融合车辆轨迹预测的高速匝道自动驾驶车辆汇入控制方法

doi:10.16097/j.cnki.1009-6744.2026.05.016

交通运输系统工程与信息 ›› 2026, Vol. 26 ›› Issue (5): 184-196.DOI: 10.16097/j.cnki.1009-6744.2026.05.016

• 智能交通系统与信息技术 • 上一篇下一篇

融合车辆轨迹预测的高速匝道自动驾驶车辆汇入控制方法

杜凯^1a，刘会洋^1a ，何启航²，于艳波^1a,3，陈丹^1a，宋京妮^*1b

1. 长安大学，a.机器人与具身智能学院，b.运输工程学院，西安710064；2.中车青岛四方车辆研究所有限公司，山东青岛266031；3.中国公路工程咨询集团有限公司，北京100093

收稿日期:2026-06-15 修回日期:2026-07-28 接受日期:2026-08-06 出版日期:2026-10-25 发布日期:2026-10-21
作者简介:杜凯（1984—），男，陕西西安人，副教授，博士。
基金资助:
陕西省自然科学基础研究计划资助项目(2026JC-YBMS-0569, 2026JC-YBMS-0822)。

Highway Ramp Merging Control Method for Autonomous Vehicles Integrating Vehicle Trajectory Prediction

DU Kai^1a, LIU Huiyang^1a, HE Qihang², YU Yanbo^1a,3, CHEN Dan^1a, SONG Jingni^*1b

1a. School of Robotics and Embodied Intelligence, 1b. School of Transportation Engineering, Chang'an University, Xi'an 710064, China; 2. CRRC Qingdao Sifang Rolling Stock Research Institute Co Ltd, Qingdao 266031, Shandong China; 3. China Highway Engineering Consulting Group Co Ltd, Beijing 100093, China

Received:2026-06-15 Revised:2026-07-28 Accepted:2026-08-06 Online:2026-10-25 Published:2026-10-21
Supported by:
Natural Science Basic Research Program of Shaanxi(2026JC-YBMS-0569, 2026JC-YBMS-0822)。

摘要/Abstract

摘要： 高速公路匝道汇入区因车辆速度差大、车流交织复杂，在交通流持续增长的背景下，已成为通行安全与效率的关键瓶颈。针对传统匝道汇入控制方法因缺乏前瞻性预判而导致汇入时机与安全性难以兼顾的问题，本文提出一种融合轨迹预测的自动驾驶车辆匝道汇入控制方法。该方法将汇入决策指令与自车未来轨迹预测信息输入纵向控制模块，通过融合宏观汇入意图与微观未来位置演化信息，生成兼顾安全和效率的优化加速度控制量。为确保准确反映真实交通环境下的驾驶特性，本文基于exiD公开数据集与西安绕城高速实地采集数据，构建真实汇入场景数据集，并以此为基础分别训练Autoformer轨迹预测模型与GWO-XGBoost（Grey Wolf Optimizer-eXtreme Gradient Boosting）横向汇入决策模型。在此基础上，构建融合Autoformer、GWO-XGBoost与DDQN（Double Deep Q-Network）的一体化汇入控制模型。仿真实验结果表明，所提模型平均汇入成功率达98.21%，显著优于XGBoost-DDQN、LC2013-DDQN和DDQN等对比模型，汇入效率与安全性方面也表现更优。进一步通过SHAP（Shapley Additive Explanations）可解释性分析发现，匝道汇入车辆当前位置、预测位置以及主干道车辆位置是影响控制输出的关键因素，验证所提模型能够有效利用时序信息实现更合理的汇入决策。

关键词: 智能交通, 汇入控制, 强化学习, 轨迹预测, 无人机数据, 建模仿真

Abstract: The ramp merging area of expressways has become a key bottleneck for traffic safety and efficiency due to the large speed difference and complex weaving of vehicle flows, especially against the backdrop of continuous traffic growth. A lane changing control method for autonomous vehicles at highway on-ramp merging areas is proposed in this paper. It aims to address the critical safety and efficiency bottlenecks caused by large speed differences and complex weaving under increasing traffic demand. The lack of predictive capability of traditional ramp merging control methods makes it hard to coordinate merging timing and safety, so the proposed method embeds merging decisions and predicted vehicle trajectories into the longitudinal control module. By combining the macro-level merging intentions with micro-level future position information, it generates an optimized acceleration profile that balances safety and efficiency. To reflect the real driving behavior, this paper builds a realistic merging scenario dataset with the public exiD dataset and field data from the Xi'an Ring Expressway. This dataset is used to train the Autoformer-based trajectory prediction model and the GWO-XGBoost lateral merging decision model. On this basis, an integrated merging control framework is developed by combining Autoformer, GWO-XGBoost, and DDQN. Simulation results show that the average merging success rate of proposed method reaches 98.21%, which is markedly higher than those of XGBoost-DDQN, LC2013-DDQN, and DDQN. The proposed method also improves merging efficiency and safety. SHAP-based interpretability analysis further shows that the current and predicted positions of the merging vehicle, together with the positions of mainline vehicles, are the dominant factors to shape the control outputs. This confirms that the model can effectively use temporal information to support more rational merging decisions.

Key words: intelligent transportation, ramp merging control, reinforcement learning, trajectory prediction, Unmanned Aerial Vehicle（UAV）data, modeling and simulation

中图分类号:

U471.15

杜凯, 刘会洋, 何启航, 于艳波, 陈丹, 宋京妮. 融合车辆轨迹预测的高速匝道自动驾驶车辆汇入控制方法[J]. 交通运输系统工程与信息, 2026, 26(5): 184-196.

DU Kai, LIU Huiyang, HE Qihang, YU Yanbo, CHEN Dan, SONG Jingni. Highway Ramp Merging Control Method for Autonomous Vehicles Integrating Vehicle Trajectory Prediction[J]. Journal of Transportation Systems Engineering and Information Technology, 2026, 26(5): 184-196.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/10.16097/j.cnki.1009-6744.2026.05.016

http://www.tseit.org.cn/CN/Y2026/V26/I5/184

参考文献

[1] CUI J X, ZHAO B Y, QU M C. An integrated lateral and longitudinal decision-making model for autonomous driving based on deep reinforcement learning[J]. Journal of Advanced Transportation, 2023, 2023(1): 1513008 1513021.

[2]TREIBER M, HENNECKE A, HELBING D. Congested traffic states in empirical observations and microscopic simulations[J]. Physical Review E, 2000, 62(2): 1805 1824.

[3]KESTING A, TREIBER M, HELBING D. General lane changing model MOBIL for car-following models[J]. Transportation Research Record, 2007, 1999: 86-94.

[4]乔良,鲍泓,玄祖兴,等.基于强化学习的无人驾驶匝道汇入模型[J]. 计算机工程,2018,44(7): 20-24. [QIAO L, BAO H, XUAN Z X, et al. Autonomous driving ramp merging model based on reinforcement learning[J]. Computer Engineering, 2018, 44(7): 20-24.]

[5]韩磊,张轮,郭为安.混合交通流环境下基于改进强化学习的可变限速控制策略[J].交通运输系统工程与信息, 2023, 23(3): 110-122. [HAN L, ZHANG L, GUO W A. Variable speed limit control based on improved dueling double deep Q network under mixed traffic environment[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(3): 110-122.]

[6]HASSELT H V, GUEZ A, SILVER D. Deep reinforcement learning with double Q-learning[J/OL]. arXiv, (2015-12-08) [2026-03-05]. https://arxiv.org/abs/ 1509.06461.

[7] NI H Y, YU G Z, CHEN P, et al. An integrated framework of lateral and longitudinal behavior decision making for autonomous driving using reinforcement learning[J]. IEEE Transactions on Vehicular Technology, 2024, 73(7): 9706-9720.

[8]LIU Q, TANG Y J, LI X Y, et al. MV-STGHAT: Multi view spatial-temporal graph hybrid attention network for decision-making of connected and autonomous vehicles [J]. IEEE Transactions on Vehicular Technology, 2025, 74(4): 5594-5609.

[9] 徐杰,裴晓飞,杨波,等.融合车辆轨迹预测的学习型自动驾驶决策[J]. 汽车安全与节能学报,2022,13(2): 317-324. [XU J, PEI X F, YANG B, et al. Learning based automatic driving decision-making integrated with vehicle trajectory prediction[J]. Journal of Automotive Safety and Energy, 2022, 13(2):317-324.]

[10] 谷梦路,葛振振,王畅,等.智能网联车辆加速车道类人化汇入控制研究[J]. 中国公路学报, 2024, 37(3): 134-146. [GU M L, GE Z Z, WANGC,et al. Human like merging control of intelligent connected vehicles on the acceleration lane[J]. China Journal of Highway and Transport, 2024, 37(3): 134-146.]

[11] KHERROUBI Z E A, AKNINE S, BACHA R. Novel decision-making strategy for connected and autonomous vehicles in highway on-ramp merging[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(8): 12490-12502.

[12] 陈英,朱加勋,田毅,等.基于K-Means-XGBoost-SHAP 集成化框架的交通事故严重程度致因分析[J].中国公路学报,2025, 38(5): 209-223. [CHEN Y, ZHU J X, TIAN Y, et al. Analysis of factors contributing to traffic accident severity based on an integrated framework of K Means-XGBoost-SHAP[J]. China Journal of Highway and Transport, 2025, 38(5): 209-223.]

[13] LI M, CUI Z H, WANG Y L, et al. An explainable Q learning method for longitudinal control of autonomous vehicles[J]. IEEE Transactions on Intelligent Transportation Systems, 2025, 26(3): 4214-4218.

[14] WU H, XU J, WANG J, et al. Autoformer: Decomposition transformers with auto correlation for long-term series forecasting[J]. Advances in Neural Information Processing Systems, 2021, 34: 22419-22430.

[1]	奇格奇, 董镇滔, 栾正凯, 关伟. 驾驶人脑电研究综述：状态感知、认知解码与人机交互协同[J]. 交通运输系统工程与信息, 2026, 26(5): 1-21.
[2]	杨静文, 齐旭东, 王萍, 夏新海. 智能网联环境下多簇交通波消散的车辆群体协同控制方法[J]. 交通运输系统工程与信息, 2026, 26(5): 79-88.
[3]	陈坚, 胡功明, 简洪波, 傅志妍. 基于社交媒体数据的交通突发事件提取方法[J]. 交通运输系统工程与信息, 2026, 26(5): 122-131.
[4]	翁金贤 , 郭豪飞, 丁海峰. 基于驾驭工程的临时集卡专用道作业区粒流协同控制方法[J]. 交通运输系统工程与信息, 2026, 26(5): 173-183.
[5]	邹复民, 江林文, 蔡祈钦, 廖律超, 赖树坤. 考虑轨迹语义与时空关联的高速公路行程时间预测方法[J]. 交通运输系统工程与信息, 2026, 26(5): 227-237.
[6]	张建华, 陈云帅. 面向时空约束的网联车匝道合流分层协同控制[J]. 交通运输系统工程与信息, 2026, 26(5): 238-247.
[7]	程国柱, 罗天健, 王文志, 梁嘉祺, 冯天军. 城市快速路施工区上游智能网联汽车协同换道方法研究[J]. 交通运输系统工程与信息, 2026, 26(5): 386-402.
[8]	薛晴婉, 张先喆, 杨林溢, 闫睿轩, 王力. 辅助驾驶环境下融合驾驶行为熵的交叉口碰撞风险预测[J]. 交通运输系统工程与信息, 2026, 26(4): 112-123.
[9]	贾彦峰, 李富贵, 曲大义, 胡星辰, 李妍妍. 人工势场增强的网联异构车队分布式非线性纵横向耦合控制[J]. 交通运输系统工程与信息, 2026, 26(4): 148-161.
[10]	何永明, 那浩轩, 金玉凤, 张龙龙. 考虑交通流态势的自动驾驶车辆换道轨迹模型[J]. 交通运输系统工程与信息, 2026, 26(4): 162-172.
[11]	王昊, 陈旭梅. 考虑异质驾驶行为的智能网联车辆超车应对策略研究[J]. 交通运输系统工程与信息, 2026, 26(4): 173-187.
[12]	马千里, 焦乙恩, 周屹恒, 贾鹏. 机场混编摆渡车队高峰小时服务调度优化[J]. 交通运输系统工程与信息, 2026, 26(4): 321-331.
[13]	程华, 张扬, 向飞, 马吉震, 唐明杰. 基于深度强化学习的机场地面保障调度方法[J]. 交通运输系统工程与信息, 2026, 26(4): 380-390.
[14]	黄凯, 谢子俊, 李濠宇, 刘芯彤, 刘志远. 基于锚点分类改进的自动驾驶车道线检测研究[J]. 交通运输系统工程与信息, 2026, 26(3): 14-24.
[15]	王正礼, 唐子墨, 郑振杰. 融合检索增强生成与思维链的交通事故责任认定大模型构建[J]. 交通运输系统工程与信息, 2026, 26(3): 83-92.

融合车辆轨迹预测的高速匝道自动驾驶车辆汇入控制方法

Highway Ramp Merging Control Method for Autonomous Vehicles Integrating Vehicle Trajectory Prediction

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics