Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 10, 2025ElectronicsOpen Access

Automated Generation of Test Scenarios for Autonomous Driving Using LLMs

View Full Paper
Ask AI
Bookmark
Share

Authors

ADAaron Agyapong DansoUBUlrich Büker

Discussion

Loading...

Member takes

Overview

This approach employs large language models to generate simulation scenarios for autonomous vehicles, highlighting key challenges and opportunities.

Key Points

  • The method significantly reduces manual effort in creating simulation scenarios for autonomous vehicles, enhancing efficiency.
  • Experimental results indicate that while static components like weather are well addressed, dynamic elements like behavior of pedestrians require further refinement.
  • This approach utilizes a blend of model-based and data-driven techniques to effectively decompose Operational Design Domains into their components.
  • Prompt engineering guides large language models in generating ScenarioRunner scripts compatible with CARLA, improving scenario generation.

Cite This Study

Danso et al. (2025) studied this question.

synapsesocial.com/papers/68c1c23d54b1d3bfb60efe38https://doi.org/10.3390/electronics14163177
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1OmniTester: Multimodal Large Language Model Driven Scenario Testing for Autonomous Vehicles2025
  2. 2Large Language Model-Based Realistic Safety-Critical Driving Video Generation2026
  3. 3Seeking to Collide: Online Safety-Critical Scenario Generation for Autonomous Driving with Retrieval Augmented Large Language Models2025 · 1 citations
  4. 4Text2Scenario: Text-Driven Scenario Generation for Autonomous Driving Test2025
  5. 5Chat2Scenario: Scenario Extraction From Dataset Through Utilization of Large Language Model2024