Automated collection management systems require a geo-spatial query, analysis and display capability to reach optimized solutions. Raster data sets such as topography and land-cover data must be flexibly integrated with vector data sets such as roads and hydrography. The system must accommodate moving objects such as aircraft, ships and ground vehicles as platforms and as targets. Remote sensing and GIS technologies provide the tools for integrating these data types in a common geo-spatial framework. ArcView, ArcInfo and ERDAS Imagine were used to demonstrate the application of these capabilities to collection management. Scripts, models and AMLs were developed to link the three packages to database and visualization software.

Introduction

Optimizing data collection from multiple sensors in a battlefield environment requires a consideration of multiple spatial data types. Automated collection management systems require the ability to assess target, threat and platform positions relative to terrain features. The objective of applying GIS technology to collection management is to provide tools for analyzing and displaying spatial information. GIS provides a common geo-spatial framework for fusing multiple information themes. .

The functionality of applying GIS for collection management was demonstrated through a simulation exercise. A Geographic Services Module (GSM) was created to provide geo-spatial data and analysis functions for the exercise. The GSM provided four main functions within the collection management simulation. The first function was the Scenario Generation Tool (SGT), an ArcView based utility for entering data to define simulation events. The second function is the Information Needs Tool (INT), an interactive interface for formulating and submitting information requests. The third function is the Run Time GIS (RTG), a queryable moving map display. The fourth and most important function, the Collection Feasibility Analysis (CFA), provides spatial analysis for automated decision making within the collection management system.

Hardware/Software Requirements

Interaction between the simulation software and GSM took place across a distributed network. Direct links were established between the Sybase simulation database and Coryphaeus three dimensional visualization software. All GIS functions were run using Esri and ERDAS software products.

This suite of software tools was chosen because of their wide range of capabilities for vector and raster processing. ArcInfo offers the most robust vector GIS with links to relational databases. Imagine provides tools for image processing in a geo-spatial environment. ArcView serves as the display and query engine, providing tools for simultaneously displaying raster and vector spatial data with tables and graphics.

The simulation hardware included three Silicon Graphics workstations and nine pentium PCs. The GSM was centralized on one of the SGIs, and one PC was equipped with ArcView. MapObjects and Esriis Internet Developeris Kit were used to serve GIS data and functions to other machines used in the simulation.

Data Set Development

A high fidelity simulation requires high resolution spatial data. It is critical, however, that the simulation data set is a realistic representation of data that will be available in an operational environment. The National Imagery and Mapping Agency (NIMA) has established formats and standards for military data sets. Many of these data sets are in interim or prototype forms and are not yet readily available on a global basis. As a result, it was necessary to create a data set which is representative of the types and formats of data that will be available in the future.

For the simulation exercise, a conflict was staged between two notional countries whose boundaries were superimposed over the southwestern U.S. Data sets at two levels of resolution were developed, a 1 km resolution data set covering the region around the battlefield and a 30m resolution data set covering the conflict area (figure 1).

The regional data set was developed with 1 km grid cells, based on data derived from USGS EROS Data Centeris Conterminous U.S. AVHRR Companion Disk and Esriis ArcWorld. The regional data set contained the following layers:

The conflict simulation area encompasses approximately 100,000 km2 of eastern New Mexico. This data set has 30m grid cells. It was derived from Landsat Thematic Mapper imagery, NIMA ARC Raster Digitized Graphics (ADRG), USGS DEM and USGS DLG.

Tasking Requirements

The tasking requirements to implement GIS in the collection management system fell into three general catagories: data set generation, data set conversion and software scripting . Data set generation was split between Imagine and ArcInfo. Imagine was used to create process raster files. The false color composites, vegetation analysis, ADRGs and DEMs were processed as .img files. The vector data sets were handled with ArcInfo. The raster files subsequently converted to GRID files for use with ArcInfo and ArcView. Additional data conversions were needed to make the data compatible with other parts of the simulation system, such as Coryphaeus. Software development and scripting was required to ensure consistent data flows between the GIS and other software components. Functions were implemented in Delphi, C++, AML, Imagine Modeler and Avenue to perform specific functions.

GIS Functionality

GSM provided GIS functionality for collection management. The four main functions, SGT, INT, RTG and CFA used the geo-spatial data set for analysis (figure 2). The functions output results which were incorporated into the simulation software.

Simulated events and actions for military exercises are often "scripted" in advance. Scenarios are adapted from real-world conflicts or detailed study of opposing force doctrine. Traditionally, the geographic elements of the scenario, such as operation areas and force movements, have been defined from paper maps and charts, often 1:250,000 Joint Operation Graphics (JOGs) and smaller scale products. GIS offers a means for automating these elements.

The Scenario Generation Tool (SGT) was written in ArcView and implemented as a series of views. There were separate views for air forces, ground forces and naval forces. The air object view allowed the analyst to input the type of platform, its departure point and time, turn points and altitudes, and its destination. Speeds and waypoint arrival times were calculated interactively, and could be modified. A similar menu was used for naval forces. Scripting ground force deployments were more complex, because routes are constrained by terrain. For some unit types, it is desirable to show them following roads. A combination of raster terrain data and vector road coverages were used as the background. In scripting a ground force movement, the analyst selected a starting point and time, and then highlighted waypoints along the desired route. This information was recorded as shape files in ArcView.

The Information Needs Tool (INT) served as an interactive interface combining Sybase functionality with the geo-spatial information in the GIS layers. The interface provided analysts with a means for identifying and selecting targets. The MapObjects interface allowed analysts to use a point and click interface with pull down menus to enter the locations of the targets. Once a target had been located and selected, the information was transmitted to the collection management system.

The Run Time GIS (RTG) served as the primary display and query interface for two dimensional map information. RTG was created as an ArcView project. The display was fed to a set of large monitors. Simulated position data updates for moving were objects transmitted from the Sybase database to RTG. Moving objects were depicted in an ArcView environment which allowed interactive swapping and updating of the background GIS data layers.

The Collection Feasibility Analysis (CFA) tool was developed to calculate the feasibility of observing a target from a given platform. The tool reasons against threat and environmental data to determine if a target can be observed. Threats such as enemy air defense artillery must be considered as constraint positioning collection platforms. Airborne platforms must stay out of range of enemy air defenses or risk being destroyed. Danger areas were defined by a buffer around each threat, corresponding to the effective range. Flight lines were kept outside the danger areas.

Elevation, vegetation and weather were used to determine visibility for each collection platform. Elevation is the first factor: a target must be visible from the platform. Terrain masking is calculated using modified visibility analysis available within ArcInfo GRID. Targets which fall within the terrain mask are eliminated as not feasible for collection. Vegetation is the second factor: targets which are located within areas of masking vegetation area are also eliminated as not feasible. A time sensitive consideration of weather is the final step: targets within areas with fog, clouds or other conditions of low visibility are excluded from consideration.

The result of spatial analysis is a definition of optimal operational areas for specific platforms and sensors. This informaton was presented as tabular information and as electronic maps. The tabular information was used for automated decision making in the collection management system. The graphics were displayed through a MapObjects interface for visualization.

Conclusions

GIS provides valuable tools for automated collection management. The GIS functions implemented for this study reduced the level of effort for scenario scripting, display and feasibility analysis. The RTG provided a flexible interactive environment for displaying simulation data. The same technology can be used to display real objects in an exercise or operational environment. The CFA demonstrated the contribution of GIS analytical functions to limiting collection areas to those which can be expected to yield results. By automating the collection management process, battlefield analysts can increase the volume and confidence of data collected, improving efficient use of resources while minimizing risks to collection assets.