Geospecific Simulation

MGT 2008 Volume: 6 Issue: 4 (July/August)

GEOSPECIFIC SIMULATION



Geospatial Data is Becoming a Cornerstone of
DoD Training and Mission Preparation.

by Marty Kauchak, MGT Correspondent

Geospatial technology is becoming a major element in the training devices and systems that give U.S. warfighters their competitive edge on the contemporary battlefield. While shortfalls prevent the wider and more effective use of this resource in training and mission preparation systems, help is on the way through the collaborative efforts of government- industry teams.

Geospatial data is used in an ever-increasing number and variety of front-line training devices and systems. This resource enables CAE’s MH-60K Medallion S and Flight Safety International’s AC-130U Vital 9 image generators (IGs) to provide visual displays that permit SOF aircrews to complete high fidelity training in those models’ full mission simulators, for example.

Raydon’s Mine-Resistant Ambush Protected Vehicle Virtual Trainer helps train basic and advanced skills for route clearance operations, including the proper response to a variety of IEDs and explosive hazards. Practical exercises set in geospecific areas of Iraq enhance this training. Another impressive example of how geospatial data were enlisted to support training audiences is USSOCOM’s special operations forces planning, rehearsal and execution preparation (SOFPREP) program.

SOFPREP serves as the command’s “intelligence focal point for the gathering of required GEOINT source data and the production of the USSOCOM Common Database in support of the SOF Mission and Training Preparation Systems (MTPS),” said Earl Miller, spokesperson, SOFPREP.

MTPS allows SOF to train, plan and rehearse their mission prior to mission start-up. The program acquires and consolidates elevation, feature, maps, imagery and intelligence source data required for database production; validates the geospatial accuracy and certifies the use of the intelligence data in completed databases and datasets; and serves as USSOCOM’s GEOINT and 3-D scene visualization database and data set archive.

A number of technology challenges and shortfalls prevent the wider and more effective use of geospatial data in training-related applications, however. The main challenges in having quality geospatial data for mission preparation and training are in acquiring high-resolution terrain elevation data, color imagery and vector feature data, Miller noted. “We spend a lot of time and effort in developing high-resolution elevation data, colorizing imagery and digitizing and conflating the vector feature data to correlate with the imagery.”

Time was also among the shortfalls eyed by one of Miller’s industry counterparts. “Turnaround time is too long—current time from collection to product generation can be anywhere from 30 to 90 days,” observed Ron Krakower, director, business development, Sarnoff Corp.

Warfighters are in a better position for missions if they are using current data, particularly in urban environments where geospatialrelated data changes on a regular basis, Krakower said, adding that there are no automated processes—most, if not all, techniques used today to generate ortho-images and 3-D models require manual processes. “This prevents large-size areas from ever being created—it is just too cumbersome.”

SCENE GENERATOR

Service-industry teams have a number of products and research and development efforts in progress that will address these and other shortfalls.

MetaVRs’ flagship Virtual Reality Scene Generator (VRSG) allows diverse DoD training audiences to prepare for missions. VRSG is a PCbased IG that provides realtime geospecific simulation coupled with game-quality graphics.

In real time, VRSG can render geocentric, roundearth, edge-blended terrain that can be visualized in a multi-channel environment with dynamic moving models, and special effects with near real-time terrain updates from multiple sensor platforms, explained Torsten Berger, senior software engineer, MetaVR. “With unconstrained freedom of movement, our customers can fly through, or drive on, 3-D terrain with a 200 km far horizon while maintaining a rate of 60 frames per second.”

VRSG is used for applications such as joint urban operations (JUO) and other ground-level training, such as the Warrior Skills Trainer convoy simulation trainer at Fort Hood, Fort Carson and other training sites; unmanned aerial vehicles, such as the Shadow 200 Tactical UAV program; and manned flight simulators, such as the A-10 full mission trainers program and the Air Force Research Lab’s joint terminal air controller virtual trainer dome prototype.

A new product offering is expected to increase VRSG’s capabilities. MetaVR’s Terrain Tools for ArcGIS is said to represent the next generation of the company’s virtual terrain generation technology. “It’s based on the ESRI ArcGIS platform, which provides a great deal of built-in support for nearly any kind of geospatial data out there,” said Berger.

Any data formats that can be brought into ArcGIS can now be used to generate virtual environments for our VRSG image generator using this product. “Since this terrain generation capability is integrated with the tools that most customers are probably already using to manage their geospatial data assets, our Terrain Tools extension enables them to leverage these assets more effectively for the creation of high-fidelity virtual environments,” Berger added.

A particular focus for MetaVR is on JUO. Contemporary, ground-level training systems and aerial training systems are traditionally separated, each operating with virtual environments that were constructed for the specific requirements of their domain.

“Ground-level environments were generally small in geographic coverage, with a greater amount of cultural content such as buildings, trees and other features. Aerial environments were optimized for flying at a high altitude over greater distances, with expansive geographic coverage but little to no ground-level cultural content,” noted Berger.

The challenge of supporting computerbased JUO training is to create a single virtual environment that has the fidelity required by many different platforms in their specific domains while maintaining the performance required by immersive training systems. “We have to make the experience of a dismounted soldier looking at a building from a meter away just as realistic as the view from a UAV looking at the same building from thousands of meters in the air, all within the same virtual environment,” said Berger.

“In constructing such an environment, we must bridge the gap between the super-highresolution ground-level source data (photographs, survey data) and the lower-resolution remotely sensed data (aerial and satellite photography, and digital elevation models) and make the transition between these data types believable.”

COMMAND, CONTROL AND INTELLIGENCE

DoD’s appetite for geospatial data for modeling and simulation (M&S) is also being fueled by including more command, control and intelligence (C2I) missions into unit (collective) training events and exercises. A variety of industry products allow operators to use this resource in C2I systems’ M&S applications.

ESRI’s software platform for C2I applications incorporates a geospatial capability. The company’s core technology for the commercial joint mapping toolkit (CJMTK) program provides a component-based toolkit for developing geospatial capability in C2I. ESRI, in collaboration with prime contractor Northrop Grumman, uses the toolkit for the overarching CJMTK geospatial appliance (CGA).

“The CGA provides a set of all the standard National Geospatial-Intelligence Agency worldwide data sets (raster product format, vector product format, and digital terrain elevation data), along with some interesting commercial data such as Natural Vue 2000 and the NASA 90-meter worldwide data elevation model,” pointed out Bill Harp, defense business development for ESRI.

These data sets, along with server technology to create maps, globes and Web services on demand, are all installed in either a rackmounted server or ruggedized tactical server suitable for immediate field deployment. “What this means is that units involved in modeling and simulation will have unprecedented, user-friendly, easy to access worldwide geospatial data at global, strategic, tactical and urban scales to support training and simulation exercises as well as actual operations,” added Harp.

MAK’s VR-Forces product is a computer- generated forces (CGF) application and software developer’s toolkit that enables customers to populate the synthetic environment with intelligent friendly and enemy forces, create missions, run scenarios and complete other capabilities. While VR-Forces provides both 2-D and 3-D visualization of the terrain for scenario creation and simulation execution, the company is eyeing a next generation of products for truer real-time performance.

Warren Katz, company co-founder and a six-term chairman of the Simulation Interoperability Standards Organization, described the capabilities of the current generation of CGF products. “Most current CGF applications utilize one of the common simulation terrain database representations such as OpenFlight, compact terrain databases and others. These database formats, having been derived from the classic visual simulation tool chain, are not generally correlated with the geospatial data used by C2 and analysis systems. This lack of correlation can produce serious anomalies in the simulation results.”

Katz’s observation that geospatial databases are not optimized for real-time performance introduces the concept of off-line production. “Because of this, there will be degradation in speed of the simulation in comparison to current real-time optimized database formats. Streaming of just-in-time geospatial data will also prove a challenge as compared to off-line production and execution of static optimized databases,” he said.

Off-line production is exactly the defect that MAK’s embryonic Geospatially Enabled Modeling and Simulation (GEMS) seeks to eliminate. GEMS is a set of functional components that link M&S with broader command and control applications. One element is the GEMS-enabled VR-Forces, developed under contract to the Army Topographic Engineering Center.

“This is a terrain processing layer inside VR-Forces that simulates on top of the native geospatial data, not a derived visual format,” explained Katz. “As VR-Forces is simulating on top of the exact same geospatial database that is used in the corresponding C2 system, the anomalies due to a lack of correlation are eliminated.”

A functional prototype of the GEMSenabled VR-Forces is currently available for demonstration to MAK potential customers interested in becoming early adopters. “We’re in the process of finalizing our VR-Forces roadmap for 2009 and are working on a firm release date for the GEMS features,” said Katz.

3-D SITE MODELS

Sarnoff offers the MapIt! software solution, which automatically generates accurate, wide-area reference images and 3-D building models by leveraging available imagery and light detection and ranging data. “These ortho-mosaics and 3-D site models provide increased situational awareness and meet the operational needs of warfighters,” explained Krakower.

MapIt! supplies geospatial data for DoD mission rehearsal system needs that include situational awareness, bomb damage assessment, collateral damage assessment, route and urban reconnaissance, and force protection. While the product’s beta form is being used in DoD operational programs, version 1.0 is expected to be fielded during fourth quarter of this year.

Solid Terrain Modeling has a product portfolio that is built on a principle recognized by generations of ground warriors. “Sand tables have been a staple for military training and mission rehearsal since the invention of sand. There is something about having a single stable reference terrain that allows groups of people to focus on the mission rather than how they should interpret a flat map,” said Lawrence Faulkner, president, Solid Terrain Modeling.

The company makes physical terrain models from digital geographic information system data, complete with full color aerial photography printed on the surface. These models are used for all aspects of mission planning and training, from learning to read topographic maps to forward observation, and from artillery control to daily mission brief/debrief in active combat zones.

“The models are computer numeric control-cut from a solid foam block according to the digital elevation model, then printed with map graphics, aerial imagery or a combination of both,” explained Faulkner. “The model gives everyone a bird’s-eye view of the area of interest and supports multiple conversations simultaneously, unlike a computer-generated view which forces everyone to look at the scene from the same position and angle. The models may be of any size and virtually any scale for which data is available.”

Solid Terrain Modeling’s defense customers include the 101st Airborne Division, 25th Infantry Division, 1st Marines, U.S. Military Academy, Defense Advanced Research Projects Agency and NATO. ESRI’s efforts to use M&S and geospatial data to better allow operators to train as they will operate with existing C2I systems are supported through advanced R&D projects with MAK Technologies and other partners.

“One thing that we are actively pursuing in geospatial R&D that will have a significant impact on modeling and simulation is the value of having a common battle management language (BML),” observed Harp. Interoperability, including BML, between battle command applications and modeling and simulation software was a key focus topic at an October 2007 ESRI-sponsored conference at George Mason University’s C4I Center. (See MGT, Volume 6, Issue 1, page 8.)

“We think that the evolving capability of BML represents a key technology that can potentially unite the disparate systems involved in battle management. Particularly, BML has the potential to create realistic train as you fight modeling and simulation capability. We are also interested in how the geospatial component of BML, or geoBML, will evolve as a subset and in parallel with BML,” concluded Harp. ♦