Evolutionary Algorithm for Parameter Optimization of Context Steering Agents

Research output: Contribution to journalArticleResearchpeer review

Authors

External Research Organisations

  • Otto-von-Guericke University Magdeburg
  • Polaritli GmbH
View graph of relations

Details

Original languageEnglish
Pages (from-to)26-35
Number of pages10
JournalIEEE Transactions on Games
Volume15
Issue number1
Publication statusPublished - 2022
Externally publishedYes

Abstract

Context Steering is a local approach to control an agent's movement in a dynamically changing scene. Recent works have formalized the context steering approach by Fray and presented a multi-objective view of the context steering problem. Combining a variety of different behaviors, which can be used multiple times in different configurations for different context maps, introduces a large number of parameters that need to be tuned to obtain well-performing agents. This work aims to use evolutionary algorithms to optimize context steering agents for various environments. A special focus lies on the evolution of agents that perform robustly across multiple variations of the same environment. To this end, we develop a real-valued encoding for a context steering agent along with three different fitness functions to represent different goals of the agent. Our experimental evaluation shows that an evolutionary optimization can produce agent configurations that perform well with respect to different tasks and show a high intra-task robustness. The proposed approach based on evolutionary optimization enables the user to optimize context steering agents such that they can explore environments while avoiding dynamic obstacles.

Keywords

    Autonomous Movement, Context Steering, Evolutionary Algorithms, Games, Multi-Criteria Optimization, Optimization, Robustness, Search problems, Steering systems, Task analysis, Tuning, robustness, context steering, Autonomous movement, multicriteria optimization, evolut- ionary algorithms (EAs)

ASJC Scopus subject areas

Cite this

Evolutionary Algorithm for Parameter Optimization of Context Steering Agents. / Dockhorn, Alexander; Kirst, Martin; Mostaghim, Sanaz et al.
In: IEEE Transactions on Games, Vol. 15, No. 1, 2022, p. 26-35.

Research output: Contribution to journalArticleResearchpeer review

Dockhorn, A, Kirst, M, Mostaghim, S, Wieczorek, M & Zille, H 2022, 'Evolutionary Algorithm for Parameter Optimization of Context Steering Agents', IEEE Transactions on Games, vol. 15, no. 1, pp. 26-35. https://doi.org/10.1109/TG.2022.3157247
Dockhorn, A., Kirst, M., Mostaghim, S., Wieczorek, M., & Zille, H. (2022). Evolutionary Algorithm for Parameter Optimization of Context Steering Agents. IEEE Transactions on Games, 15(1), 26-35. https://doi.org/10.1109/TG.2022.3157247
Dockhorn A, Kirst M, Mostaghim S, Wieczorek M, Zille H. Evolutionary Algorithm for Parameter Optimization of Context Steering Agents. IEEE Transactions on Games. 2022;15(1):26-35. doi: 10.1109/TG.2022.3157247
Dockhorn, Alexander ; Kirst, Martin ; Mostaghim, Sanaz et al. / Evolutionary Algorithm for Parameter Optimization of Context Steering Agents. In: IEEE Transactions on Games. 2022 ; Vol. 15, No. 1. pp. 26-35.
Download
@article{7fe547aff33b46b98e72f19fc8c1db50,
title = "Evolutionary Algorithm for Parameter Optimization of Context Steering Agents",
abstract = "Context Steering is a local approach to control an agent's movement in a dynamically changing scene. Recent works have formalized the context steering approach by Fray and presented a multi-objective view of the context steering problem. Combining a variety of different behaviors, which can be used multiple times in different configurations for different context maps, introduces a large number of parameters that need to be tuned to obtain well-performing agents. This work aims to use evolutionary algorithms to optimize context steering agents for various environments. A special focus lies on the evolution of agents that perform robustly across multiple variations of the same environment. To this end, we develop a real-valued encoding for a context steering agent along with three different fitness functions to represent different goals of the agent. Our experimental evaluation shows that an evolutionary optimization can produce agent configurations that perform well with respect to different tasks and show a high intra-task robustness. The proposed approach based on evolutionary optimization enables the user to optimize context steering agents such that they can explore environments while avoiding dynamic obstacles.",
keywords = "Autonomous Movement, Context Steering, Evolutionary Algorithms, Games, Multi-Criteria Optimization, Optimization, Robustness, Search problems, Steering systems, Task analysis, Tuning, robustness, context steering, Autonomous movement, multicriteria optimization, evolut- ionary algorithms (EAs)",
author = "Alexander Dockhorn and Martin Kirst and Sanaz Mostaghim and Martin Wieczorek and Heiner Zille",
note = "Publisher Copyright: {\textcopyright} 2018 IEEE.",
year = "2022",
doi = "10.1109/TG.2022.3157247",
language = "English",
volume = "15",
pages = "26--35",
number = "1",

}

Download

TY - JOUR

T1 - Evolutionary Algorithm for Parameter Optimization of Context Steering Agents

AU - Dockhorn, Alexander

AU - Kirst, Martin

AU - Mostaghim, Sanaz

AU - Wieczorek, Martin

AU - Zille, Heiner

N1 - Publisher Copyright: © 2018 IEEE.

PY - 2022

Y1 - 2022

N2 - Context Steering is a local approach to control an agent's movement in a dynamically changing scene. Recent works have formalized the context steering approach by Fray and presented a multi-objective view of the context steering problem. Combining a variety of different behaviors, which can be used multiple times in different configurations for different context maps, introduces a large number of parameters that need to be tuned to obtain well-performing agents. This work aims to use evolutionary algorithms to optimize context steering agents for various environments. A special focus lies on the evolution of agents that perform robustly across multiple variations of the same environment. To this end, we develop a real-valued encoding for a context steering agent along with three different fitness functions to represent different goals of the agent. Our experimental evaluation shows that an evolutionary optimization can produce agent configurations that perform well with respect to different tasks and show a high intra-task robustness. The proposed approach based on evolutionary optimization enables the user to optimize context steering agents such that they can explore environments while avoiding dynamic obstacles.

AB - Context Steering is a local approach to control an agent's movement in a dynamically changing scene. Recent works have formalized the context steering approach by Fray and presented a multi-objective view of the context steering problem. Combining a variety of different behaviors, which can be used multiple times in different configurations for different context maps, introduces a large number of parameters that need to be tuned to obtain well-performing agents. This work aims to use evolutionary algorithms to optimize context steering agents for various environments. A special focus lies on the evolution of agents that perform robustly across multiple variations of the same environment. To this end, we develop a real-valued encoding for a context steering agent along with three different fitness functions to represent different goals of the agent. Our experimental evaluation shows that an evolutionary optimization can produce agent configurations that perform well with respect to different tasks and show a high intra-task robustness. The proposed approach based on evolutionary optimization enables the user to optimize context steering agents such that they can explore environments while avoiding dynamic obstacles.

KW - Autonomous Movement

KW - Context Steering

KW - Evolutionary Algorithms

KW - Games

KW - Multi-Criteria Optimization

KW - Optimization

KW - Robustness

KW - Search problems

KW - Steering systems

KW - Task analysis

KW - Tuning

KW - robustness

KW - context steering

KW - Autonomous movement

KW - multicriteria optimization

KW - evolut- ionary algorithms (EAs)

UR - http://www.scopus.com/inward/record.url?scp=85126286055&partnerID=8YFLogxK

U2 - 10.1109/TG.2022.3157247

DO - 10.1109/TG.2022.3157247

M3 - Article

AN - SCOPUS:85126286055

VL - 15

SP - 26

EP - 35

JO - IEEE Transactions on Games

JF - IEEE Transactions on Games

SN - 2475-1502

IS - 1

ER -

By the same author(s)