Can LLMs Understand Social Norms in Autonomous Driving Games?

TL;DR

The paper investigates whether Large Language Models can understand and form social norms in autonomous driving scenarios by deploying GPT-4-based agents within Markov games. It introduces a prompt-chaining framework where each agent in a multi-agent system makes decisions based on system and user prompts, enabling analysis of norm emergence. Two driving scenarios are explored: an unsignalized intersection and a highway platoon, with results showing the emergence of norms such as yielding at intersections and forming platoons on highways. The study highlights the strong operability and analyzability of LLM-based agents for experimental design and suggests future work on more complex scenarios and comparisons with human behavior to enhance socially aware autonomous driving systems.

Abstract

Social norm is defined as a shared standard of acceptable behavior in a society. The emergence of social norms fosters coordination among agents without any hard-coded rules, which is crucial for the large-scale deployment of AVs in an intelligent transportation system. This paper explores the application of LLMs in understanding and modeling social norms in autonomous driving games. We introduce LLMs into autonomous driving games as intelligent agents who make decisions according to text prompts. These agents are referred to as LLM-based agents. Our framework involves LLM-based agents playing Markov games in a multi-agent system (MAS), allowing us to investigate the emergence of social norms among individual agents. We aim to identify social norms by designing prompts and utilizing LLMs on textual information related to the environment setup and the observations of LLM-based agents. Using the OpenAI Chat API powered by GPT-4.0, we conduct experiments to simulate interactions and evaluate the performance of LLM-based agents in two driving scenarios: unsignalized intersection and highway platoon. The results show that LLM-based agents can handle dynamically changing environments in Markov games, and social norms evolve among LLM-based agents in both scenarios. In the intersection game, LLM-based agents tend to adopt a conservative driving policy when facing a potential car crash. The advantage of LLM-based agents in games lies in their strong operability and analyzability, which facilitate experimental design.

Figures (10)

• Figure 1: LLM for autonomous driving games
• Figure 2: Game scenarios
• Figure 3: Prompt design
• Figure 4: Responses (temperature = 2.0)
• Figure 5: Prompts and responses - Scenario 1