&lt;title&gt;Autonomous agents as air combat simulation adversaries&lt;/title&gt;

Gunsch, Gregg H.; Dyer, Douglas E.; Gerken, Mark; Merkle, Laurence D.; Whelan, Mike

doi:10.1117/12.141754

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our piece of the puzzle is to implement the basic capabilities for autonomous aircraft agents which can operate in stand-alone or distributed simulations. The current capabilities of our prototype MAXIM, although modest, were developed as a vehicle for a variety of research thrusts [2]. In this simulation environment the autonomous agents must maintain a real-time pace in the midst of network communication delays, information losses, computer resource competition, and the uncertainties inherent in interacting with humans.…”

Section: The Simulation Environmentmentioning

confidence: 99%

On applying machine learning to develop air combat simulation agents

Gunsch

Mezera

Gordon

1993 4th Annual Conference on AI, Simulation and Planning in High Autonomy Systems

View full text Add to dashboard Cite

This paper describes several approaches for utilizing machine learning technologies towards improving the capabilities of autonomous, simulation-based agents.For an autonomous agent to be robust, it must be able t o plan its activities, react quickly t o unforeseen events, and execute planned or modified behaviors to achieve goals. W e have begun t o develop autonomous agents which exhibit appropriate behaviors for simulated air combat, providing intelligent, realistic adversaries and cooperative allies. Building such agents is not trivial, and the techniqves of machine learning hold great promise for extending the capabilities of hand-coded systems. In this paper we describe the application of some of these techniques, past successes, and current research directions.

show abstract

Section: The Simulation Environmentmentioning

confidence: 99%