控制与信息技术Issue(5)：46-51,6.DOI:10.13889/j.issn.2096-5427.2025.05.006
延长卡车队列行驶里程的周期性重排序算法
Periodic Reordering Algorithm for Extending Driving Ranges of Truck Platoons
摘要
Abstract
To address the limitations of fixed platoon strategies in cooperation control within truck platoons—specifically their inability to effectively respond to real-time variations in vehicle operating states,which leads to unbalanced fuel consumption and restricted driving ranges—this study proposes a dynamic decision-making framework based on deep reinforcement learning(DRL)for periodic reordering.This framework employs a deep Q-network(DQN)as its core algorithm to construct a dynamic learning mechanism through continuous agent—environment interactions.Utilizing a multi-dimensional environmental perception model and an adaptive decision-making module,the agent autonomously determines the optimal timing for platoon reordering according to the current fuel levels of individual trucks,thereby achieving dynamic fuel balance within platoons.This approach overcomes the technical bottlenecks of traditional fixed platoon strategies in real-time responsiveness and energy distribution,effectively mitigating the premature fuel depletion of individual trucks that limits platoon driving ranges.Experimental results demonstrated that the proposed framework increased platoon driving ranges by approximately 5.0%under identical transportation conditions,with minimal influence from the period variable.The findings provide a novel technical pathway for energy-efficient platoon control in complex traffic environments,and the core concept can be further extended to multi-object cooperation optimization applications.关键词
卡车编队行驶/车队控制/深度强化学习/多目标协同优化/深度Q网络Key words
truck platooning/platoon control/deep reinforcement learning/multi-object cooperation optimization/deep Q-network(DQN)分类
交通运输引用本文复制引用
杨智凯,郭少攀,刘淼,肖龙..延长卡车队列行驶里程的周期性重排序算法[J].控制与信息技术,2025,(5):46-51,6.基金项目
国家自然科学基金项目(62203212) （62203212）
江苏省高等学校基础科学(自然科学)研究项目(25KJB120003) （自然科学）
