← Back to the feed
AI & softwareResearch findings

Evaluation of Spatial Awareness and Traffic Rule Compliance in LLM-Based Autonomous Driving

Original published 2023.12.11 21:56 KSTAdded to Blackdeck 2026.11.15 07:25 KST

What has changed?

Quantitative assessment of spatial awareness and traffic rule adherence in LLM-based autonomous driving systems

Key points

  1. LLMs can convert surrounding object coordinates and velocities into text prompts to generate driving decisions
  2. Spatial awareness and traffic rule compliance are quantitatively evaluated
  3. A proof-of-concept system for LLM-driven vehicle operation is developed

Connect it to your work

The system inputs coordinate and velocity data of surrounding objects as text prompts to generate driving decisions

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.

  1. arxivEvaluation of Large Language Models for Decision Making in Autonomous Driving