面向高密度交通场景的自动驾驶运动规划

doi:10.3778/j.issn.1002-8331.2304-0208

摘要/Abstract

摘要： 针对现有自动驾驶运动规划方法在提取状态信息时忽略了周边车辆交互、在高密度交通场景下规划效果不理想的问题，提出了一种联合图神经网络和深度强化学习的运动规划模型。基于图神经网络提出了一种自动驾驶车辆的交互式特征提取方法，再基于双延迟深度确定性策略梯度（twin delayed deep deterministic policy gradient，TD3）算法从交互式特征中预测车辆的动作，从而实现运动规划。将所提模型与目前自动驾驶运动规划模型LSTM+TD3、TD3及深度确定性策略梯度（deep deterministic policy gradient，DDPG）模型进行对比，在PGDrive驾驶模拟器中进行训练和测试的实验结果表明，在高密度交通场景中，所提方法的训练奖励值和测试成功率相比于对比方法提升了约36%、43%、23%及13、19、53个百分点。表明所提方法能有效解决自动驾驶周边车辆的交互式信息感知问题，更好地实现自动驾驶运动规划。

关键词: 自动驾驶, 运动规划, 交互式特征, 图神经网络, 强化学习

Abstract: Aiming at the problem that the existing motion planning methods for autonomous driving ignore the interaction of surrounding vehicles when extracting state information and the bad planning effect in dense traffic scenarios, a motion planning model combined with graph neural network and deep reinforcement learning is proposed. Firstly, based on the graph neural network, an interactive feature representation method of self-driving vehicles is proposed to extract spatial interaction features of multiple traffic participants. In this case, a learning strategy for motion planning is designed based on twin delayed deep deterministic policy gradient (TD3), and the next action is predicted from the interactive features so as to realize motion planning. The proposed method is compared with the current motion planning model LSTM+TD3,TD3 and deep deterministic policy gradient (DDPG) for autonomous driving, in dense traffic scenarios, the experimental results of training and testing in the PGDrive driving simulator increased by 36%, 43%, 23% and 13, 19, 53?percentage points compared with the comparison method, which means the proposed method can effectively solve the problem of interactive information perception of surrounding vehicles for better motion planning of autonomous driving.

Key words: autonomous driving, motion planning, interactive feature, graph neural network, reinforcement learning

肖雨微, 姚溪子, 胡学敏, 罗显志. 面向高密度交通场景的自动驾驶运动规划[J]. 计算机工程与应用, 2024, 60(14): 114-122.

XIAO Yuwei, YAO Xizi, HU Xuemin, LUO Xianzhi. Motion Planning for Autonomous Driving in Dense Traffic Scenarios[J]. Computer Engineering and Applications, 2024, 60(14): 114-122.

参考文献

[1] 张斌, 何明, 陈希亮, 等. 改进DDPG算法在自动驾驶中的应用[J].计算机工程与应用, 2019, 55(10): 264-270.
ZHANG B, HE M, CHEN X L, et al. Self-driving via improved DDPG algorithm[J]. Computer Engineering and Applications, 2019, 55(10): 264-270.
[2] PAXTON C, RAMAN V, HAGER G D, et al. Combining neural networks and tree search for task and motion planning in challenging environments[C]//Proceedings of the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2017: 6059-6066.
[3] HU X, TANG B, CHEN L, et al. Learning a deep cascaded neural network for multiple motion commands prediction in autonomous driving[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(12): 7585-7596.
[4] CHEN L, HU X, TIAN W, et al. Parallel planning: a new motion planning framework for autonomous driving[J]. IEEE/CAA Journal of Automatica Sinica, 2018, 6(1): 236-246.
[5] SHI T, WANG P, CHENG X, et al. Driving decision and control for autonomous lane change based on deep reinforcement learning[J]. arXiv:1904.10171, 2019.
[6] KAI A, DEISENROTH M P, BRUNDAGE M, et al. Deep reinforcement learning: a brief survey[J].IEEE Signal Processing Magazine, 2017, 34(6): 26-38.
[7] KARAMAN S, WALTER M R, PEREZ A, et al. Anytime motion planning using the RRT[C]//Proceedings of the 2011 IEEE International Conference on Robotics and Automation, 2011: 1478-1483.
[8] FARAG W, SALEH Z. Tuning of PID track followers for autonomous driving[C]//Proceedings of the 2018 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies, 2018: 1-7.
[9] WANG P, GAO S, LI L, et al. Obstacle avoidance path planning design for autonomous driving vehicles based on an improved artificial potential field algorithm[J]. Energies, 2019, 12(12): 2342.
[10] BOJARSKI M, TESTA D D, DWORAKOWSKI D, et al. End to end learning for self-driving cars[J].arXiv: 1604. 07316, 2016.
[11] PAN Y, CHENG C A, SAIGOL K, et al. 2016 imitation learning for agile autonomous driving[J].International Journal of Robotics Research, 2020, 39(2/3): 286-302.
[12] HAWKE J, SHEN R, GURAU C, et al. Urban driving with conditional imitation learning[C]//Proceedings of the 2020 IEEE International Conference on Robotics and Automation, 2020: 251-257.
[13] WANG P, CHAN C Y, FORTELLE A. A reinforcement learning based approach for automated lane change maneuvers[C]//Proceedings of the 2018 IEEE Intelligent Vehicles Symposium, 2018: 1379-1384.
[14] KENDALL A, HAWKE J, JANZ D, et al. Learning to drive in a day[C]//Proceedings of the 2019 International Conference on Robotics and Automation, 2019: 8248-8254.
[15] BAUTISTA-MONTESANO R, GALLUZZI R, RUAN K, et al. Autonomous navigation at unsignalized intersections: a coupled reinforcement learning and model predictive control approach[J].Transportation Research, Part C: Emerging Technologies, 2022, 139(2):103662.
[16] FUJIMOTO S, HOOF H V, MEGER D.Addressing function approximation error in actor-critic methods[C]//Proceedings of the International Conference on Machine Learning, 2018: 1587-1596.
[17] LIAO Y, YU G, CHEN P, et al.Modelling personalised car-following behaviour:a memory-based deep reinforcement learning approach[J].Transportmetrica A: Transport Science, 2022: 1-29.
[18] HASSELT H V, GUEZ A, SILVER D.Deep reinforcement learning with double Q-learning[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2016:2094-2100.
[19] WANG Z, SCHAUL T, HESSEL M, et al. Dueling network architectures for deep reinforcement learning[C]//Proceedings of the International Conference on Machine Learning, 2016: 1995-2003.
[20] XU H, YANG G, YU F, et al.End-to-end learning of driving models from large-scale video datasets[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2174-2182.
[21] GORI M, MONFARDINI G, SCARSELLI F. A new model for learning in graph domains[C]//Proceedings of the IEEE International Joint Conference on Neural Networks, 2005: 729-734.
[22] KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[C]//Proceedings of the International Conference on Learning Representation, 2017.
[23] ZHAO L, SONG Y, ZHANG C, et al. T-GCN: a temporal graph convolutional network for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(9):3848-3858.
[24] LIU Q, LI X, YUAN S, et al. Decision-making technology for autonomous vehicles learning-based methods, applications and future outlook[C]//Proceedings of the 2021 IEEE International Intelligent Transportation Systems Conference, 2021: 30-37.
[25] 何倩倩, 孙静宇, 曾亚竹. 基于邻域感知图神经网络的会话推荐[J]. 计算机工程与应用, 2022, 58(9): 107-115.
HE Q Q, SUN J Y, ZENG Y Z. Neighborhood awareness graph neural networks for session-based recommendation[J].Computer Engineering and Applications, 2022, 58(9): 107-115.
[26] 刘鑫, 梅红岩, 王嘉豪, 等. 图神经网络推荐方法研究[J].计算机工程与应用, 2022, 58(10): 41-49.
LIU X, MEI H Y, WANG J H, et al. Research on graph neural network recommendation method[J]. Computer Engineering and Applications, 2022, 58(10): 41-49.
[27] VAN SEIJEN H, FATEMI M, ROMOFF J, et al. Hybrid reward architecture for reinforcement learning[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems, 2017: 5398-5408.
[28] LI Q, PENG Z, ZHANG Q, et al.Improving the generalization of end-to-end driving through procedural generation[J]. arXiv:2012.13681, 2020.
[29] MICALE D, COSTANTINO G, MATTEUCCI I, et al. CAHOOT: a context-aware vehicular intrusion detection system[C]//Proceedings of the 2022 IEEE International Conference on Trust, Security and Privacy in Computing and Communications, 2022: 1211-1218.
[30] PENG Z, LI Q, LIU C, et al. Safe driving via expert guided policy optimization[C]//Proceedings of the Conference on Robot Learning, 2022: 1554-1563.
[31] CHEN Y, HAN W, ZHU Q H, et al. Target-driven obstacle avoidance algorithm based on DDPG for connected autonomous vehicles[J]. EURASIP Journal on Advances in Signal Processing, 2022, 2022(1):1-22.
[32] CHEN L, HU X, TANG B, et al. Conditional DQN-based motion planning with fuzzy logic for autonomous driving[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(4): 2966-2977.
[33] LI Q, PENG Z, FENG L, et al. Metadrive: composing diverse driving scenarios for generalizable reinforcement learning[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022: 3461-3475.