Frenetic-lib: An extensible framework for search-based generation of road structures for ADS testing

Ezequiel Castellano; Stefan Klikovits; Ahmet Cetinkaya; Paolo Arcaini

doi:10.1016/j.scico.2023.102996

Frenetic-lib: An extensible framework for search-based generation of road structures for ADS testing

Ezequiel Castellano
, Stefan Klikovits
, Ahmet Cetinkaya
, Paolo Arcaini

Institute of Business Informatics - Software Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

Being capable of identifying significant safety shortcomings, search-based methods are a core tool for testing automated driving system (ADS) technologies. In this domain, Frenetic has proven to be a popular and very effective tool, searching and identifying diverse sets of roads that point out potentially faulty ADS behaviour. This paper presents Frenetic-lib, a Python library that captures Frenetic’s novel combination of road representation and genetic algorithm, and makes it generally available in a customisable way. Next to the capacity to integrate additional ADS simulators, Frenetic-lib further creates new research opportunities on search-based road testing, novel road representations and mutation operators.

Original language	English
Article number	102996
Number of pages	10
Journal	Science of Computer Programming
Volume	230
DOIs	https://doi.org/10.1016/j.scico.2023.102996
Publication status	Published - Aug 2023

Fields of science

102006 Computer supported cooperative work (CSCW)
102015 Information systems
102016 IT security
102020 Medical informatics
102022 Software development
102027 Web engineering
102034 Cyber-physical systems
509026 Digitalisation research
502032 Quality management
502050 Business informatics
503015 Subject didactics of technical sciences

JKU Focus areas

Digital Transformation

Access to Document

10.1016/j.scico.2023.102996

Cite this

@article{5d15851ad23448d9819f66e6066293f9,

title = "Frenetic-lib: An extensible framework for search-based generation of road structures for ADS testing",

abstract = "Being capable of identifying significant safety shortcomings, search-based methods are a core tool for testing automated driving system (ADS) technologies. In this domain, Frenetic has proven to be a popular and very effective tool, searching and identifying diverse sets of roads that point out potentially faulty ADS behaviour. This paper presents Frenetic-lib, a Python library that captures Frenetic{\textquoteright}s novel combination of road representation and genetic algorithm, and makes it generally available in a customisable way. Next to the capacity to integrate additional ADS simulators, Frenetic-lib further creates new research opportunities on search-based road testing, novel road representations and mutation operators.",

author = "Ezequiel Castellano and Stefan Klikovits and Ahmet Cetinkaya and Paolo Arcaini",

year = "2023",

month = aug,

doi = "10.1016/j.scico.2023.102996",

language = "English",

volume = "230",

journal = "Science of Computer Programming",

issn = "0167-6423",

publisher = "Elsevier North-Holland, Inc. Amsterdam, The Netherlands, The Netherlands",

}

TY - JOUR

T1 - Frenetic-lib: An extensible framework for search-based generation of road structures for ADS testing

AU - Castellano, Ezequiel

AU - Klikovits, Stefan

AU - Cetinkaya, Ahmet

AU - Arcaini, Paolo

PY - 2023/8

Y1 - 2023/8

N2 - Being capable of identifying significant safety shortcomings, search-based methods are a core tool for testing automated driving system (ADS) technologies. In this domain, Frenetic has proven to be a popular and very effective tool, searching and identifying diverse sets of roads that point out potentially faulty ADS behaviour. This paper presents Frenetic-lib, a Python library that captures Frenetic’s novel combination of road representation and genetic algorithm, and makes it generally available in a customisable way. Next to the capacity to integrate additional ADS simulators, Frenetic-lib further creates new research opportunities on search-based road testing, novel road representations and mutation operators.

AB - Being capable of identifying significant safety shortcomings, search-based methods are a core tool for testing automated driving system (ADS) technologies. In this domain, Frenetic has proven to be a popular and very effective tool, searching and identifying diverse sets of roads that point out potentially faulty ADS behaviour. This paper presents Frenetic-lib, a Python library that captures Frenetic’s novel combination of road representation and genetic algorithm, and makes it generally available in a customisable way. Next to the capacity to integrate additional ADS simulators, Frenetic-lib further creates new research opportunities on search-based road testing, novel road representations and mutation operators.

UR - https://www.sciencedirect.com/science/article/pii/S0167642323000783?via%3Dihub

UR - https://www.scopus.com/pages/publications/85165044862

U2 - 10.1016/j.scico.2023.102996

DO - 10.1016/j.scico.2023.102996

M3 - Article

SN - 0167-6423

VL - 230

JO - Science of Computer Programming

JF - Science of Computer Programming

M1 - 102996

ER -

Frenetic-lib: An extensible framework for search-based generation of road structures for ADS testing

Abstract

Fields of science

JKU Focus areas

Access to Document

Other files and links

Cite this