LearningFlow: Automated Policy Learning Workflow for Urban Driving with Large Language Models

TL;DR

LearningFlow is an automated workflow that uses multiple large language models to improve reinforcement learning for urban driving, reducing manual effort and enhancing policy performance in complex environments.

Contribution

This paper introduces LearningFlow, a novel LLM-based framework that automates reward and curriculum generation for RL in urban driving, improving efficiency and generalization.

Findings

Outperforms existing methods in reward and curriculum generation

Achieves higher driving policy performance in CARLA simulator

Demonstrates robust generalization across tasks and RL algorithms

Abstract

Recent advancements in reinforcement learning (RL) demonstrate the significant potential in autonomous driving. Despite this promise, challenges such as the manual design of reward functions and low sample efficiency in complex environments continue to impede the development of safe and effective driving policies. To tackle these issues, we introduce LearningFlow, an innovative automated policy learning workflow tailored to urban driving. This framework leverages the collaboration of multiple large language model (LLM) agents throughout the RL training process. LearningFlow includes a curriculum sequence generation process and a reward generation process, which work in tandem to guide the RL policy by generating tailored training curricula and reward functions. Particularly, each process is supported by an analysis agent that evaluates training progress and provides critical insights to the generation agent. Through the collaborative efforts of these LLM agents, LearningFlow automates policy learning across a series of complex driving tasks, and it significantly reduces the reliance on manual reward function design while enhancing sample efficiency. Comprehensive experiments are conducted in the high-fidelity CARLA simulator, along with comparisons with other existing methods, to demonstrate the efficacy of our proposed approach. The results demonstrate that LearningFlow excels in generating rewards and curricula. It also achieves superior performance and robust generalization across various driving tasks, as well as commendable adaptation to different RL algorithms.