都市快轨交通

融合大模型与近端策略优化的列车自动驾驶智能控制研究

PDF下载

高凯 ¹^,² , 朱力 ¹ , 苗佳 ¹ , 刘磊 ¹

都市快轨交通 | 人工智能 2026,39(4): 114-122

收起

都市快轨交通 |人工智能 2026 , 39 (4) : 114 -122

融合大模型与近端策略优化的列车自动驾驶智能控制研究

全屏

高凯¹^,², 朱力¹, 苗佳¹, 刘磊¹

作者信息

^1．北京交通大学自主运行技术研究所，北京 100044

^2．广西交控智维科技发展有限公司，南宁 530200

作者简介:

高凯，男，博士研究生，从事智能列车控制、智能运维等研究，25115018@bjtu.edu.cn

An Intelligent Automatic Train OperationTrain Control Method Integrating Large Language Models and Proximal Policy Optimization

Kai Gao¹^,², Li Zhu¹, Jia Miao¹, Lei Liu¹

Affiliations

^1.Institute of Autonomous Operation Technology, Beijing Jiaotong University, Beijing 100044

^2.Guangxi Traffic Control Zhiwei Technology Development Co., Ltd., Nanning 530200

doi: 10.3969/j.issn.1672-6073.2026.04.013

文章导航

摘要

收起

在城市轨道交通智能化发展背景下，列车自动驾驶需在节能与准点性之间实现有效权衡，而现有调度规章多以自然语言形式表达，难以直接用于控制建模。为此，本文提出一种融合大语言模型（large language model, LLM）与近端策略优化（proximal policy optimization, PPO）的列车自动驾驶控制方法。该方法利用LLM对调度规章与运行经验进行语义解析，并在知识图谱（knowledge graph, KG）约束下自动构建包含状态空间、动作集合、奖励函数与安全约束的马尔可夫决策过程（markov decision process, MDP）模型。在此基础上，引入动作屏蔽机制与奖励权重自适应调节策略，采用PPO算法在仿真环境中训练安全可控的自动驾驶策略。仿真结果表明：与传统规则控制、模型预测控制（model predictive control, MPC）和深度Q网络（deep Q-network, DQN）方法相比，本文方法的平均牵引能耗分别降低约2.5%、1.5%和1.0%，平均晚点时间分别降低约13.7%、19.6%和16.3%，且平均jerk值降至0.59m/s³，验证了LLM语义建模与强化学习相结合在列车自动驾驶控制中的有效性。进一步的半实物仿真结果表明，知识图谱校验能够降低语义解析中的规则遗漏与物理不一致风险，动作屏蔽机制可将限速和安全间隔等硬约束嵌入策略执行过程，奖励权重调节机制能够依据能耗、晚点和舒适度指标动态调整多目标权衡。所提方法形成了“语义建模—策略优化—反馈调节”的闭环，为自然语言调度知识参与列车自动驾驶控制建模提供了可解释的实现思路。

关键词

轨道交通 / 列车自动驾驶 / 大语言模型 / 近端策略优化 / 语义建模 / 多目标控制

Abstract

收起

With advancement of intelligent urban rail transit, automatic train operation needs to achieve an effective balance between energy efficiency and punctuality. However, existing dispatching regulations are mostly expressed in natural language, making them difficult to directly incorporate into control modeling. To address this problem, this paper proposes an automated train operation control method integrating large language models (LLMs) and proximal policy optimization (PPO). The proposed method employs an LLMs to perform semantic parsing of dispatching regulations and operational experience, and automatically constructs a Markov decision process (MDP) model under the constraints of a knowledge graph (KG). The generated MDP includes the state space, action set, reward function, and safety constraints. On this basis, an action masking mechanism and an adaptive reward-weight adjustment strategy are introduced, and PPO is used to train a safe and controllable automated train operation policy in a simulation environment. Simulation results show that, compared with conventional rule-based control, Model Predictive Control (MPC), and Deep Q-Network (DQN), the proposed method reduces the average traction energy consumption by approximately 2.5%, 1.5%, and 1.0%, respectively, and reduces the average delay time by approximately 13.7%, 19.6%, and 16.3%, respectively. In addition, the average jerk is reduced to 0.59 m/s³. These results verify the effectiveness of combining LLM-based semantic modeling with reinforcement learning for automated train operation control. Further hardware-in-the-loop tests indicate that KG-based consistency checking can reduce rule omissions and physical inconsistencies in semantic parsing. The action masking mechanism embeds hard constraints, including speed limits and safe headway distances, into policy execution, while the adaptive reward-weight mechanism adjusts the trade-off among energy consumption, punctuality, and ride comfort according to operational feedback. The proposed method therefore establishes a closed-loop framework of semantic modeling, policy optimization, and feedback adjustment, providing an interpretable approach for incorporating natural-language dispatching knowledge into automated train operation control.

Key words

rail transit / automated train operation / large language model / proximal policy optimization / semantic modeling / multi-objective control

引用本文

高凯, 朱力, 苗佳, 刘磊. 融合大模型与近端策略优化的列车自动驾驶智能控制研究. 都市快轨交通, 2026 , 39 (4) : 114 -122 . DOI: 10.3969/j.issn.1672-6073.2026.04.013

Kai Gao, Li Zhu, Jia Miao, Lei Liu. An Intelligent Automatic Train OperationTrain Control Method Integrating Large Language Models and Proximal Policy Optimization[J]. Urban Rapid Rail Transit, 2026 , 39 (4) : 114 -122 . DOI: 10.3969/j.issn.1672-6073.2026.04.013

基金

收起

北京市联合基金(L251011)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

2026年第39卷第4期

PDF下载

引用本文

BibTeX

文章信息

doi: 10.3969/j.issn.1672-6073.2026.04.013

接收时间：2025-07-21
首发时间：2026-09-20

补充材料

相关文章

文章信息

作者

出版历史

收稿日期：2025-07-21
修回日期：2026-04-13

基金

北京市联合基金(L251011)

作者信息

^1．北京交通大学自主运行技术研究所，北京 100044

^2．广西交控智维科技发展有限公司，南宁 530200

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/dskgjt/CN/10.3969/j.issn.1672-6073.2026.04.013

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT