1. Introduction
Military operations such as Desert Storm are not successful by accident. Leaders and soldiers spend years training for such contingencies. This training process increasingly includes time working with simulations and simulators. The purpose of these is to place soldiers in situations which replicate those they will experience in actual combat, to stress their reactive and decision making capabilities, and to give them the opportunity to make mistakes where the consequences are not lethal.
The military defines any training that is not real combat to be simulation. This has lead to a division of simulations into three broad categories: Live, Virtual, and Constructive. The movie "Top Gun" focused on live simulation, soldiers operate their real equipment in mock engagements. Ground forces participate in similar maneuvers, armored and infantry forces don laser gear and engage a threat force in non-lethal combat. These exercises allow combat troops to experience the rigors of living and working in the field and force them to fight against a well trained reactive opponent.
Virtual Simulation is similar to live except that the equipment is replaced with simulated mockups and the field of battle is generated by a computer. In these simulators soldiers practice many of the same tactics but at a much lower operational cost and with greater freedom in taking risks. These tanks do not burn fuel, break down, or destroy terrain, thus reducing much of the expense involved in Live Simulation. Operators are free to pull the triggers of their weapons and see the results of simulated tank rounds impacting a target. They may attempt to ford rivers, destroy building, and generally engage in behavior that is too dangerous to attempt in live simulation. Since the battlefield is sculpted by the computer it can be made to look exactly like that on which the trainee expects to operate. Live simulations, on the other hand, are limited to the terrain that can be found naturally at training sites.
Constructive Simulation, also known as wargaming, derives its name from the fact that the pieces operating on the battlefield are not individual tanks and aircraft but a construction of many different types of equipment into a single aggregated unit like an armor company, artillery battery, etc. (figure 1). The first two types of simulations are used to train individuals operating equipment, this equipment is in turn controlled by leaders in command posts who see the battle in a more abstract form. Constructive simulations allow these commanders to face situations and make decisions under the stress of time and limited resources just as they will during actual combat. In a classroom situation it is easy to present students with problems in which the answers have been determined and solving them is a homework assignment. Constructive simulations immerse these commanders in a situation where the enemy is highly trained, experienced, and just as determined to win the war. This enemy is unpredictable and does not always operate as the books say he should. Here soldiers discover whether the tactics they have been taught really work, here they develop confidence in their ability to operate as a team and win wars.
2. Constructive Simulation
Simulations have typically been developed by a sponsor with an interest in a specific part of the combat environment. As a result separate models exist which portray ground, air, sea, intelligence, electronic warfare, and logistics operations. These originally operated separately but are beginning to be integrated to provide a joint operational environment in which all of the military branches support each other. Some of the prime simulations of these areas are:
Corps Battle Simulation (CBS) - Developed by the NASA Jet Propulsion Laboratory, this simulation focuses on ground combat as practiced by the Army. It includes models of terrain, movement, direct engagement, artillery fire, helicopter strikes, special forces infiltration, and to a lesser extent aircraft operations, intelligence, and logistics.
Air Warfare Simulation (AWSIM) - Developed at the NATO Warrior Preparation Center in Germany, this simulation represents Air Force operations. It includes fixed- wing air strikes, air-to-air combat, surface-to-air missile defense, air base operations (refueling, runway capacity, and aircraft maintenance), and electronic warfare.
Tactical Simulation (TACSIM) - Developed for the Army, TACSIM collects intelligence on enemy forces and provides the information to military units responsible for analyzing it to determine what the enemy is doing and planning.
Combat Service Support Training Simulation System (CSSTSS) - Developed by the Army to replicate the operations that support the combat troops. The models include resupply of ammunition, fuel, and food; clearing the battlefield of the dead and wounded; medical operations; and providing transport for moving assets into and out of combat situations.
Joint Electronic Warfare / Electronic Combat Simulation (JECEWSI) - This model replicates the operations of electronic assets such as jammers and radars on the battlefield. Since radars are used extensively to collect information about the enemy and provide early warning of air strikes, the use of radar jammers can have a significant impact on the outcome of the battle.
Enhanced Naval War Gaming Simulation (ENWGS) - This simulates the operations of naval battle groups. These may engage each other or may project their firepower into a land-based fight. Models include ship maneuvers, submarine activities, aircraft operations, and electronic warfare.
Marine Air Ground Task Force Tactical Warfare Simulation (MTWS) - This simulates Marines operating independently or as part of a task force. The models include amphibious ship-to-shore operations, ground and air combat, engineering, and intelligence collection.
As we saw in Desert Storm, all of the military branches support one another in opposing the enemy. Since simulations must replicate this reality, it must be possible to operate all types of simulated units in a single training exercise. This could be done by producing a single gigantic simulation that contains everything or by linking the above simulations into a single "virtual simulation" (figure 2). The latter solution is being pursued to allow each service to develop and control its own simulations but still operate together. The Aggregate Level Simulation protocol (ALSP) has been developed to join these into a single distributed solution to the problem. ALSP is not the only attempt to put simulated entities together in a virtual environment. The virtual simulations mentioned earlier are developing the Distributed Interaction Simulation (DIS) project and a Joint Simulation System (JSIMS) is in the works to close the gap between the virtual and constructive worlds.
3. Simulation Personnel
Creating these types of simulations requires a wide variety of people. The actual construction workers on these projects can be broken into seven major categories (figure 3). The project begins and ends with the customer - the military. Everything is put in motion because the military needs and requests a new capability. The first step is to design the simulation to meet the specifications provided by the customer. Since the systems are computer based, the next step is to assemble the pieces that make up the hardware that will host the simulation. Next, the system developers provide the operating system, editors, configuration tools, and other pieces of the development environment. Once this is done the actual simulation developers begin to produce the software that is the heart of the simulation. A finished simulation must then be trained to the people who will operate it. Finally, the operators put the simulation machine into motion to produce the combat environment experienced by the military, who are in training exercises.
4. Simulation Framework
To explore the inner workings of a constructive simulation we have created a basic framework. Experience has shown that most simulations contain these basic pieces regardless of their specific functionality. Figure 4 shows the six basic pieces of a simulation.
Simulation Engine. At the center is the engine that produces the ground combat, air strikes, intelligence collection, or naval maneuvers desired. This contains representations of the terrain, weather, combat units, and equipment characteristics. These are operated on by orders received from the training audience and their opposing forces (OPFOR). Mathematic algorithms then execute the orders to produce the combat events and their outcomes.
Scenario Build. The simulation engine can not operate until a data set containing military equipment descriptions, operating characteristics, terrain, and force placement has been supplied. The data may naturally exist in hundreds of different documents, libraries, and databases in varying formats. Scenario build is responsible for creating this data in a form that the simulation can use. This tool began as an operator using a text editor to build a formatted data file. Human errors prompted the development of menu driven systems which control the formatting of the data and eliminate "fat finger" errors. More advanced systems are being developed in which the data is entered graphically, reducing the amount of time required to prepare for an exercise.
After Action Review. When an exercise is finished, the performance of the units being trained must be measured and the information presented in an understandable format. The AAR tools are responsible for collecting, analyzing, and presenting this information. As computer analysis becomes more prevalent these reviews have grown and are now expected one or two times every day. This allows the exercise sponsors to grasp what is happening and adjust the exercise to maximize the lessons learned by the training audience.
Controllers. The simulation is run by controllers who are responsible for starting, stopping, controlling the rate of time advancement, and maintaining the system. These typically use a graphical user interface which allows a few people to monitor and control a simulation training hundreds of soldiers. They are responsible for ensuring that the system operates efficiently during the scheduled training period. Controllers include the OPFOR commanders who are responsible for fighting against the trainee to ensure that he experiences a truly professional and tenacious foe.
Trainee. The entire point of the simulation is to stimulate the trainee. In some cases these people will be operating a simulation specific user interface designed to allow them to enter orders and operations in a format understandable by the computer. Unfortunately, this creates a simulation artifact. It does not teach the operator to work in the environment he or she will use during actual war, though the decision process may be much the same. As the military becomes more computerized, it becomes more natural for the simulation to invisibly connect to the trainee's organic equipment and stimulate it directly.
Networks. Simulations seldom operate alone. This piece joins multiple simulations and performs the data transformations that are needed to allow independently developed simulations to understand one another.
5. Current Problems
Since defense spending is declining, simulation as a training aid is increasingly important. This interest has resulted in projects to address some of the nagging problems of building and operating simulations. Some of these are shown in figure 5.
Building a scenario is a very time consuming process. As mentioned earlier this has traditionally been done with a simple text editor like "edt" or "eve". Several projects have developed menu based data entry forms, but they have been only moderately successful. The typical method for building a combat scenario is to take an existing data file and edit it to meet the needs of the upcoming exercise. Menu based systems are usually more of a burden in this process than an aid. As a result the most experienced operators have returned to the text editors to work with the data. It typically requires three to four months to coordinate and build the data file. When simulations like CBS, AWSIM, and TACSIM are joined together the scenarios must be built to synchronize with each other, requiring a significant amount of coordination. Graphically oriented scenario builders are needed that allow the user to work on map backgrounds and see data in graphic forms. Another great benefit would be a single data source which could contain all of the information needed by all of the simulations and could then export it in the specific formats needed by each simulation. Such a project is underway and promises significant exercise preparation savings.
Controllers are an overhead expense in operating an exercise. These people must be a jack-of-all-trades: simulation operator, software debugger, database builder, military expert, and network manager. Since such people are not in great supply, and to save costs, new simulations need to operate with a fraction of the current number of controllers. This requires that a simulation be self- sufficient, a database be bug free, and a controller interfaces be intuitive and powerful.
After action reviews (AARs) have grown in importance and popularity over the last few years. As simulations expand, these need to become more powerful and flexible. A single workstation needs to be able to access and analyze all of the data stored on a distributed network of simulation nodes. This analysis then needs to be organized into a format that is statistically accurate and oriented toward a military audience. Automated AAR has just begun to be able to collect and correlate the complex, many stepped causes of combat events. It is no longer enough to show what happened, it must be able to show the many steps that lead up to and caused the event.
As mentioned earlier, the interface with the training audience is increasingly be done via connections to organic computing equipment. The goal is to place commanders in simulated combat without having to change their natural operating environment at all. They should never have to learn about simulation input terminals, modeling nuances, or the limitations of their options caused by the simulation. As the military becomes more computerized this is becoming easier and much more necessary.
The simulation engine must become smarter and more powerful. A great deal of power can be added by networking multiple simulations together to create a more complete and realistic operating environment for the trainees. The most difficult part of these interfaces is that the original simulations were not designed to operate together. As a result, the way in which a military unit is represented in each is different. Making these match so they can operate believably together requires much ingenuity. The military simulation world needs a standard way to represent the most common military objects, be they tanks, companies of tanks, or command posts. Computer Generated Forces (CGF) are also a hot topic right now. The need exists to create forces which can be controlled intelligently by the computer rather than humans. This will reduce the limitations placed on exercises by the availability of experienced human controllers. Finally, there is a growing desire to join constructive simulations to virtual simulators. One benefit of this is that a constructive simulation can operate hundreds of virtual entities with a few controllers. These can then become teammates or targets for the trainees in the virtual simulators. Conversely, the detail and human decision making found in the virtual simulators can add realism to the mathematically structured events in the constructive simulations.
6. Technical Skills
Since this article is intended for an IEEE audience we should explore the place for engineers and computer scientists in the field. Though engineers can be found in all of the levels shown in figure 3, they are most prevalent in the simulation development layer. Figure 6 matches some of the skills used with the major blocks of the simulation framework.
Scenario Builders need to be proficient with human computer interfaces, database structures, and cognitive understanding. The goal is to create a tool that is functionally strong, but that is easy for the operator to grasp and use.
Control requires simulation experience and human computer interfaces. The experience of running a military simulation is invaluable. Events and demands occur that can never be predicted and sometime never met. But, experience will tell the person what types of solutions are available and which of those are acceptable.
AAR requires human computer interfaces, data visualization, statistics, and virtual reality. Statistics will insure that the results of the analysis are sound rather than misleading. The other skills create tools that present information in a form that can be absorbed by the audience. VR is becoming an important form of data presenta- tion. Though this is usually not immersive, it is useful in conveying messages that summary charts can not encapsulate.
Connecting the simulation to the trainee's organic equipment is an electrical engineer and networker's domain. Experience with the military will show what types of connections to expect and what can be done without interfering with the training audience's natural environment.
The simulation section requires physical and cognitive modeling, operations research, probability, knowledge based systems, and military experience. Mathematicians are able to build the models accurately, but they can not know what needs to be built without the expertise of a military person.
Networking multiple simulations together may be similar to connecting to the trainee's equipment, or it may involve creating a parallel or distributed computer application. The structure of the data being exchanged is important in order to convey the most information with the least amount of network traffic.
7. Conclusion
The future of military simulation appears to be very bright. Since this type of training is an order of magnitude less expensive that live simulation, it should be a natural growth area in a shrinking defense environment. It allows the United States to train a force that is ready to perform their mission efficiently, which is an absolute necessity when the size of the force is being decreased.
Current hot-beds of activity in military simulation include Orlando, Florida; Northern Virginia; Los Angeles, California; and Denver, Colorado.
References:
Allen, Thomas B., War Games, Berkley Books, New York, NY, 1987.
Corps Battle Simulation: Executive Overview, Jet Propulsion Laboratory, Pasadena, California, 1993.
The DIS Vision: A Map to the Future of Distributed Simulation, Institute for Simulation and Training, University of Central Florida, Orlando, January, 1993.
Somerville, Robert M. S., The Military Frontier, Time-Life Books, Alexandria, Virginia, 1988.
Singley, George T., "Distributed Interactive Simulation - A Preview", Army Research, Development, and Acquisition Bulletin, March-April, 1993.
Smith, Roger D., "Analytical Computer Simulation of a Complete Battlefield Environment", Simulation, 1992.
Smith, Roger D., editor, Proceedings of the 1994 Electronic Conference on Constructive Training Simulation, May 1994. Elecsim94
About the Author:
ROGER SMITH is a Principal Simulation Engineer with Mystech Associates. He is responsible for developing simulations and tools to support the training missions of US and Allied forces. These have included air and ground combat models, intelligence collection and analysis algorithms, after action review systems, and system management tools. He has an M.S. in Statistics from Texas Tech University and a B.S. in Applied Mathematics from the University of Southern Colorado.