The ten-year technology horizon is dominated by developments in the digital domain

he battlespace is increasingly dynamic and challenging. The volume of information available to decision-makers is ever increasing. The uncertainties about the battlespace, whether due to cyber attack, miscommunication, mechanical fault, or other problem, prevents rapid, effective, informed decision-making. The mere application of so-called “big data” technologies cannot address the scope of the coming problems. In an environment with great uncertainty and large amounts of data, autonomous intelligent agents on the battlefield can provide the military planner with an invaluable aid for observing, planning, integrating, and interpreting to achieve mission goals. However, current intelligent systems cannot provide the required decision-making or uncertainty reduction support. The state of the art for automated intelligent behavior relies heavily on the use of predefined scripts, which prevents both adaptability in the intelligent agent response and support for human intent. The prerequisite for successfully creating autonomous intelligent battlefield agents are the Artificial Intelligence (AI) technologies for machine learning, user intent, prediction of actions, and emergent behavior. Improvements across these areas of AI can provide autonomous intelligent behavior in battlefield agents that goes beyond scripted behavior and is far more useful. Simulation can be used to address requirements development and testing of battlefield intelligent agents. Simulation can be used to promote integration of battlefield intelligent agents into military operations at all scales across all the levels of command. In the paper we discuss battlefield intelligent agents and simulation’s role in bringing about their next generation. The paper introduces the technical challenges obstructing battlefield intelligent agents, the challenges of situation awareness, and the AI shortfalls to be addressed. The paper discusses the roles that simulation should play in order to research, develop, and integrate battlefield intelligent agent technologies into operational practice. Finally, the paper discusses the use of simulation technologies to assemble battlefield intelligent agents. T SIMULATION FOR RESEARCH, DEVELOPMENT, AND TEST OF BATTLEFIELD INTELLIGENT AGENTS Sheila B. Banks, Ph.D. Martin R. Stytz, Ph.D. Calculated Insight Georgetown, Calculated Insight Orlando, Fl 32828 Washington, DC (407) 353-0566 (407) 497-4407, (703) 338-2997 [email protected] [email protected] , [email protected] ABSTRACT he battlespace is increasingly dynamic and challenging. The volume of information available to decision-makers is ever increasing. The uncertainties about the battlespace, whether due to cyber attack, miscommunication, mechanical fault, or other problem, prevents rapid, effective, informed decision-making. The mere application of so-called “big data” technologies cannot address the scope of the coming problems. In an environment with great uncertainty and large amounts of data, autonomous intelligent agents on the battlefield can provide the military planner with an invaluable aid for observing, planning, integrating, andhe battlespace is increasingly dynamic and challenging. The volume of information available to decision-makers is ever increasing. The uncertainties about the battlespace, whether due to cyber attack, miscommunication, mechanical fault, or other problem, prevents rapid, effective, informed decision-making. The mere application of so-called “big data” technologies cannot address the scope of the coming problems. In an environment with great uncertainty and large amounts of data, autonomous intelligent agents on the battlefield can provide the military planner with an invaluable aid for observing, planning, integrating, and interpreting to achieve mission goals. However, current intelligent systems cannot provide the required decision-making or uncertainty reduction support. The state of the art for automated intelligent behavior relies heavily on the use of predefined scripts, which prevents both adaptability in the intelligent agent response and support for human intent. The prerequisite for successfully creating autonomous intelligent battlefield agents are the Artificial Intelligence (AI) technologies for machine learning, user intent, prediction of actions, and emergent behavior. Improvements across these areas of AI can provide autonomous intelligent behavior in battlefield agents that goes beyond scripted behavior and is far more useful. Simulation can be used to address requirements development and testing of battlefield intelligent agents. Simulation can be used to promote integration of battlefield intelligent agents into military operations at all scales across all the levels of command. In the paper we discuss battlefield intelligent agents and simulation’s role in bringing about their next generation. The paper introduces the technical challenges obstructing battlefield intelligent agents, the challenges of situation awareness, and the AI shortfalls to be addressed. The paper discusses the roles that simulation should play in order to research, develop, and integrate battlefield intelligent agent technologies into operational practice. Finally, the paper discusses the use of simulation technologies to assemble battlefield intelligent agents.