A Decision Planning Method for Unstructured Road Scenarios Based on Deep Reinforcement Learning
Main Article Content
Abstract
The rapid development of autonomous driving has intensified research on decision-making in unstructured road scenarios. Conventional rule-based methods often suffer from poor adaptability, limited efficiency gains, and inadequate economic performance in such environments. This paper presents a Deep Q-Network (DQN)-based approach that defines observation states and decision actions, with a reward function incorporating efficiency, economy, safety, and comfort. Simulations in a mining road environment show that the method outperforms traditional approaches, enhancing decision-making capabilities in unstructured scenarios and offering new perspectives for autonomous driving in complex environments.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
Xiong Lu, Kang Yuchen, Zhang Peizhi, et al. Research on the Behavior Decision-making System of Unmanned Vehicles [J]. Automotive Technology, 2018(8): 9..
Hubmann C, Becker M, Althoff D, et al. Decision making for autonomous driving considering interaction and uncertain prediction of surrounding vehicles[C]//2017 IEEE intelligent vehicles symposium (IV). IEEE, 2017: 1671-1678.
Yurtsever E, Lambert J, Carballo A, et al. A survey of autonomous driving: Common practices and emerging technologies[J]. IEEE access, 2020, 8: 58443-58469.
Wang Xiaoyuan, Yang Xinyue. Research on Driving Behavior Decision Mechanism Based on Decision Tree [J]. Journal of System Simulation, 2008, 20(2): 6.
Aksjonov A, Kyrki V. Rule-based decision-making system for autonomous vehicles at intersections with mixed traffic environment[C]//2021 IEEE International Intelligent Transportation Systems Conference (ITSC). IEEE, 2021: 660-666.
Bouchard F, Sedwards S, Czarnecki K. A rule-based behaviour planner for autonomous driving[C]//International Joint Conference on Rules and Reasoning. Cham: Springer International Publishing, 2022: 263-279.
Xu C, Zhao W, Liu J, et al. An integrated decision-making framework for highway autonomous driving using combined learning and rule-based algorithm[J]. IEEE Transactions on Vehicular Technology, 2022, 71(4): 3621-3632.
Talebpour A, Mahmassani H S, Hamdar S H. Modeling lane-changing behavior in a connected environment: A game theory approach[J]. Transportation Research Procedia, 2015, 7: 420-440.
Aksjonov A, Kyrki V. Rule-based decision-making system for autonomous vehicles at intersections with mixed traffic environment[C]//2021 IEEE International Intelligent Transportation Systems Conference (ITSC). IEEE, 2021: 660-666.
Hwang S, Lee K, Jeon H, et al. Autonomous vehicle cut-in algorithm for lane-merging scenarios via policy-based reinforcement learning nested within finite-state machine[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(10): 17594-17606.
Kiran B R, Sobh I, Talpaert V, et al. Deep reinforcement learning for autonomous driving: A survey[J]. IEEE transactions on intelligent transportation systems, 2021, 23(6): 4909-4926.
Xia Wei, Li Huiyun. Autonomous Driving Strategy Learning Method Based on Deep Reinforcement Learning [J]. Ensemble Technology, 2017, 6(3): 12.
Arulkumaran K, Deisenroth M P, Brundage M, et al. Deep reinforcement learning: A brief survey[J]. IEEE Signal Processing Magazine, 2017, 34(6): 26-38.
Wu J, Yang H, Yang L, et al. Human-guided deep reinforcement learning for optimal decision making of autonomous vehicles[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2024.
Yuan M, Shan J, Mi K. Deep reinforcement learning based game-theoretic decision-making for autonomous vehicles[J]. IEEE Robotics and Automation Letters, 2021, 7(2): 818-825.
Liao J, Liu T, Tang X, et al. Decision-making strategy on highway for autonomous vehicles using deep reinforcement learning[J]. IEEE Access, 2020, 8: 177804-177814.
Shi Bowen. Research on humanized autonomous driving behavior decision-making based on deep reinforcement learning[D]. Jilin University, 2022.
Wang Tinghan, Luo Yugong, Liu Jinxin, et al. End-to-end autonomous driving strategy based on deep deterministic policy gradient algorithm considering state distribution[J]. Journal of Tsinghua University: Science and Technology, 2021.