CORTEX-AVD: A Framework for CORner Case Testing and EXploration in Autonomous Vehicle Development
Gabriel Kenji Godoy Shimanuki, Alexandre Moreira Nascimento, Lucio Flavio Vismari, João Batista Camargo, Jr., Jorge Rady de Almeida, Jr., Paulo Sergio Cugnasca · 2025
Read CORTEX-AVD: A Framework for CORner Case Testing and EXploration in Autonomous Vehicle Development by Gabriel Kenji Godoy Shimanuki, Alexandre Moreira Nascimento, Lucio Flavio Vismari, João Batista Camargo, Jr., Jorge Rady de Almeida, Jr., Paulo Sergio Cugnasca (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-03989.txt.
Description
Gabriel Shimanuki and co-authors present a 2025 framework for testing self driving cars on rare, risky cases. Such corner cases, like a sudden crossing or a surprise swerve, are hard to gather in real life and slow to build by hand in simulation. Their open source tool aims to make these cases faster and more varied.
The framework, CORTEX-AVD, joins a car simulator with a scene language to build cases from text prompts. A genetic search then tunes the scene values to raise the chance of high risk events. Its score weighs distance, time, speed, and crash chance together. The study also gives a benchmark for comparing search based case building. The authors say the tool can speed up safety work for self driving cars.
Catalog record
| CREATOR | Gabriel Kenji Godoy Shimanuki, Alexandre Moreira Nascimento, Lucio Flavio Vismari, João Batista Camargo, Jr., Jorge Rady de Almeida, Jr., Paulo Sergio Cugnasca |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-03989