Document Open Access Logo

HASCO: A Hybrid AI Simulation Compiler for Semantic Accident Reconstruction

Authors Edin Jelačić , Rong Gu , Cristina Seceleanu , Ning Xiong , Peter Backeman , Tiberiu Seceleanu , Zhennan Fei , Ali Nouri

PDF

File

Thumbnail PDF
PDF
OASIcs.AEiC.2026.4.pdf
  • Filesize: 3 MB
  • 22 pages

Document Identifiers

Subject Classification

ACM Subject Classification
  • Computing methodologies → Natural language generation
Keywords
  • Autonomous Driving
  • OpenSCENARIO
  • Large Language Models
  • Scenario Generation
  • Semantic Reconstruction

Metrics

Abstract

The validation of Automated Driving Systems (ADSs) has shifted from distance-based metrics to Scenario-Based Testing (SBT). Large Language Models (LLMs) have emerged as powerful tools with potential for generating vehicular scenarios at scale. However, generative models, used for direct simulation synthesis, produce inadequate output, therefore necessitating a more structured compilation approach. In this regard, we present HASCO (Hybrid AI Simulation COmpiler), a system that translates natural-language driving scene specifications into executable simulation artifacts (XOSC/XODR files) for the esmini/OpenSCENARIO ecosystem. While LLMs excel at narrative parsing, we demonstrate that direct synthesis of simulation artifacts fails in the vast majority of cases due to hallucinated physics or schema violations. To resolve this, HASCO treats scenario creation as a compilation task rather than a generative one. The pipeline supports three compilation paths: direct synthesis, a Python intermediate (via scenariogeneration), and an ontology-guided path that grounds intent into an intermediate representation (IR) before compilation. We further evaluate a self-judging mechanism for automated repair. Across six operating modes evaluated on 40 real-world accident reports, the ontology-guided compiler and Python-based compiler achieve 95% and 90% executability rates, respectively (compared to 5% for direct synthesis). Additionally, we evaluate outputs on semantic fidelity, positioning HASCO as a robust tool for forensic scene reconstruction.

Cite As Get BibTex

Edin Jelačić, Rong Gu, Cristina Seceleanu, Ning Xiong, Peter Backeman, Tiberiu Seceleanu, Zhennan Fei, and Ali Nouri. HASCO: A Hybrid AI Simulation Compiler for Semantic Accident Reconstruction. In 30th Ada-Europe International Conference on Reliable Software Technologies (AEiC 2026). Open Access Series in Informatics (OASIcs), Volume 143, pp. 4:1-4:22, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026) https://doi.org/10.4230/OASIcs.AEiC.2026.4

Author Details

Edin Jelačić
  • Department of Computer Science and Engineering, Mälardalen University, Sweden
Rong Gu
  • Prover Technology AB, Stockholm, Sweden
Cristina Seceleanu
  • Department of Computer Science and Engineering, Mälardalen University, Sweden
Ning Xiong
  • Department of Computer Science and Engineering, Mälardalen University, Sweden
Peter Backeman
  • Department of Computer Science and Engineering, Mälardalen University, Sweden
Tiberiu Seceleanu
  • Department of Computer Science and Engineering, Mälardalen University, Sweden
Zhennan Fei
  • Volvo Cars, Göteborg, Sweden
  • Chalmers University of Technology, Göteborg, Sweden
Ali Nouri
  • Volvo Cars, Göteborg, Sweden
  • Chalmers University of Technology, Göteborg, Sweden

Funding

We gratefully acknowledge the Swedish Knowledge Foundation via the PerFlex (Performant and Flexible digital Systems through Verifiable Artificial Intelligence) project, grant nr. 20220033, as well as Mälardalen University via the TSS-INSID (In-Silico Driving Assurance Using Machine-Learning-Powered Verification and Synthesis for Safe Autonomous Vehicles) project, of the Trusted Smart Systems initiative, which supported this research.

References

  1. ASAM e.V. ASAM OpenSCENARIO®: Standard for Dynamic Content in Driving Simulation, 2022. Version 1.x (XML format). URL: https://www.asam.net/standards/detail/openscenario/.
  2. Amy Aukema, Kate Berman, Travis Gaydos, Ted Sienknecht, Chou-Lin Chen, Chris Wiacek, Tim Czapp, and Schuyler St. Real-world effectiveness of model year 2015-2020 advanced driver assistance systems. In 27th International Technical Conference on the Enhanced Safety of Vehicles (ESV) National Highway Traffic Safety Administration., 2023. Google Scholar
  3. Gerrit Bagschik, Till Menzel, and Markus Maurer. Ontology based scene creation for the development of automated vehicles. In 2018 IEEE Intelligent Vehicles Symposium (IV), pages 1813-1820. IEEE, 06 2018. URL: https://doi.org/10.1109/IVS.2018.8500632.
  4. Daniel Bogdoll, Stefani Guneshka, and J Marius Zöllner. One ontology to rule them all: Corner case scenarios for autonomous driving, pages 409-425. Lecture Notes in Computer Science. Springer Nature Switzerland, 2023. Google Scholar
  5. Yao Deng, Jiaohong Yao, Zhi Tu, Xi Zheng, Mengshi Zhang, and Tianyi Zhang. TARGET: Automated scenario generation from traffic rules for testing autonomous vehicles via validated LLM-guided knowledge extraction. arXiv [cs.SE], 05 2023. URL: https://arxiv.org/abs/2305.06018.
  6. Alexey Dosovitskiy, German Ros, Felipe Codevilla, Antonio Lopez, and Vladlen Koltun. CARLA: An open urban driving simulator. In Sergey Levine, Vincent Vanhoucke, and Ken Goldberg, editors, Proceedings of the 1st Annual Conference on Robot Learning, volume 78 of Proceedings of Machine Learning Research, pages 1-16. PMLR, 13-15 Nov 2017. URL: https://proceedings.mlr.press/v78/dosovitskiy17a.html.
  7. Karim Elmaaroufi, Devan Shanker, Ana Cismaru, Marcell Vazquez-Chanlatte, Alberto Sangiovanni-Vincentelli, Matei Zaharia, and Sanjit A Seshia. ScenicNL: Generating probabilistic scenario programs from natural language. arXiv [cs.SE], 2024. URL: https://arxiv.org/abs/2405.03709.
  8. esmini development team. esmini: Basic OpenSCENARIO player and validator. https://github.com/esmini/esmini, 2025. Accessed: 2026-01-20.
  9. Roland Galbas, Marcus Nolte, Ulrich Eberle, Hardi Hungar, Henning Mosebach, Nayel Fabian Salem, Helmut Schittenhelm, Jan Reich, Thomas Kirschbaum, and Lukas Westhofen. VV methods safety assurance position paper. resreport, Robert Bosch GmbH und BMW AG, 2024. Google Scholar
  10. Yuan Gao, Mattia Piccinini, Korbinian Moller, Amr Alanwar, and Johannes Betz. From words to collisions: LLM-guided evaluation and adversarial generation of safety-critical driving scenarios. arXiv [cs.AI], 2025. URL: https://arxiv.org/abs/2502.02145.
  11. An Guo, Yuan Zhou, Haoxiang Tian, Chunrong Fang, Yunjian Sun, Weisong Sun, Xinyu Gao, Anh Tuan Luu, Yang Liu, and Zhenyu Chen. SoVAR: Build generalizable scenarios from accident reports for autonomous driving testing. In Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering, pages 268-280. ACM, 10 2024. URL: https://doi.org/10.1145/3691620.3695037.
  12. Pin Ji, Yang Feng, Zongtai Li, Xiangchi Zhou, Jia Liu, Jun Sun, and Zhihong Zhao. Txt2Sce: Scenario generation for autonomous driving system testing based on textual reports. arXiv [cs.SE], 2025. URL: https://arxiv.org/abs/2509.02150.
  13. Hoseon Kim, Jieun Ko, Cheol Oh, and Seoungbum Kim. Evaluation of autonomous driving safety by operational design domains (ODD) in mixed traffic. Sustainability, 16:9672, 11 2024. Google Scholar
  14. Stefan Jacobs Madlen Haarbach. Lieferwagen stand auf Radweg: Radfahrerin stirbt bei Unfall in Berlin-Friedrichshain-Mahnwache am Freitag, May 2021. URL: https://www.tagesspiegel.de/berlin/polizei-justiz/radfahrerin-stirbt-bei-unfall-in-berlin-friedrichshain--mahnwache-am-freitag-6174754.html.
  15. Till Menzel, Gerrit Bagschik, and Markus Maurer. Scenarios for development, test and validation of automated vehicles. arXiv [cs.SE], 01 2018. URL: https://arxiv.org/abs/1801.08598.
  16. Christian Neurohr, Lukas Westhofen, Tabea Henning, Thies de Graaff, Eike Mohlmann, and Eckard Bode. Fundamental considerations around scenario-based testing for automated driving. In 2020 IEEE Intelligent Vehicles Symposium (IV), pages 121-127. IEEE, 10 2020. Google Scholar
  17. Ali Nouri, Beatriz Cabrero-Daniel, Zhennan Fei, Krishna Ronanki, Håkan Sivencrona, and Christian Berger. Large language models in code co-generation for safe autonomous vehicles. arXiv [cs.SE], 05 2025. URL: https://arxiv.org/abs/2505.19658.
  18. Sebastian Pagel, Flo, Michi Scholz, Muhammed Kerem Kahraman, Pranav Ashok, victorjarlow, cxruan, Egan, Eric Xue, boettol, Emil Knabe, Hüseyin Aydın, and m0uH. grepthat/libOpenDRIVE: v0.5.0. Google Scholar
  19. Paolo Pareti and George Konstantinidis. A review of SHACL: From data validation to schema reasoning for RDF graphs, pages 115-144. Lecture Notes in Computer Science. Springer International Publishing, 2022. Google Scholar
  20. pyoscx. scenariogeneration: A python library for generating openscenario and opendrive files, 2024. Accessed: 2025-02-11. URL: https://github.com/pyoscx/scenariogeneration.
  21. Stefan Riedmaier, Thomas Ponn, Dieter Ludwig, Bernhard Schick, and Frank Diermeyer. Survey on scenario-based safety assessment of automated vehicles. IEEE Access, 8:87456-87477, 2020. URL: https://doi.org/10.1109/ACCESS.2020.2993730.
  22. Guodong Rong, Byung Hyun Shin, Hadi Tabatabaee, Qiang Lu, Steve Lemke, Mārtiņš Možeiko, Eric Boise, Geehoon Uhm, Mark Gerow, Shalin Mehta, Eugene Agafonov, Tae Hyung Kim, Eric Sterner, Keunhae Ushiroda, Michael Reyes, Dmitry Zelenkovsky, and Seonman Kim. LGSVL simulator: A high fidelity simulator for autonomous driving. arXiv [cs.RO], 2020. URL: https://arxiv.org/abs/2005.03778.
  23. Simon Ulbrich, Till Menzel, Andreas Reschka, Fabian Schuldt, and Markus Maurer. Defining and substantiating the terms scene, situation, and scenario for automated driving. In 2015 IEEE 18th International Conference on Intelligent Transportation Systems, pages 982-988. IEEE, 09 2015. URL: https://doi.org/10.1109/ITSC.2015.164.
  24. Walther Wachenfeld and Hermann Winner. The release of autonomous vehicles, pages 425-449. Springer Berlin Heidelberg, 2016. Google Scholar
  25. Hermann Winner, Karsten Lemmer, Thomas Form, and Jens Mazzega. PEGASUS—first steps for the safe introduction of automated driving, pages 185-195. Springer International Publishing, 2019. Google Scholar
  26. Yongqi Zhao, Wenbo Xiao, Tomislav Mihalj, Jia Hu, and Arno Eichberger. Chat2Scenario: Scenario extraction from dataset through utilization of large language model. In 2024 IEEE Intelligent Vehicles Symposium (IV). IEEE, 06 2024. Google Scholar
  27. Ji Zhou, Yongqi Zhao, Yixian Hu, Hexuan Li, Zhengguo Gu, Nan Xu, and Arno Eichberger. Can AI generate more comprehensive test scenarios? review on automated driving systems test scenario generation methods. arXiv [cs.SE], 12 2025. URL: https://arxiv.org/abs/2512.15422.
  28. Maximilian Zipfl, Nina Koch, and J Marius Zöllner. A comprehensive review on ontologies for scenario-based testing in the context of autonomous driving. In 2023 IEEE Intelligent Vehicles Symposium (IV), pages 1-7. IEEE, 06 2023. URL: https://doi.org/10.1109/IV55152.2023.10186681.
Any Issues?
X

Feedback on the Current Page

CAPTCHA

Thanks for your feedback!

Feedback submitted to Dagstuhl Publishing

Could not send message

Please try again later or send an E-mail
© 2023-2026 Schloss Dagstuhl – LZI GmbH Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact