Facilitating Autonomous Driving Tasks With Large Language Models

We explore how large language models (LLMs) can expedite and automate the learning process for autonomous driving tasks. This involves harnessing LLM knowledge to shape a learning framework and utilizing LLMs to guide the learning process. We conduct a case study to demonstrate LLMs’ ability to export driving rules. LLM outputs may not be entirely reliable for the direct handling of driving decisions due to potential inaccuracies and inconsistencies. To address these issues, we propose integrating LLM knowledge with statistical learning. This enables LLMs to export task-specific knowledge as symbolic rules, forming the initial learning structure. Rule weights are calculated based on statistical salience derived from training data, resulting in a set of weighted rules for robust decision making. Furthermore, this set of weighted rules preserves strong semantics, allowing LLMs to comprehend and make modifications based on varying needs. Simulations using a highway driving simulator validate the effectiveness of our approach.

Authors

• Mengyao Wu China University of Petroleum (East China)
• F. Richard Yu Carleton University
• Peter Xiaoping Liu
• Ying He Nanyang Technological University

Keywords

• Safety
• Decision making
• Autonomous vehicles
• Reinforcement learning
• Statistical learning
• Intelligent systems
• Chatbots
• Large language models
• Autonomous driving
• Decision-making
• Machine Learning
• Training Data
• Learning Process
• Machine Learning Methods
• Learning Task
• Set Of Rules
• Hallucinations
• Decision Rules
• Decision Model
• Reward Function
• Design Expert
• Learning Rule
• Understanding Of The World
• Imitation Learning
• Robust Decision
• Potential Use Cases
• Reinforcement Learning Process
• Small Amount Of Data
• Deep Reinforcement Learning
• Lane Change
• Transition State
• Output Label
• Noisy Data
• Knowledge Sharing