Automatically Learning Formal Models from Autonomous Driving Software
The correctness of autonomous driving software is of utmost importance, as incorrect behavior may have catastrophic consequences. Formal model-based engineering techniques can help guarantee correctness and thereby allow the safe deployment of autonomous vehicles. However, challenges exist for wides...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/4/643 |
| _version_ | 1797480916705083392 |
|---|---|
| author | Yuvaraj Selvaraj Ashfaq Farooqui Ghazaleh Panahandeh Wolfgang Ahrendt Martin Fabian |
| author_facet | Yuvaraj Selvaraj Ashfaq Farooqui Ghazaleh Panahandeh Wolfgang Ahrendt Martin Fabian |
| author_sort | Yuvaraj Selvaraj |
| collection | DOAJ |
| description | The correctness of autonomous driving software is of utmost importance, as incorrect behavior may have catastrophic consequences. Formal model-based engineering techniques can help guarantee correctness and thereby allow the safe deployment of autonomous vehicles. However, challenges exist for widespread industrial adoption of formal methods. One of these challenges is the model construction problem. Manual construction of formal models is time-consuming, error-prone, and intractable for large systems. Automating model construction would be a big step towards widespread industrial adoption of formal methods for system development, re-engineering, and reverse engineering. This article applies <i>active learning</i> techniques to obtain formal models of an existing (under development) autonomous driving software module implemented in MATLAB. This demonstrates the feasibility of automated learning for automotive industrial use. Additionally, practical challenges in applying automata learning, and possible directions for integrating automata learning into the automotive software development workflow, are discussed. |
| first_indexed | 2024-03-09T22:07:03Z |
| format | Article |
| id | doaj.art-f7d0bb2993ab4dccad5149d21d9467c9 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-09T22:07:03Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-f7d0bb2993ab4dccad5149d21d9467c92023-11-23T19:40:39ZengMDPI AGElectronics2079-92922022-02-0111464310.3390/electronics11040643Automatically Learning Formal Models from Autonomous Driving SoftwareYuvaraj Selvaraj0Ashfaq Farooqui1Ghazaleh Panahandeh2Wolfgang Ahrendt3Martin Fabian4Zenseact, 417 56 Gothenburg, SwedenDepartment of Electrical Engineering, Chalmers University of Technology, 412 96 Gothenburg, SwedenZenseact, 417 56 Gothenburg, SwedenDepartment of Computer Science and Engineering, Chalmers University of Technology, 412 96 Gothenburg, SwedenDepartment of Electrical Engineering, Chalmers University of Technology, 412 96 Gothenburg, SwedenThe correctness of autonomous driving software is of utmost importance, as incorrect behavior may have catastrophic consequences. Formal model-based engineering techniques can help guarantee correctness and thereby allow the safe deployment of autonomous vehicles. However, challenges exist for widespread industrial adoption of formal methods. One of these challenges is the model construction problem. Manual construction of formal models is time-consuming, error-prone, and intractable for large systems. Automating model construction would be a big step towards widespread industrial adoption of formal methods for system development, re-engineering, and reverse engineering. This article applies <i>active learning</i> techniques to obtain formal models of an existing (under development) autonomous driving software module implemented in MATLAB. This demonstrates the feasibility of automated learning for automotive industrial use. Additionally, practical challenges in applying automata learning, and possible directions for integrating automata learning into the automotive software development workflow, are discussed.https://www.mdpi.com/2079-9292/11/4/643autonomous drivingactive learningformal methodsmodel-based engineeringautomata learning |
| spellingShingle | Yuvaraj Selvaraj Ashfaq Farooqui Ghazaleh Panahandeh Wolfgang Ahrendt Martin Fabian Automatically Learning Formal Models from Autonomous Driving Software Electronics autonomous driving active learning formal methods model-based engineering automata learning |
| title | Automatically Learning Formal Models from Autonomous Driving Software |
| title_full | Automatically Learning Formal Models from Autonomous Driving Software |
| title_fullStr | Automatically Learning Formal Models from Autonomous Driving Software |
| title_full_unstemmed | Automatically Learning Formal Models from Autonomous Driving Software |
| title_short | Automatically Learning Formal Models from Autonomous Driving Software |
| title_sort | automatically learning formal models from autonomous driving software |
| topic | autonomous driving active learning formal methods model-based engineering automata learning |
| url | https://www.mdpi.com/2079-9292/11/4/643 |
| work_keys_str_mv | AT yuvarajselvaraj automaticallylearningformalmodelsfromautonomousdrivingsoftware AT ashfaqfarooqui automaticallylearningformalmodelsfromautonomousdrivingsoftware AT ghazalehpanahandeh automaticallylearningformalmodelsfromautonomousdrivingsoftware AT wolfgangahrendt automaticallylearningformalmodelsfromautonomousdrivingsoftware AT martinfabian automaticallylearningformalmodelsfromautonomousdrivingsoftware |