Human-centric Reward Optimization for Reinforcement Learning-based Automated Driving using Large Language Models
This addresses the problem of achieving flexible and precise human-like driving behavior in automated systems, representing an incremental advance in RL-based AD.
The paper tackles the challenge of making Reinforcement Learning-based Automated Driving agents more human-like by using large language models to optimize reward functions, resulting in improved anthropomorphic behavior and performance.
One of the key challenges in current Reinforcement Learning (RL)-based Automated Driving (AD) agents is achieving flexible, precise, and human-like behavior cost-effectively. This paper introduces an innovative approach that uses large language models (LLMs) to intuitively and effectively optimize RL reward functions in a human-centric way. We developed a framework where instructions and dynamic environment descriptions are input into the LLM. The LLM then utilizes this information to assist in generating rewards, thereby steering the behavior of RL agents towards patterns that more closely resemble human driving. The experimental results demonstrate that this approach not only makes RL agents more anthropomorphic but also achieves better performance. Additionally, various strategies for reward-proxy and reward-shaping are investigated, revealing the significant impact of prompt design on shaping an AD vehicle's behavior. These findings offer a promising direction for the development of more advanced, human-like automated driving systems. Our experimental data and source code can be found here