Learning of Intelligent Controllers for Autonomous Unmanned Combat Aerial Vehicles by Genetic Cascading Fuzzy Methods

Ernest, Nicholas; Cohen, Kelly; Schumacher, Corey; Casbeer, David W.

doi:10.4271/2014-01-2174

Cited by 15 publications

(12 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An 80% success rate in 20 trials of each mission was deemed acceptable for an amount of SDMs. This success rate was deemed acceptable here as, unlike prior work, the addition of EWAR stations brings much more uncertainty, especially in these unrealistically long training missions [8]. A higher success rate is required for nontraining missions.…”

Section: Training Mission Setupmentioning

confidence: 92%

“…The basic concepts of GAs and FISs are assumed and will not be discussed. Though reviewed in prior work, included is an overview of genetic fuzzy systems [8].…”

Section: Methodsmentioning

confidence: 99%

“…The GFT created in this work, the Learning Enhanced Tactical Handling Algorithm (LETHA), has evolved from a genetic cascading fuzzy system [8]. The Hoplological Autonomous Defend and Engage Simulation (HADES) is an efficient, low-fidelity simulation used by LETHA to train the fuzzy tree, though the resulting controllers have no need of a low-fidelity environment and can be adaptable to more realistic simulators [8].…”

Section: Problem Statementmentioning

confidence: 99%

“…The Hoplological Autonomous Defend and Engage Simulation (HADES) is an efficient, low-fidelity simulation used by LETHA to train the fuzzy tree, though the resulting controllers have no need of a low-fidelity environment and can be adaptable to more realistic simulators [8]. All models utilized within this study are unitless and have been created through conversation with Air Force Research Laboratory in an effort to create a difficult control problem with some level of realism.…”

Section: Problem Statementmentioning

confidence: 99%

“…Prior work has taught some important lessons with regards to the training of a GFT in this environment [8]. Each MF in every FIS should be tested, but extremely long missions that test the same scenarios multiple times are unnecessary.…”

Section: Trainingmentioning

confidence: 99%

See 4 more Smart Citations

Genetic Fuzzy Trees and their Application Towards Autonomous Training and Control of a Squadron of Unmanned Combat Aerial Vehicles

et al. 2015

Self Cite

View full text Add to dashboard Cite

This study introduces the technique of Genetic Fuzzy Trees (GFTs) through novel application to an air combat control problem of an autonomous squadron of Unmanned Combat Aerial Vehicles (UCAVs) equipped with next-generation defensive systems. GFTs are a natural evolution to Genetic Fuzzy Systems, in which multiple cascading fuzzy systems are optimized by genetic methods. In this problem a team of UCAV's must traverse through a battle space and counter enemy threats, utilize imperfect systems, cope with uncertainty, and successfully destroy critical targets. Enemy threats take the form of Air Interceptors (AIs), Surface to Air Missile (SAM) sites, and Electronic WARfare (EWAR) stations. Simultaneous training and tuning a multitude of Fuzzy Inference Systems (FISs), with varying degrees of connectivity, is performed through the use of an optimized Genetic Algorithm (GA). The GFT presented in this study, the Learning Enhanced Tactical Handling Algorithm (LETHA), is able to create controllers with the presence of deep learning, resilience to uncertainties, and adaptability to changing scenarios. These resulting deterministic fuzzy controllers are easily understandable by operators, are of very high performance and efficiency, and are consistently capable of completing new and different missions not trained for.

show abstract

Section: Training Mission Setupmentioning

confidence: 92%

“…The basic concepts of GAs and FISs are assumed and will not be discussed. Though reviewed in prior work, included is an overview of genetic fuzzy systems [8].…”

Section: Methodsmentioning

confidence: 99%