期刊
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
卷 23, 期 6, 页码 5068-5078出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TITS.2020.3046646
关键词
Autonomous vehicles; Reinforcement learning; Probabilistic logic; Roads; Task analysis; Autonomous automobiles; Graphical models; Autonomous driving; deep reinforcement learning; probabilistic graphical model; interpretability
资金
- DENSO International at America
This article introduces an interpretable deep reinforcement learning method for end-to-end autonomous driving, which utilizes a sequential latent environment model for handling complex urban scenarios and significantly reducing the sample complexity of reinforcement learning. Comparative tests in a realistic driving simulator demonstrate that the method outperforms many baseline models including DQN, DDPG, TD3, and SAC in crowded urban environments.
Unlike popular modularized framework, end-to-end autonomous driving seeks to solve the perception, decision and control problems in an integrated way, which can be more adapting to new scenarios and easier to generalize at scale. However, existing end-to-end approaches are often lack of interpretability, and can only deal with simple driving tasks like lane keeping. In this article, we propose an interpretable deep reinforcement learning method for end-to-end autonomous driving, which is able to handle complex urban scenarios. A sequential latent environment model is introduced and learned jointly with the reinforcement learning process. With this latent model, a semantic birdeye mask can be generated, which is enforced to connect with certain intermediate properties in today's modularized framework for the purpose of explaining the behaviors of learned policy. The latent space also significantly reduces the sample complexity of reinforcement learning. Comparison tests in a realistic driving simulator show that the performance of our method in urban scenarios with crowded surrounding vehicles dominates many baselines including DQN, DDPG, TD3 and SAC. Moreover, through masked outputs, the learned model is able to provide a better explanation of how the car reasons about the driving environment.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据