Surrogate-Assisted High-Dimensional Optimization on Microscopic Traffic Simulators

Bernhard Werth; Erik Pitzer; Gerald Ostermayer; Michael Affenzeller

Surrogate-Assisted High-Dimensional Optimization on Microscopic Traffic Simulators

Bernhard Werth
, Erik Pitzer
, Gerald Ostermayer
, Michael Affenzeller

Institute of Formal Models and Verification

Research output: Chapter in Book/Report/Conference proceeding › Conference proceedings › peer-review

Abstract

Microscopic traffic simulation is able to capture many details of a traffic system, which makes it inherently interesting for simulation-based optimization. However, the considerable computational effort required for a single simulation run limits the use of standard heuristic optimization techniques and encourages the use of surrogate models to facilitate the search for an optimal solution. In this work, a grey-box surrogate model for microscopic traffic simulations is presented which allows the optimization of high-dimensional traffic optimization problems without relying on geographic or simulation-specific knowledge.

Original language	English
Title of host publication	Proceedings of the 30th European Modeling and Simulation Symposium EMSS2018
Number of pages	7
Publication status	Published - 2018

Fields of science

102 Computer Sciences
102001 Artificial intelligence
102011 Formal languages
102022 Software development
102031 Theoretical computer science
603109 Logic
202006 Computer hardware

JKU Focus areas

Computation in Informatics and Mathematics

Cite this

@inproceedings{abdd694198a54c158a9aa9caa11a281f,

title = "Surrogate-Assisted High-Dimensional Optimization on Microscopic Traffic Simulators",

abstract = "Microscopic traffic simulation is able to capture many details of a traffic system, which makes it inherently interesting for simulation-based optimization. However, the considerable computational effort required for a single simulation run limits the use of standard heuristic optimization techniques and encourages the use of surrogate models to facilitate the search for an optimal solution. In this work, a grey-box surrogate model for microscopic traffic simulations is presented which allows the optimization of high-dimensional traffic optimization problems without relying on geographic or simulation-specific knowledge.",

author = "Bernhard Werth and Erik Pitzer and Gerald Ostermayer and Michael Affenzeller",

year = "2018",

language = "English",

booktitle = "Proceedings of the 30th European Modeling and Simulation Symposium EMSS2018",

}

TY - GEN

T1 - Surrogate-Assisted High-Dimensional Optimization on Microscopic Traffic Simulators

AU - Werth, Bernhard

AU - Pitzer, Erik

AU - Ostermayer, Gerald

AU - Affenzeller, Michael

PY - 2018

Y1 - 2018

N2 - Microscopic traffic simulation is able to capture many details of a traffic system, which makes it inherently interesting for simulation-based optimization. However, the considerable computational effort required for a single simulation run limits the use of standard heuristic optimization techniques and encourages the use of surrogate models to facilitate the search for an optimal solution. In this work, a grey-box surrogate model for microscopic traffic simulations is presented which allows the optimization of high-dimensional traffic optimization problems without relying on geographic or simulation-specific knowledge.

AB - Microscopic traffic simulation is able to capture many details of a traffic system, which makes it inherently interesting for simulation-based optimization. However, the considerable computational effort required for a single simulation run limits the use of standard heuristic optimization techniques and encourages the use of surrogate models to facilitate the search for an optimal solution. In this work, a grey-box surrogate model for microscopic traffic simulations is presented which allows the optimization of high-dimensional traffic optimization problems without relying on geographic or simulation-specific knowledge.

M3 - Conference proceedings

BT - Proceedings of the 30th European Modeling and Simulation Symposium EMSS2018

ER -