Wiki › Project

Driving scenario generator

Project · research · 2025

Driving scenario generator is an experimental tool, still in development, that uses a large language model to write driving scenarios, such as a car merging onto a highway, for testing automated driving systems. He worked on it in mid-2025, at the start of his research at JAIST on safety testing for automated driving.

How it works

A prompt is assembled from three kinds of reusable pieces, each kept as a file in the prompts folder:

instruction tokens
what to generate;
base scenarios
a scenario to start from (optional);
experience tokens
lessons from earlier runs (optional).

A Python command-line tool combines them, sends the prompt to an OpenAI model (or to a mock client for testing without the API), and saves the result.

Status

The changelog is honest about the first version's limit: it produced descriptive scenarios, written as prose, which can't be fed directly to a simulator for testing. The structured, ontology-based side of the problem became his ADS safety-testing ontology and semantic extraction pipeline.

See also

In the knowledge graph

Driving scenario generator and the 11 things it is linked to. Open any dot for its page, or explore it in the full graph.

Semantic extraction pipeline · work · mentionsSemantic extracti…ADS safety-testing ontology · work · RelatedADS ontologyFlask OpenAI chatbot · work · links hereFlask OpenAI chat…Python · skill · Built withPythonLarge language models · skill · Built withLLMsOpenAI API · skill · Built withOpenAI APIMachine learning and data · theme · ThemesMachine learning …Research tooling · theme · ThemesResearch toolingOpen source on GitHub · person, role or place · ContextOpen sourceJoint researcher at JAIST · person, role or place · mentionsJoint researcherOpenAI models · service or platform · Works withOpenAI models Drivingscenario ge…
11 linked in the graph: 3 works3 skills2 themes2 people, roles and places1 service or platform
When
2023202420252026
Jun 2025 · 2 commits

As data

Everything on this page is read from triples in the knowledge graph. These are the 22 it states about Driving scenario generator, and 2 more its ontology implies; the console runs SPARQL over the whole graph, here in your browser.

Download
@prefix id: <https://ask.chaipat.cc/id/> .
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix cj: <https://ask.chaipat.cc/ns#> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .

id:ScenarioGenerator
    a cj:Project ;
    rdfs:label "Driving scenario generator" ;
    cj:description "LLM-driven generator of driving scenarios (e.g. highway merges) from prompts and past experiences; in development." ;
    cj:context id:OpenSource ;
    cj:repository "https://github.com/AppleBoiy/driving_scenario_generator"^^xsd:anyURI ;
    cj:firstCommit "2025-06-02"^^xsd:date ;
    cj:lastCommit "2025-06-02"^^xsd:date ;
    cj:commitCount 2 ;
    cj:kind "research" ;
    cj:theme id:ThemeML , id:ThemeResearchTools ;
    cj:uses id:Python , id:LLMs , id:OpenAIAPI ;
    cj:tier 2 ;
    cj:integrates id:OpenAI ;
    cj:relatedTo id:ADSOntology ;
    cj:mentions id:RoleJAIST , id:SemanticExtraction ;
    cj:article "https://ask.chaipat.cc/wiki/ScenarioGenerator"^^xsd:anyURI ;
    cj:activeIn id:Y2025 ;
    cj:activity _:b146 .
Works 2
ADS ontologyFlask OpenAI chatbot