AI & softwareResearch findings
Proposal of ADRD: LLM-Driven Rule-Based Autonomous Driving System
Original published 2025.06.17 17:18 KSTAdded to Blackdeck 2026.11.08 10:43 KST
What has changed?
First integration of LLMs with rule-based systems for autonomous driving decision-making
Key points
- Proposes an interpretable rule-based autonomous driving framework using LLMs
- Generates context-aware rule-based driving tactics via integrated information and agent modules
- Iteratively refines tactics through continuous interaction with a testing module
Connect it to your work
Collects driving context and generates rule-based tactics through iterative testing
What remains unproven
This research note is based on an abstract; check the original for test conditions and application scope.
Originals used for this note
Links open the original publisher in a new tab.