Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ARISE -- Adaptive Refinement and Iterative Scenario Engineering

Konstantin Poddubnyy, Igor Vozniak, Nils Lipp, Ivan Burmistrov, Davit Hovhannisyan, Christian Mueller, Philipp Slusallek

TL;DR

The paper addresses the challenge of generating diverse, executable synthetic traffic scenarios for autonomous driving using natural language prompts. It introduces ARISE, a multi-stage, LLM-driven pipeline that converts prompts into Scenic scripts through an adaptive, iterative Test-and-Repair Loop. Key contributions include expanded semantic extraction, an updated Scenic knowledge base, and automated testing and repair that substantially improve execution success rates and semantic conformity, as well as a robust evaluation demonstrating improved reliability across scenarios and LLMs. This approach enables fully automated scenario generation for safety-critical testing in CARLA and downstream evaluation in SafeBench, reducing manual intervention and increasing the reliability of training data for collision-free trajectory planning systems.

Abstract

The effectiveness of collision-free trajectory planners depends on the quality and diversity of training data, especially for rare scenarios. A widely used approach to improve dataset diversity involves generating realistic synthetic traffic scenarios. However, producing such scenarios remains difficult due to the precision required when scripting them manually or generating them in a single pass. Natural language offers a flexible way to describe scenarios, but existing text-to-simulation pipelines often rely on static snippet retrieval, limited grammar, single-pass decoding, or lack robust executability checks. Moreover, they depend heavily on constrained LLM prompting with minimal post-processing. To address these limitations, we introduce ARISE - Adaptive Refinement and Iterative Scenario Engineering, a multi-stage tool that converts natural language prompts into executable Scenic scripts through iterative LLM-guided refinement. After each generation, ARISE tests script executability in simulation software, feeding structured diagnostics back to the LLM until both syntactic and functional requirements are met. This process significantly reduces the need for manual intervention. Through extensive evaluation, ARISE outperforms the baseline in generating semantically accurate and executable traffic scenarios with greater reliability and robustness.

ARISE -- Adaptive Refinement and Iterative Scenario Engineering

TL;DR

The paper addresses the challenge of generating diverse, executable synthetic traffic scenarios for autonomous driving using natural language prompts. It introduces ARISE, a multi-stage, LLM-driven pipeline that converts prompts into Scenic scripts through an adaptive, iterative Test-and-Repair Loop. Key contributions include expanded semantic extraction, an updated Scenic knowledge base, and automated testing and repair that substantially improve execution success rates and semantic conformity, as well as a robust evaluation demonstrating improved reliability across scenarios and LLMs. This approach enables fully automated scenario generation for safety-critical testing in CARLA and downstream evaluation in SafeBench, reducing manual intervention and increasing the reliability of training data for collision-free trajectory planning systems.

Abstract

The effectiveness of collision-free trajectory planners depends on the quality and diversity of training data, especially for rare scenarios. A widely used approach to improve dataset diversity involves generating realistic synthetic traffic scenarios. However, producing such scenarios remains difficult due to the precision required when scripting them manually or generating them in a single pass. Natural language offers a flexible way to describe scenarios, but existing text-to-simulation pipelines often rely on static snippet retrieval, limited grammar, single-pass decoding, or lack robust executability checks. Moreover, they depend heavily on constrained LLM prompting with minimal post-processing. To address these limitations, we introduce ARISE - Adaptive Refinement and Iterative Scenario Engineering, a multi-stage tool that converts natural language prompts into executable Scenic scripts through iterative LLM-guided refinement. After each generation, ARISE tests script executability in simulation software, feeding structured diagnostics back to the LLM until both syntactic and functional requirements are met. This process significantly reduces the need for manual intervention. Through extensive evaluation, ARISE outperforms the baseline in generating semantically accurate and executable traffic scenarios with greater reliability and robustness.
Paper Structure (18 sections, 3 equations, 3 figures, 4 tables)

This paper contains 18 sections, 3 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Synthetic Traffic Scenario Generation
  4. LLM-empowered Frameworks
  5. LLM-based Scenario Generation Using Scenic
  6. Methodology
  7. Scenario Prompts
  8. Module A: Extraction
  9. Module B: Retrieval
  10. Module C: Script Assembly
  11. Module D: Test-and-Repair Loop (TRL)
  12. Evaluations
  13. Metrics
  14. Quantitative Evaluations
  15. Qualitative Evaluations
  16. ...and 3 more sections

Figures (3)

  • Figure 1: Prompt-based traffic scenario generation (Carla rendered).
  • Figure 2: ARISE pipeline (modules A-D) and LLM-driven semantics evaluator.
  • Figure 3: A sequence of scenario events: during nighttime with light fog, the adversarial vehicle (blue rectangle) performs a lane change, merging into the same lane as the ego vehicle (red rectangle), and subsequently brakes in front of an obstacle.