Detecting intelligent agent behavior with environment abstraction in complex air combat systems

Author(s):  
S. Mittal ◽  
M. J. Doyle ◽  
E. Watz
2009 ◽  
Vol 8 (5) ◽  
pp. 668-677 ◽  
Author(s):  
Steven K.C. Lo ◽  
Huan-Chao Keh ◽  
Yi-Hung Lin ◽  
Wang Jo-Chi

Author(s):  
Michael W. Floyd ◽  
Justin Karneeb ◽  
Philip Moore ◽  
David W. Aha

We describe the Tactical Battle Manager (TBM), an intelligent agent that uses several integrated artificial intelligence techniques to control an autonomous unmanned aerial vehicle in simulated beyond-visual-range (BVR) air combat scenarios. The TBM incorporates goal reasoning, automated planning, opponent behavior recognition, state prediction, and discrepancy detection to operate in a real-time, dynamic, uncertain, and adversarial environment. We describe evidence from our empirical study that the TBM significantly outperforms an expert-scripted agent in BVR scenarios. We also report the results of an ablation study which indicates that all components of our agent architecture are needed to maximize mission performance.


Kybernetes ◽  
2001 ◽  
Vol 30 (2) ◽  
pp. 166-178 ◽  
Author(s):  
David G. Schwartz ◽  
Dov Te’eni

2021 ◽  
Vol 6 (2) ◽  
pp. 32-38
Author(s):  
Hanny Haryanto ◽  
Ardiawan Bagus Harisa ◽  
Indra Gamayanto

Game replayability is very important in serious game to maximize the understanding for the learning content. The replayability is the result from the gameplay experience. Games have the advantage of providing a fun experience, and immersion is a vital element in game design to produce the experience. However, the design of immersion in games is often not well conceptualized so that it does not produce the expected experience. This study uses Appreciative Learning based reward system, which focuses on positive things such as peak achievements, opportunities, exploration of potential and optimism for the future. The reward activity consists of four stages, namely Discovery, Dream, Design and Destiny. Reward personalization is done by regulating reward behavior using artificial intelligence which runs in all four stages. Appreciative Learning will be used to design immersive experiences consisting of sensory, imaginary and challenge-based immersion, which are the three main elements of immersive games. Intelligent agent behavior is modeled using the Finite State Machine. This study produces an immersive reward design that is applied to the concept of Appreciative Learning in designing a serious game.


2019 ◽  
Vol 2019 (4) ◽  
pp. 7-22
Author(s):  
Georges Bridel ◽  
Zdobyslaw Goraj ◽  
Lukasz Kiszkowiak ◽  
Jean-Georges Brévot ◽  
Jean-Pierre Devaux ◽  
...  

Abstract Advanced jet training still relies on old concepts and solutions that are no longer efficient when considering the current and forthcoming changes in air combat. The cost of those old solutions to develop and maintain combat pilot skills are important, adding even more constraints to the training limitations. The requirement of having a trainer aircraft able to perform also light combat aircraft operational mission is adding unnecessary complexity and cost without any real operational advantages to air combat mission training. Thanks to emerging technologies, the JANUS project will study the feasibility of a brand-new concept of agile manoeuvrable training aircraft and an integrated training system, able to provide a live, virtual and constructive environment. The JANUS concept is based on a lightweight, low-cost, high energy aircraft associated to a ground based Integrated Training System providing simulated and emulated signals, simulated and real opponents, combined with real-time feedback on pilot’s physiological characteristics: traditionally embedded sensors are replaced with emulated signals, simulated opponents are proposed to the pilot, enabling out of sight engagement. JANUS is also providing new cost effective and more realistic solutions for “Red air aircraft” missions, organised in so-called “Aggressor Squadrons”.


Sign in / Sign up

Export Citation Format

Share Document