Tailoring a Cognitive Model for Situation Awareness using Machine Learning

Richard Koopmanschap, Mark Hoogendoorn, Jan Joris Roessingh

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Using a pure machine learning approach to enable the generation of behavior for agents in serious gaming applications can be problematic, because such applications often require human-like behavior for agents that interact with human players. Such human-like behavior is not guaranteed with e.g. basic reinforcement learning schemes. Cognitive models can be very useful to establish human-like behavior in an agent. However, they require ample domain knowledge that might be difficult to obtain. In this paper, a cognitive model is taken as a basis, and the addition of scenario specific information is for a large part automated by means of machine learning techniques. The performance of the approach of automatically adding scenario specific information is rigorously evaluated using a case study in the domain of fighter air combat. An evolutionary algorithm is proposed for automatically tailoring a cognitive model for situation awareness of fighter pilots. The standard algorithm and several extensions are evaluated with respect to performance in air combat. The results show that it is possible to apply the algorithm to optimize belief networks for cognitive models of intelligent agents (adversarial fighters) in the aforementioned domain, thereby reducing the effort required to elicit knowledge from experts, while retaining the required ‘human-like’ behavior.
LanguageEnglish
Pages36-48
Number of pages13
JournalApplied Intelligence
Volume42
Issue number1
DOIs
Publication statusPublished - 2015

Fingerprint

Learning systems
Intelligent agents
Reinforcement learning
Bayesian networks
Air
Evolutionary algorithms

Keywords

  • Cognitive model
  • Evolutionary algorithm
  • Machine learning
  • Neural networks
  • Simulation
  • Situation awareness

Cite this

Koopmanschap, Richard ; Hoogendoorn, Mark ; Roessingh, Jan Joris. / Tailoring a Cognitive Model for Situation Awareness using Machine Learning. In: Applied Intelligence. 2015 ; Vol. 42, No. 1. pp. 36-48.
@article{d3e7d1a459024ffb95a8f1a38516b543,
title = "Tailoring a Cognitive Model for Situation Awareness using Machine Learning",
abstract = "Using a pure machine learning approach to enable the generation of behavior for agents in serious gaming applications can be problematic, because such applications often require human-like behavior for agents that interact with human players. Such human-like behavior is not guaranteed with e.g. basic reinforcement learning schemes. Cognitive models can be very useful to establish human-like behavior in an agent. However, they require ample domain knowledge that might be difficult to obtain. In this paper, a cognitive model is taken as a basis, and the addition of scenario specific information is for a large part automated by means of machine learning techniques. The performance of the approach of automatically adding scenario specific information is rigorously evaluated using a case study in the domain of fighter air combat. An evolutionary algorithm is proposed for automatically tailoring a cognitive model for situation awareness of fighter pilots. The standard algorithm and several extensions are evaluated with respect to performance in air combat. The results show that it is possible to apply the algorithm to optimize belief networks for cognitive models of intelligent agents (adversarial fighters) in the aforementioned domain, thereby reducing the effort required to elicit knowledge from experts, while retaining the required ‘human-like’ behavior.",
keywords = "Cognitive model, Evolutionary algorithm, Machine learning, Neural networks, Simulation, Situation awareness",
author = "Richard Koopmanschap and Mark Hoogendoorn and Roessingh, {Jan Joris}",
year = "2015",
doi = "10.1007/s10489-014-0584-3",
language = "English",
volume = "42",
pages = "36--48",
journal = "Applied Intelligence",
issn = "0924-669X",
publisher = "Springer Verlag",
number = "1",

}

Tailoring a Cognitive Model for Situation Awareness using Machine Learning. / Koopmanschap, Richard; Hoogendoorn, Mark; Roessingh, Jan Joris.

In: Applied Intelligence, Vol. 42, No. 1, 2015, p. 36-48.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Tailoring a Cognitive Model for Situation Awareness using Machine Learning

AU - Koopmanschap, Richard

AU - Hoogendoorn, Mark

AU - Roessingh, Jan Joris

PY - 2015

Y1 - 2015

N2 - Using a pure machine learning approach to enable the generation of behavior for agents in serious gaming applications can be problematic, because such applications often require human-like behavior for agents that interact with human players. Such human-like behavior is not guaranteed with e.g. basic reinforcement learning schemes. Cognitive models can be very useful to establish human-like behavior in an agent. However, they require ample domain knowledge that might be difficult to obtain. In this paper, a cognitive model is taken as a basis, and the addition of scenario specific information is for a large part automated by means of machine learning techniques. The performance of the approach of automatically adding scenario specific information is rigorously evaluated using a case study in the domain of fighter air combat. An evolutionary algorithm is proposed for automatically tailoring a cognitive model for situation awareness of fighter pilots. The standard algorithm and several extensions are evaluated with respect to performance in air combat. The results show that it is possible to apply the algorithm to optimize belief networks for cognitive models of intelligent agents (adversarial fighters) in the aforementioned domain, thereby reducing the effort required to elicit knowledge from experts, while retaining the required ‘human-like’ behavior.

AB - Using a pure machine learning approach to enable the generation of behavior for agents in serious gaming applications can be problematic, because such applications often require human-like behavior for agents that interact with human players. Such human-like behavior is not guaranteed with e.g. basic reinforcement learning schemes. Cognitive models can be very useful to establish human-like behavior in an agent. However, they require ample domain knowledge that might be difficult to obtain. In this paper, a cognitive model is taken as a basis, and the addition of scenario specific information is for a large part automated by means of machine learning techniques. The performance of the approach of automatically adding scenario specific information is rigorously evaluated using a case study in the domain of fighter air combat. An evolutionary algorithm is proposed for automatically tailoring a cognitive model for situation awareness of fighter pilots. The standard algorithm and several extensions are evaluated with respect to performance in air combat. The results show that it is possible to apply the algorithm to optimize belief networks for cognitive models of intelligent agents (adversarial fighters) in the aforementioned domain, thereby reducing the effort required to elicit knowledge from experts, while retaining the required ‘human-like’ behavior.

KW - Cognitive model

KW - Evolutionary algorithm

KW - Machine learning

KW - Neural networks

KW - Simulation

KW - Situation awareness

UR - http://www.scopus.com/inward/record.url?scp=84920504751&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920504751&partnerID=8YFLogxK

U2 - 10.1007/s10489-014-0584-3

DO - 10.1007/s10489-014-0584-3

M3 - Article

VL - 42

SP - 36

EP - 48

JO - Applied Intelligence

T2 - Applied Intelligence

JF - Applied Intelligence

SN - 0924-669X

IS - 1

ER -