1

AIAA 2001-2061

VIRTUAL REALITY PARACHUTE SIMULATIONFOR TRAINING AND MISSION REHEARSAL

Jeffrey R. Hogue*R. Wade Allen†

Systems Technology, Inc.Hawthorne, CA

Jerry MacDonald‡Cliff Schmucker**SSK Industries, Inc.

Lebanon, OH

Steve Markham††Valentine Technologies, Ltd.

Hampshire, UK

Arvid Harmsen‡‡Automatisering en Adviesbureau

Stuurboord, The Netherlands

ABSTRACT

This paper describes recent developments in a virtualreality (VR) training device developed to teach parachuteguidance and control from deployment through landing.Parachute training simulation originated over 12 yearsago for training to improve the safety and performance ofsmokejumper and military parachutists. Whileimprovements and developments continue for thesefunctions and are discussed in this paper, progress hasbeen made in applying this concept to a number of otherapplications such as sport jumping , arcade and themepark entertainment, museums, aircrew emergencies , andoperational mission planning and rehearsal includingGPS navigation.

INTRODUCTION

This parachute simulator1 currently allows the jumper todeploy a parachute during free fall, visually check thecanopy for proper deployment2, mitigate malfunctions,and cutaway and deploy a reserve parachute if required,then exert control and guidance through to landing.

* Principal Specialist, Senior AIAA Member† Technical Director, AIAA Member‡ Project Engineer** President,†† Technical Director‡‡ Technical Director

Copyright © 2001 by Systems Technology, Inc. Published by the American Institute of Aeronautics andAstronautics with Permission.

A range of visual data base selections are available for agiven jump, and methods are now being developed toallow rapid preparation of visual and wind field databases for rehearsal of specific mission objectives. Thisprocess for developing mission rehearsal visual and windscenarios will incorporate available digital terrainprofiles, satellite or aerial photographic imagery ofground terrain and weather information. The simulatorhas also been interfaced with a commercial GPSnavigation device designed for parachuting, whichallows training in the use of this equipment for guidanceand navigation. This combined system is described inthis paper.

This parachute simulator has been modified andupgraded to provide various features to meetrequirements of a range of applications and theinterrelating benefits of these enhancements are thesubjects of this paper.

USDA FS FS-14/SF-10A Simulator Developments

The parachute simulator dynamics allow for properrepresentation of response to control inputs and windfields. Modeling the turning, pitching, rolling anddescent response of a given parachute required data onperformance and other information on significantnonlinear behavior that affects control sensitivity andoscillatory response. For example a new USDA ForestService (FS) parachute3, the FS-14 has improvedcharacteristics over the FS-12, including: faster forwardspeed, ability to fly backwards in brakes, slower decentrate, more rapid turn rate, and has been supplied in threedifferent sizes.

The higher performance of this parachute designrequired the development of a new version of the training

2

Figure 2. Wind Visualization Through StreamerSimulation

flight simulator4 for Forest Service. This version wasadopted this year by all Forest Service training centers.It incorporates these improved flight dynamicscharacteristics and a number of enhanced trainingfeatures.

These include a simulator scene developed to model aspecific real-world Forest Service training landing zonewith the difficult landing challenges typical of theirrough terrain fire fighting operations, shown in Figure 1.Scene development for parachute simulation is a difficulttask because due to the flight vehicle’s steep glide slopeangles and the parachutists need to look directly down atthe landing zone below, towards the horizon fornavigation and collision avoidance, and completelyoverhead to assess canopy condition. As a consequence,scenes must include the large overall details required forflight simulations with the small size details encounteredin simulations for ground vehicles, with the furtherrequirement that objects look correct from overhead aswell as from horizontal viewpoint. This wide field ofview is precisely what mandated the use of VRtechnology.

Figure 1. Simulated Montana USDA FS Landing Zone

Invisible 3D features like ambient winds can be adifficult concept to convey to a student. The streamervisualization feature, illustrated in Figure 2, shows a linerepresenting a 3-d view of the actual path taken by astreamer dropped from directly above the selected target.This allows the instructor to demonstrate the effects ofwind change with altitude and alert the student to anypotential problem areas. When this feature is in use, thestudent’s monitor shows the jump scene selected on thestartup options screen with a yellow curving line startingat the initial altitude selected over the jump spot. Thesimulator computes a simulated path that streamer wouldtake when dropped in the particular wind field selected.

Previous post-simulated jump view options wereavailable to show the jump exactly as seen by the jumperor via a remote view of the jumper and jump scene froma view angle that can be moved by a joystick. While thisperspective was particularly useful in viewing the effectsof malfunctions and jumper motions, it and the previousjumper view playback mode presented problems in termsof understanding and critiquing of the navigational andcollision avoidance tactics adopted by the trainee. Forthis reason and at smokejumper request, a jump reviewwas added where the observer’s eyepoint tracks along thewindline above the jumper, and the joystick is not used.The jumper in the previous run is marked with a circle todistinguish it from any jump partner. This observation isuseful for understanding the parachutist’s path over theground and relative to other parachutists.

The Forest Service has developed a new version of theirsimulator-based training syllabus. The large size versionof this parachute has been adopted and has been boughtin large quantity by the US Army Special OperationsForces (SOF) as the SF10A. The SOF has also mandatedtheir own version of the Forest Service training program,including VR parachute simulators.

Arcade Entertainment Developments

A version of the parachute simulator has been adaptedfor arcade entertainment systems. The application,shown in Figure 3, was designed and is manufactured byIllusion Inc. This version is based on the MS Windowsplatform, and uses the OpenGVS rendering engine. (Theearlier DOS versions of the parachute simulator used aproprietary TGE library and file format from Triac Inc.).The Windows graphics library uses the widely adoptedOpen Flight file format originated by Silicon Graphics

3

Figure 4. Simulation scene of MMFS at Yuma AZ

Figure 3. Arcade Version

3D Models Cultural Features

Scenario Topography

MeterologicalDataWind

FieldDynamic Elements

Terrain GeometryCultural FeaturesAtmospheric ConditionsDynamic ElementsTime of Day

Raw Terrain Data(DEM, DTEDImagery, etc.)

TerrainDatabase

GenerationSoftware

ScenarioDatabaseGenerator

Real-TimeJump

Simulation

CommercialModel

Libraries

3D GraphicsModelingSoftware

ImageEditing

Software

ModelLibrary

Terrain-CorrelatedWind FieldGeneration

Figure 5. Process for Generating Mission Rehearsal Simulation Scenes

Inc. The parachute dynamics, head tracker, and jumperinput sensor interface were ported to Windows by STI aspart of this effort. Windows allows more versatilehardware compatibility and the use of a broader range ofprogram and graphics scene generation tools.

Mission Planning and Rehearsal Developments

STI is in the middle of a Phase II Small BusinessInnovative Research contract from the SpecialOperations Command (SOCOM) to develop a version toenable rapid generation of real world scenes based ondigital terrain, photos, and weather data. This processinvolves the porting the current MS-DOS Windowsimplementation with a more modern graphics userinterface (GUI), much better compatibility with modernPC hardware, and readily incorporate networking so thatactual (rather than simply pre-recorded) interactionbetween parachutists will occur. Figure 4 shows two ram-air jumpers jumping above a scene replicating theMilitary Free Fall School (MMFS) area near YumaArizona. Figure 5 shows the process of generating realworld simulator scenes.

4

a) without head tracker b) with head tracker

Figure 6. VR Head Mounted Display (HMD)

c) full gear including oxygen maskand flight gloves

a) HMD worn with elastic brow band b)Tracker attached to helmet

Figure 7. HMD Modified to Wear Under Helmet with Tracker Attached

Aircrew Emergency Training

The process of delivering and installing simulators inquantity for Aviation Physiology and Life Supporttraining to the US Air Force Air Combat, Reserve, AirEducation and Training Commands, Air National Guard,NASA, and the US Navy Naval Operational MedicineInstitute Aviation Survival Training, provided anopportunity for review of this simulation by a widevariety of experienced life support and SERE (Survival,Evasion, Rescue, and Escape) instructors with a numberof suggestions which were added for improved trainingcapabilities5. In particular, scoring criteria were revisedto accept a broader off-wind landing angle to align withaircrew parachute landing (PLF) goals and to allow forthe high reverse landing speeds which may beencountered on the best of landings under the strongwinds which can be encountered during emergencyparachuting.

Additional aircrew training related improvements are indevelopment. These include a VR system designed to beworn with a flight helmet and improved riser force

sensors. Almost all current installations use a VR headmounted display (HMD) which can be worn alone withan elastic band between the ear pieces, or with an trackersubstituted for the elastic band as shown in Figure 6.

Aircrew emergency procedures call for raising the visorand removing the oxygen mask (while wearing flightgloves) during descent under parachute canopy. Sinceaircrew would only experience actual parachuteprocedures and equipment operation as a dire andhopefully extremely infrequent occurrence, it wasperceived as particularly important that trainingexperiences replicate the actual event as closely (thoughsafely) as possible.

Although there have been earlier attempts to use theHMD with the elastic band under a flight helmet andattach the tracker to the rear of the helmet, this had beenunsuccessful due to interference between the HMDearpieces/earphones and the helmet. As a result, theseprocedures were taught prior to removing the flighthelmet and donning the combined VR HMD/tracker for

5

Figure 9. US Navy Aviation Survival Training Scene

Figure 8. Using Personnel Lowering Device

Figure 10. O2, AAD, and Manual Emergency Ripcords

the simulated parachuting experience. However, theaircrew emergency training community expressed astrong desire to address complete training scenario issuesin a single device in a continuous training experience.

In response to these concerns, the HMD has now beenmodified to remove the earpieces and attach the elasticband to the brow portion as illustrated in Figure 7a.Audio is supplied to the normal helmet earpieces, and thetracker is attached to the rear of the helmet with the visorcut away to clear the HMD as illustrated in Figure 7b.

Use of a Personnel Lowering Device (PLD) is required bycertain aircrew emergency procedures. A lowering strapis snapped to the riser straps and then the harnessfasteners are released (Figure 8). Sturdier riser sensorsare being developed which include damping devices towithstand the abrupt shock from this sudden drop andunloading. This capability will remove the need to trainfor this procedure in a separate hanging harness device.

Aircrew emergencies can occur under far more difficultconditions than operational jumps, which have limits onterrain, weather, etc. Obviously these limits do not applyto emergency situations when mishaps can occur overany terrain. Thus, simulators were supplied to the USNavy Aviation Survival Training Centers (and previousUS Air Force Life Support units) which featured weatherconditions such as rain, fog, overcast, and improvednight lighting renditions.

Additional scenes emphasizing over water and coastallocations were also supplied. The Instructor’s screenreminds of the need to coach the trainee below 200 feetaltitude to minimize maneuvering, prepare for parachutelanding fall, and watch the horizon. A typical jump sceneis shown in Figure 9.

Some aircrew ride in aircraft with ejection seats; othersget parachutes but must bailout (egress) in emergencies.As shown in Figure 10, these parachutes are typicallyequipped with emergency manual ripcords (metalhandle), a ripcord to arm an automatic activation device(AAD) with a red knob which then opens the parachutecontainer when the jumper is below a preset altitudewhile in some systems simultaneously exceeding aspecified descent rate, and a ripcord with a green roundknob to release oxygen. Ripcords have been featured inthe operational version of the program for some time andare now included in the emergency aircrew version aswell. The AAD feature has now been added to the sportversion of the program.

6

Figure 11. VR Parachute Simulator Suspension Frame

Figure 12. Horizontal Start Hanging Harness

New Jumper Suspension Frames

Hanging harness training has long been mandated foroperational, and especially for aircrew-emergencyparachute training. The trainee hangs in an actualharness, suspended from above by riser straps, wearingflight suit, gloves, helmet with oxygen mask and visor,etc. Some of these rigs were suspended from a ring-attached overhead to a single point. These existingsystems often had pulleys attached to the ring adjacent tothe risers, and control lines were then run back andattached to the wall behind the jumper. When the jumperpulled on the lines, he was physically rotated in thedirection he pulled. This was seen as advantageous insystems without a simulator, even though the control-motion-to-visual correlation was poor. Earliest versionsof the VR parachute simulator system were installed withexisting hanging harnesses. Attaching the controller boxto the wall or floor modified these systems, and thecontrol lines were routed from the box down through thepulleys to the jumper.

Some initial aircrew training installations used the singlering harness attachment discussed above, but it was clearthat undesirable motions occurred during the simulatedjump. More recently the frame shown in Figure 11 hasbeen developed to give a consistent installation thatstabilizes the trainee, and also allows for installation offour riser strap sensors that are used in conjunction with

malfunction training.. With a VR simulator, the lack ofsynchronization between physical motion cues and visualmotion due to the dynamics of the simulated parachutewould be worsened further by the effects of the physicalmotion on the jumper’s head tracker. When possible, tominimize this problem, horizontal suspension rings wereattached via lines tied to fasteners anchored in adjacentwalls.

The simulator now allows runs to start in free fall at upto 25,000 ft. altitude. A frame enhancement has beendeveloped which attaches to the bottom of the parachuteharness with a standard 3-ring release. For good openingparachutes, one part of a 2 line main ripcord is thenrouted though this release, and the other part is routedthrough the main ripcord sensor. The simulator jumperis pulled up to a horizontal position as shown in Figure12. When the ripcord is pulled, the simulated parachuteopens and the jumper feels himself moving to a verticalposition.

When a high speed malfunction is selected by theinstructor, one part of a 2 line reserve ripcord is thenrouted though the harness pull-up bottom release, and theother part is routed through the reserve ripcord sensor.Thus in this case, when the jumper pulls the mainripcord, he stays physically horizontal, but when he pullsthe reserve, he physically moves from a horizontal tovertical position.

GPS Guidance System Simulator Based Training and Rehearsal

OPANAS (Operational Parachute Navigation System) isa GPS, altimeter, and magnetic compass based guidancesystem originally developed by NAVOCAP S.A. andenhanced by SSK Industries for use by HAHO (HighAltitude High Opening) jumpers. The parachutist viewsposition, altitude, and relative position to the target.

7

Figure 12. On-Target system withOPANAS, MMP, and ParaSim™

MMPComputer OPANAS

ParaSimComputerJumper

TogglesRisers

RipcordsHead Tracker

3D Scene

TimePositionHeading Programming,

Simulated Data:GPS, Altimeter,Magnetometer

Exit PointTarget Location

Winds AloftCanopy Performance

HeadingMoving Map

Figure 13. Data Flow Diagram for the On-Target Training System

Prior to the mission, the guidance system is programmedwith course data derived from target coordinates, forecastwind aloft data and canopy performance information, anddisplays in moving map format a flight path and the real-time position of the jumper.

Mission Management Planner (MMP) is a SSK softwareprogram that runs on a PC computer (typically a laptopwhich can be attached to the guidance system) thatpermits complete HAHO mission planning. Missioninformation is entered into the MMP, including targetcoordinates and elevation, forecast wind data and canopyperformance. An exit area is calculated and displayed onthe mapping program. The MMP is also used to programthe guidance system.

Field reports from HAHO troops had indicated that thelogistics and expense of practicing HAHO jumps limitthe training available. Less than acceptable results areobtained when the jumps are performed in practice oractual insertion activities. STI and SSK addressed thisproblem by developing the On-Target System, whichcombines the OPANAS and MMP with the VR trainingsimulator to enable HAHO troops to fly simulatedmissions using realistic wind, meteorological and terraininformation. Figure 12 shows the components of the On-Target system and Figure 13 shows the data flowbetween them.

The On-Target System permits multiple practice flightsto be completed prior to a practice jump or actualmission. In the case of mission training, the MMP alsoprograms the parachute simulator (ParaSim™). TheMMP is used as a control panel during simulatedtraining missions. MMP allows the instructor to varycertain simulated conditions during mission planning,such as adjusting the winds aloft from the forecastvalues. The instructor can also monitor on the MMP thepath followed by the jumper in real time duringsimulation.

Sport Parachuting

Over the last few years, sport parachutes have evolvedinto very high performance, small size, very high wingloading designs. These canopies require high skill levelsto be flown with any degree of safety, so there is anobvious motivation for the frequent requests for simulatortraining. While these parachutes are sold based onpublished flight data for forward and vertical descentspeeds, the data provided is not necessarily for the sameflight condition; and suspended weight, altitude,temperature, and barometric pressure are not stated butare required to accurately simulate the parachutedynamics. In addition, the simulation needs data forbraking and turning inputs.

The answer to this problem was to develop a flight testaerodynamic data acquisition system specifically toproduce data for simulation modeling. Importantparameters to measure include forward and verticalinertial speeds (together with measured suspendedweight, temperature, barometric pressure, and altitude)at full flight and the full range of brakes, and turn rates,bank angles and vertical acceleration through the fullrange of toggle deflection. Consistent data sets had beenpreviously obtained from Para-Flite, Inc., manufacturerof MC-5/MT-1XX series, and other US militarycanopies. The SSK system described below and shown inFigure 14 is based on instrumentation similar to thatused by Para-Flite to collect performance data for theircanopies.

8

PC Computer(Off-Line)Brauniger

RecordingBarograph

BarometricAltitudeSensor

GPS

TotalAirspeedSensor

HorizontalCamcorder

VerticalCamcorder

Heading Angle

Toggle Positions

In-FlightData

Recorder

3 AxisAccelerometer

Sensors

Toggleand RiserSensors

3 AxisRate GyroSensors

Future Enhancements

Figure 14. Parachute Data Acquisition System

The SSK system measures canopy performance using abarograph package designed for use under paragliders.The barograph, an IQ-Competition model manufacturedby Brauniger Flugelectronic GmbH of Weilheim,Germany, measures time, altitude, rate of descent, andtotal airspeed. Given w, rate of descent (verticalairspeed) and V, total airspeed, simple trigonometrycomputes horizontal airspeed. A GPS is connected to thebarograph to measure position and heading data.

The instrument package calculates altitude by measuringthe pressure difference between it’s location and a pre-setreference pressure, typically the surface pressure or sealevel pressure. Airspeed is measured by a calibratedpropeller sensor suspended beneath the jumper, clear ofairflow interference.

Data are recorded once per second in the instrumentpackage, and up to 50 flights can be recorded. The datais downloaded into a PC through a serial connection anda software program supplied with the barograph. Thesoftware program downloads the raw recorded data andalso prepares graphs of the altitude, airspeed, and verticalspeed.

GPS information sent to the instrument package updatesonce every two seconds, and is not useful as a headingindicator during rapid turns. To measure turn rates, adigital camcorder, pointing horizontally, is mounted tothe jumper, who uses a well-defined landmark (such asa road) for an initial heading and reference line. Thisalso gives bank angle information during turns. Another camcorder, vertically oriented, records toggleand riser inputs. Under development is a system that willinclude a 3-axis accelerometer and gyro system to collectmore detailed canopy performance data.

Conclusions

The availability of a parachute flight simulator has leadto the development of a variety of particular requirementsfor a wide range of applications. These requirementshave been met by a number of program improvementsthat benefit all of these applications.

References

1. Hogue, Jeffrey R., Johnson, Walter A., Allen, R.Wade, A Simulator Solution for the Parachute CanopyControl and Guidance Training Problem, SAE Paper920984, 1992 Society of Automotive EngineersAerospace Atlantic Conference and Exposition, Dayton,Ohio, April 7-10, 1992

2. Hogue, Jeffrey R., Frederick G. Anderson, Cecy A.Pelz, R. Wade Allen, Robert Gates, Steve Markham,Arvid Harmsen, Beyond the Basics: Enhanced ParachuteSimulation Training, presented at the 1999 ParachuteIndustry Association International Symposium, January10-14, 1999, San Diego, California.

3. Pierce, Dave, The FS-14 - An Improved SmokejumperParachute Canopy, OE02P27, March 1995

4. Hogue, Jeffrey R., Johnson, Walter A., Allen, R.Wade, Pierce, Dave, A Smokejumpers' ParachuteManeuvering Training Simulator, AIAA-91-0829,American Institute of Aeronautics and Astronautics 11thAerodynamic Decelerator Systems TechnologyConference, San Diego, California, April 9-11, 1991

5. Hogue, Jeffrey R., Frederick G. Anderson, Cecy A.Pelz, R. Wade Allen, Steve Markham, Arvid Harmsen,Parachute Simulat ion Enhancements forPost-Ejection/Egress Training, 36th Annual SAFESymposium, September 14-16, 1998, Phoenix, AZ.