ResearchAI Engineering & DevTools AI Research

Chat2Scenic Boosts Autonomous Driving Scenario Generation with RAG.

Yuan Gao, Wenting Miao, Mattia Piccinini, Haoyu Wang, Qunying Song, Johannes Betz· July 17, 2026 View original

Summary

Chat2Scenic is an iterative RAG-based framework that generates diverse, regulation-compliant test scenarios for autonomous driving systems using a chatbot interface and DSL scripts. It significantly outperforms existing methods in compilation success rate and framework accuracy.

Autonomous driving systems require extensive testing with diverse, regulation-compliant scenarios. Generating these scenarios automatically, especially as executable scripts in a Domain Specific Language (DSL), has been a persistent challenge. Existing methods struggle with either scalability or success rates. Researchers have introduced Chat2Scenic, an innovative iterative framework leveraging Retrieval-Augmented Generation (RAG). This system features a chatbot interface for interactive scenario refinement and integrates RAG to ensure scenario generation adheres to regulatory knowledge and DSL syntax. Chat2Scenic demonstrates superior performance compared to current state-of-the-art methods, achieving a 76.42% compilation success rate and 58.17% framework accuracy. The project also includes an open benchmark of 123 scenarios from various regulations, and its code is open-sourced to foster further research.

Why it matters

This framework offers a more efficient and reliable way to create complex test scenarios for autonomous vehicles, accelerating development and improving safety validation. Professionals in automotive AI can leverage this to streamline their testing pipelines.

How to implement this in your domain

1Explore the open-source Chat2Scenic code to understand its architecture and RAG implementation.
2Integrate the framework into existing autonomous driving simulation environments for scenario generation.
3Utilize the interactive chatbot interface to refine and customize specific test scenarios based on regulatory requirements.
4Contribute to or adapt the provided open benchmark to validate new autonomous driving features.

Who benefits

AutomotiveRoboticsSoftware DevelopmentRegulatory Compliance

Key takeaways

Chat2Scenic significantly improves autonomous driving scenario generation using an iterative RAG framework.
The system offers a chatbot interface for interactive refinement and grounds generation in regulatory knowledge.
It achieves higher compilation success rates and framework accuracy than previous methods.
The open-source release and benchmark facilitate further research and adoption in the industry.

Original post by Yuan Gao, Wenting Miao, Mattia Piccinini, Haoyu Wang, Qunying Song, Johannes Betz

"arXiv:2607.14387v1 Announce Type: new Abstract: Validating autonomous driving systems requires diverse, regulation-compliant test scenarios. In simulation-based testing, scenarios are defined as executable scripts. Yet automatically generating such scripts from regulatory descrip…"

View on X

Primary sources