AD4RL: Autonomous Driving Benchmarks for Offline Reinforcement Learning with Value-based Dataset
This work addresses the need for more realistic benchmarks in offline reinforcement learning for autonomous driving, though it is incremental as it focuses on dataset creation rather than novel algorithmic advances.
The paper tackles the lack of realistic datasets for offline reinforcement learning by providing 19 autonomous driving datasets, including real-world human driver data, and benchmarks seven algorithms across three driving scenarios, establishing a foundation for practical research.
Offline reinforcement learning has emerged as a promising technology by enhancing its practicality through the use of pre-collected large datasets. Despite its practical benefits, most algorithm development research in offline reinforcement learning still relies on game tasks with synthetic datasets. To address such limitations, this paper provides autonomous driving datasets and benchmarks for offline reinforcement learning research. We provide 19 datasets, including real-world human driver's datasets, and seven popular offline reinforcement learning algorithms in three realistic driving scenarios. We also provide a unified decision-making process model that can operate effectively across different scenarios, serving as a reference framework in algorithm design. Our research lays the groundwork for further collaborations in the community to explore practical aspects of existing reinforcement learning methods. Dataset and codes can be found in https://sites.google.com/view/ad4rl.