, theautoland simulation ignores the LPD deckstatus until the UAV arrives at hold positionM0. LPD is then interrogated looking for thefirst LPD green deck. The decision to wait forLPD green deck ensures that the UAV descentcan be safely achieved. Once the LPD greendeck is acquired, the UAV descends to thedeck. During decent, LPD continues to monitorthe ship motion and will signal to the UAV toabort should an unusual ship displacementoccur that takes the deck out of limits.

Figure 20 shows the UAV in transition toposition M0 (where the green trace is equal to3). It holds briefly at M0 and then descends tothe deck. On recovery there may have been abounce indicated by a sudden sharp rise andthen definitively indicates recovery by thevalue 6. Closer inspection of the recoveryperiod showed that, as soon as the UAVarrived at the M0 wait position, the EI wasshowing green deck. In this case the autolandsystem operated well. The HLA appeared tohave passed the green deck signal; the UAVdescended over a 5-second period to the deck,which was within limits for recovery.

-Manned Simulation compared to recordeddata

Several test pilots were involved in both theUS and UK simulations. One of the US testpilots was also selected to conduct the LPDevaluation aboard USS PREBLE (DDG 88)which followed the last US simulated DI test. Between the two programs, hundreds of

Out-of-limit

6

12

evolutions were conducted in conditions withvarious relative wave direction and waveheight. Winds were kept between 10 to 30knots vectored in the direction of the relativewave angles (winds are computed as aconstant force). Ship speed was maintainedmostly at 10 knots, while some testing wasconducted at 20 knots. The visibility waseither day or night with several scenariosconducted during rain or snowstorms.

Simulation flights focused on the test matrix.As most aspects of the flight and shipcharacteristics are cross-referenced, it wasrelatively easy to develop tendencies andcause and effect principles during the courseof the test. The three primary study graphicsare presented in Figures 21 - 23.

Flight 1 – 10150909LPDoff day

Figure 21 - Day

Figure 21 shows a time history, with LPD off,the recovery event occurred on a green-amberor safe deck, and the launch occurred fromquiescent deck. T h e correspondingtranslational traces showed similardisplacements at launch. Oleo compression(figure 22) appeared normal with the tailwheel striking firmly first, but with highengine torque measured at several pointsthrough the evolution (figure 23). This mightbe attributed to pilot adjustment to simulatorflight operations.

Figure 22 – Force on Wheels

Figure 23 – Engine Torque

Figure 24 (UK example) LPD on, is composedof the launch and recovery events, EI, ship’sroll, pitch and yaw traces along with the deckenergy levels.

Figure 24 - Day

Oleo compression trace appeared to show apeak compression on the nose gear (Figure25).

13

Figure 25 – Oleo Compression Trace

Figure 26 – Day Operations

Figure 26 time history, with LPD off, therecovery event occurred on a quiescent deck,but the launch happened from a high amberor caution deck. The deck was very nearlyout-of-limit. The corresponding translationaltraces showed similar large displacements atlaunch. Oleo compression appeared normal,but high engine torque was measured atlaunch (figure 27). Figure 28 comparesboarding times with and without the LPDindicator. The figure also compares pilotresponses in the MFS and the MerlinSimulator. From the figure, boarding timesare improved, particularly at night, with theLPD illuminated.

Figure 27 – Engine Torque

Figure 28 – Boarding Times

COMPARED TEST SUMMARY

DECK RECOVERY - MANNED

The LPD was applied as a visual landing aidand operated as a federate. The MannedFlight Simulator was modified to implement afederated operation allowing individualsimulation components to be replaced with aminimum of change to the other components. Among the issues analyzed was thefundamental question as to whether or not theLPD could be used to improve launch andrecovery activities. The answer to thatquestion would manifest itself in the recordeddata and would be supported by pilotcommentary and observations.

As mentioned earlier, one of the key factorsrelated to increased operational capability inlanding helicopters aboard ship is the abilityto repeatedly launch and recover safely from aship moving in response to the seaway. The

Nose wheel

14

successful repetition of the same event raisesthe overall confidence in conducting thelaunch and recovery evolution. One of theobjectives in using the LPD is to recover on aquiescent or near quiescent deck, regardless ofthe condition of the seaway. The primaryobjective is to assess operational improvementas a function of environmental conditions, withand without LPD. The metric of success is thechoice of recovery with LPD on quiescent ornear quiescent deck which equates to aminimum of ship motion. The data for thismetric is recorded and displayed.

Figure 29 displays the average distribution, bypercentage, of LPD status during launch andrecovery events. The distribution (markedreal-world) represents the combined results ofthe participating pilots. The diagram comparesthe at-sea result with the equivalent simulatedsum. The two cases are similar but thesimulated case contained a greater number oflaunch and recovery events. This may be dueto the simulated environment encouragingtaking greater risk on the part of the pilot oroperator. There were no red deck eventsrecorded while LPD was ON. The nighttimeevents with LPD-ON appear approximatelythe same between test environments.

Figure 29 – Distribution of LPD Status

Figure 30 displays the launch and recoveryevents during a particular session of mannedoperations. As in the tendencies recorded inthe simulator and at sea, launch and recoveryevents with the LPD switched on occur inlower ship motion while launch and recoveryevents with the LPD off shows near random

results. The chart displays the correspondingtable of launch and recovery events for theentire day session. With respect to the percentdistribution of deck energies, 44% of theattempts occurred with LPD off from a greenquiescent deck whilst 92% were to/from agreen deck with the LPD on. Green-amber orsafe deck accounted for 40% of cases with theLPD off, while only 8 % of the remainingevents with LPD on were to a green-amberdeck. About 16% of the attempts occurred toan amber deck with the LPD off. There wereno amber events recorded with the LPD on. The session did not record launch and recoveryevents to a red deck.

Figure 30- L and R Event Summary

Another key factor related to increasedoperational capability in landing helicoptersonboard ship, is the ability to repeatedlylaunch and recover safely from a ship movingin response to the seaway. One of theobjectives in using the LPD is to rapidly butsafely recover to a quiescent or near quiescentdeck, regardless of the condition of theseaway.

Figure 31 is divided into Day and Nightoperations, with and without LPD for thelaunch and land events. Referring to the Dayportion of the graphic, with LPD off, it tooklonger to maneuver the aircraft and for thepilot to achieve a landing solution than withthe LPD on. Referring to the night portion ofthe graphic, with LPD off, the same tendenciesare exacerbated at night. The improvedrecovery times are attributed to improvedconfidence on the part of the pilots making thelanding decision. The quicker recovery time ofnight evolutions to day evolutions is attributed

15

to the availability of fewer cues for the pilot toachieve a landing solution. The deck statusconditions were also recorded and studiedduring the simulated DI.

Figure 31- Boarding Times to the Deck

DECK RECOVERY - UNMANNED

A test of concept was performed during theUSN Joint Project Office MAVUS project usingthe Bombardier, Inc CL-327 co-axial UAV. The project culminated with an at-seaTECHEVAL ending in 2003. The programfeatured a number of “firsts” including anautoland proof-of-concept using the EIapproach. Figure 32 displays an early MAVUStest using the CL-227 version UAV on an FFG7 class vessel. The time history graphics in thecentre of the figure displays simulated motionsalong with the EI computation indicating tothe ground control station (later the TCS)when to command UAV descent. The airvehicle tended to hover at about 3 metres overa Recovery Assist, Securing, and Traversing(RAST) track controlled custom made grid. Onland signal, the air vehicle descended at about1 metre/second, landing on the grid in lessthan 4 seconds from the low hover position. This is well within the rise time minimum ofthe FFG 8 ship class.

Figure 32- CL-327 x FFG39 MAVUS TrialHere aircraft stability at touchdown on or nearthe grid in real-time is calculated using shipmotion as a function of the aircraft model. The

aircraft model is considered an extension ofthe ship. The aircraft experiences shiptransferred forces and moments, which createrectilinear and angular accelerations on theair vehicle. The accelerations can benumerically integrated to determine theposition and attitude of the helicopter relativeto the ship as function of time for various shipmotions. In essence, the aircraft motion is theresult of the sum of all forces to which it isexposed. This is the inspiration to use the EItoday, to measure and predict deck motion tocomplete launch and recovery events. Figure33 displays EI based measures from a recenttest of the MQ-8B Fire Scout aboard USSMCINERNEY (FFG!8).

Figure 33 – Motion Characterization

The Quad chart shown in Figure 33 contains atime history trace containing rise and fall timeevents along with the corresponding ship pitchand roll traces. For an eventual autolandsystem to function, an autoland commandwould be sent to the air vehicle during hoverin an appropriate designated position over thedeck. Assuming a descent rate similar to othermaritime helicopters, the aircraft touchesdown well within the rise time of the ship. Still referring to Figure 33, the lower leftcorner of the Quad chart displays a typical 24hour period of ship motion recordings showingthe distribution of deck energies per hourrecording, and the hours in which flightoperations occurred.

As in the manned version of the test, a keyfactor related to increased operational

16

capability in landing VTOL or fixed wingaircraft onboard ship is the ability torepeatedly launch and recover safely from aship moving in response to the seaway. Theoperational benefits of an unmanned airvehicle are increased if it can be safelycaptured autonomously, i.e. without the aid ofan experienced LSO to a quiescent ship. Theaircraft will still need to be flown to the deck,but the computer resolves the difficultassessment of quiescent point identification.

CONCLUSIONS

The primary goal for conducting dynamicinterface analysis is to expand existingoperating envelopes and increase air vehicleavailability thereby improving overall navaleffectiveness. The objective of dynamicinterface study is to determine the maximumsafe air vehicle/ship platform operationallimitations. Given an air/ship system andinherent operational limitations, DI strives toincrease tactical flexibility for any set ofenvironmental conditions. Modeling andsimulation is used to rapidly delineate systemlimitations. The calculated system limitationsprovide experimental DI with the necessarydata to more effectively set testing strategy toprobe the limiting conditions.

The paper described the development of asimulation that functions through a HLAFederat ion creat ing a reasonablerepresentation of real world operations. Thisis achieved within a controlled environmentpermitting greater opportunity to evaluate acandidate system well before the system isbrought to sea. Initial at-sea testing for bothmanned and unmanned air vehicles shows afavourable tendency to reflect predictionsmade by simulated computations. Whilstthere remain some improvements to be made,the demonstrations have been, thusfar,successful.

In the development of this study, an overviewof the ship motion and dynamic interfacesimulations and modeling has been describedwith the emphasis on undercarriageencountered forces and air vehicle responsestability. Validation of the results is a prioritybecause of the potential problems affecting

ship-helicopter operating deck limits to beprogrammed for air vehicle automaticrecovery. Beyond the basic problem of dataverification and validation, the analyticprocedure demonstrated above may be used tocross-correlate between proposed aircraft-shipdeck limits and the vehicle expected physicalresponses.

While the focus of the report was on airvehicle final approach and recovery, deckissues significant to air vehicles after recoveryinclude chock and chain, aircraft on deckmanipulation, handling and servicing.

REFERENCES

[1] S. White, R. Reading (2001). NATO/PfP HLAFederation of VTOL Operations SupportingSimulation Based Acquisition. 01E-SIW-032.Brussels

[2] I. Cox, G. Turner, J. Duncan (2005). Applying aNetworked Architecture to the Merlin HelicopterSimulator. Royal Aeronautical Society Conference.London.

[3] Bokus, G. and Ferrier, B (1995). Development,Integration and Test of an automatic UAVRecovery Sytem for the CL227 VTOL-UAV.American Helicopter Society. Ft. Worth.

[4] P. Crossland, B. Ferrier, T R. Applebee (2004).Providing Decisiion Making Support to ReduceOpeator Workload for Shipboard Air Operations.Royal Institute of Naval Architects. London.

[5] I. Woodrow, D. Spilling, A. McCallum (2002).The Interoperable Simulation of Air Vehicles andShip Air Wakes within a Multinational SimulationFramework. 02E-SIW-047. Bussels.

[6] Ferrier, B (1997). “Étude Analytique d’Interfacedynamique aéronef-navire.” Projet de l’Indicateurd’appontage. Thèse de doctorat . ÉcolePolytechnique de Montréal. Montréal. Montréal.

[7] Lumsden, R Bruce (2001). Helicopter/Ship DeckLanding Guidance Systems. Royal AeronauticalSociety. London.

[8] Ferrier, B (2006). “Safe Expanded ShipOperations by Quiescent Deck Recognition”. Scan

17

Eagle x Type 23 Deck Availability. UK060417-1.Bristol.

ACKNOWLEDGEMENTS

This article is dedicated to the loving memory ofDr. Peter J. F. O’Reilly, a founding father of thedynamic interface discipline, and one of the firstpeople to propose the use of the Energy Index t oautomatically recover Unmanned Air Vehicles. Theauthors gratefully acknowledge the contributionsmade to this article by CDR Jeffrey R. Von Hor,USN (ret) and LCDR Stephen Crockatt, RNboth formally Instructors at the US Naval Test PilotSchool.

BIOGRAPHIES

Dr. Bernard Ferrier isEngineering Head of the ONRAircraft- Ship DynamicIn ter face Program atHoffman Engineering Corp. The DI Office Programincludes the design and

manufacture of the Landing Period Designator,assembly and conduct of simulation programsrelated to dynamic interface focusing on theassessment of a wide variety of air vehicles, shipboard handling systems and ship classes. Prior tojoining Hoffman and BMT Syntek TechnologiesFerrier led the Anteon Corporation’s (Analysis &Technology now General Dynamics InformationTechnology) Dynamic Interface Program for thelast six years covering a wide variety of UAV, USV,and manned- ship projects. Prior to joiningAnteon, Ferrier led the CL227 interface program atBombardier, Inc (aka Canadair) in Montréal,Québec Canada and Arlington, Virginia. Prior tojoining Bombardier, Ferrier was a rotor dynamistand project leader of the dynamic interface projectof the AH-64 at the McDonnell Douglas HelicopterCompany (now Boeing) in Culver City (California)and Mesa (Arizona). He received his last doctoratein helicopter/ship interface engineering at the ÉcolePolytechnique de Montréal (Québec) Canada

Dr. John Duncan is ProgramManager of the UK MODDefence Equipment &Support, Directorate Safetyand Engineering, Sea SystemGroup’s Simulation Based

Acquisition. The Group’s focus is on ship boardmaritime technology development and modellingtechniques used for the interface of manned an

unmanned air and sea vehicles. DR. JOHN M.DUNCAN is the chairman of the NATO NavalGroup 6 Sub-Group 61 on virtual ships. He waspreviously chairman of the NATO Specialist Teamon Simulation Based Design and VirtualPrototyping for ship acquisition. He led the ST-SBDVP development of Allied Naval EngineeringPublication 61 on ship virtual prototyping. He isalso leading application of long-haul distributedsimulation to address systems interoperabilityrequirements for the CVF, next generation UKaircraft carrier. Dr. Duncan received his Ph.Dfrom Durham University, Durham (UK).

John Nelson is a rotarywing ship sui tabi l i tyengineer assigned at theN a v a l A i r SystemsCommand at PatuxentRiver, MD. He has 12 yearsof experience testing fixed

and rotary wing aircraft aboard ship. His projectsinclude EA-6B and F/A-18E/F carrier suitabilitytests and shipboard tests with numerous helicoptersand the MV-22B tiltrotor. He was lead engineer forthe USS SAN ANTONIO (LPD 17) First of Class testprogram. He is currently assigned to the MQ-8BFire Scout program and conducted the first MQ-8Bshipboard flight tests. He received his B.S. fromWorcester Polytechnic Institute and his M.S. fromSyracuse University.

Dean Carico is a senior aerospace engineer in theRotary Wing Ship SuitabilityBranch in the Integrated SystemsEvaluation, Experimentation, and Test (ISEET) Department at theNaval Air Systems Command atPatuxent River, MD. He has over39 years experience working onNavy and Army flight test and

related analytic/simulations programs. Dean wasthe lead engineer in the Navy’s first rotorcraftoperational flight trainer (OFT) evaluation for SH-2F Device 2F106. As the Rotary Wing Aircraft TestDirectorate Simulation Specialist during the lateeighties, Dean was lead engineer in a multi-yearprogram on Rotorcraft Simulation to SupportFlight Testing. As the first Dynamic InterfaceSection Head, Dean initiated a combined flight testand analytic program in 1983. Dean was lead inan internal science and technology program thatfocused on determining the effect of math modelcomplexity in analyzing the rotorcraft shipboardlanding task. He is currently the Navy lead in a

18

high performance computing program that focuseson developing analytical options to improve flighttest performance, stability and control. He hasalso generated several small business innovativeresearch programs that focus on enhancingrotorcraft land and ship-based flight testing. Deanhas masters’ degrees in Aerospace Engineeringfrom Princeton and in Engineering Science fromthe Navy Postgraduate School, and is anengineering graduate from the USNTPS. Hereceived the Meritorious Civilian Service Award fortesting in a combat zone in 1973, and the RichardL. Wernecke Award for technical excellence inrotorcraft test and evaluation in 1997. He receivedthe SFTE Director’s award in Sep 2003.

David J. Ludwig is an aircrafttechnologies program officer atthe Office of Naval Researchworking a variety of UnmannedAir Vehicle and Rotary WingAircraft technology developmentprograms. Mr Ludwig began hiscareer at the Naval Air Test

Center, Patuxent River, MD. Early in his career, heserved as an aerial refueling (AR) technicalspecialist conducting a variety of AR developmentaltests of the KC-10 and KC-135 tanker systems, aerialrefueling pods, and the S-3 tanker system. He was alead mechanical systems engineer for the V-22Integrated Test Team during the EMD program andthe H-1 Upgrades program. He served as Maritimeand Rotary Wing Mechanical Systems andPropulsion Branch Head within the Integrated Testand Evaluation Department at Naval Air SystemsCommand. Mr Ludwig earned his Bachelor ofScience degree in Mechanical Engineering from theUniversity of Maryland and is a 1992 graduate ofthe U.S Naval Test Pilot School. Mr Ludwig iscurrently pursuing a Masters of Science degree inAerospace Engineering at the Florida Institute ofTechnology.

19