Academic journal article Military Review

Intelligent Agents in the Command Post

Academic journal article Military Review

Intelligent Agents in the Command Post

Article excerpt

A COMMANDER'S ABILITY to make good decisions under adverse conditions can decide a battery. Seventy-two hours into an operation, he may be running on catnaps and caffeine.

At the tactical operations center, the operations officer briefs three possible courses of action (COAs) for a new mission. The operations officer recommends the first COA with conviction, then explains its advantages. Something about the COA is troubling the commander. Perhaps fatigue, stress or some other distraction is preventing him from recalling the lesson learned from other battlefields or training events that would affect his decision. The commander makes his decision but has a feeling that there is a better way--he needs more time, more information or a clearer head.

The decisions commanders, staff officers and warfighters make are driven by education, training, experience and personal preference. A lifetime of learning allows them to set goals, assess conditions, identify and evaluate alternatives, and make quick, complex decisions. The human mind can make remarkable decisions under extraordinary conditions-an ability technology lacks. However, the mind is subject to adverse effects by environmental factors such as fatigue, stress and hunger. It is well documented that decisions made under these conditions are generally inferior to those made by well-rested, fed, comfortable and relaxed leaders. The US military strives to select commanders based on their proven ability to make good decisions under adverse conditions.

The opening scenario provides a good example of decision making under adverse conditions. Consider the same adverse conditions-with another decision-making tool for commanders.

A computerized intelligent agent is programmed with lessons learned from classrooms and training. The software runs on the same computers that are used to type memos and create briefing slides. It accesses the data in the battle command systems and planning tools, including powerful networked computer systems, handwritten notes and sketches.

The intelligent agent is oblivious to time, temperature and a commander's physical and mental condition. Based on all available information, it evaluates the COAs and lists each strength, weakness and issue in a matter of seconds. The commander quickly scans the list, discards some COAs and considers others until he reaches the element critical to the decision. Based on his judgment, his staff s recommendations and a few key factors that he might otherwise not have had the energy or awareness to identify, he decides.

The strengths, weaknesses or issues might be based on planning considerations learned by a lieutenant in basic course; discovered by a captain during an after-action review; or noted during senior service college. They might be based on new enemy tactics observed during a battle yesterday. The intelligent agent combines the things a computer does best-sorting and sifting through massive data while remaining immune to fatigue, stress or environmental factors-with the things a human can do best, such as learning from experience. The agent combines doctrine and tactics with lessons learned throughout military history. It does not replace the commander or make the commander's decisions; it provides concise, relevant and explainable information the commander can consider when making decisions. This is how we can use intelligent agents in the command post of the future.

Learning Agents Laboratory

Automated decision support systems, expert systems and intelligent agents are not new but have played a limited role in military command and control or support systems. Even with today's rapid growth in computing power and connectivity, most software products that claim to be intelligent do not solve complex, real-world problems. Government, industrial and academic researchers have made recent progress in moving intelligent systems from hype to reality.

A novel approach to creating and using intelligent agents to solve complex military problems is under way at the George Mason University (GMU) Learning Agents Laboratory (LALAB). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.