摘要
近年来,一种基于虚拟轨道导向的胶轮承载新型轨道交通制式逐渐成为研究的热点,作为一种新兴的轨道交通车辆,虚拟轨道车辆在诸多关键技术上面临挑战。文章从电子导向及虚拟轨道技术、环境感知技术、路权分配策略及动力学承载特性四方面入手,对研究现状、技术原理进行了详细剖析,探讨了虚拟轨道车辆未来发展方向。研究结果表明,电子导向及虚拟轨道技术主要聚焦于在可靠导向信息的基础上,转向系统最优控制策略的选择,传统控制策略可靠性高,对于多因素、多变量的控制系统,基于神经网络的控制算法具有更大优势,但在普适性方面仍需提高;视觉、激光、微波等多源感知技术可以为车辆提供大量的环境信息数据,关键数据的冗余、互补及筛选是虚拟轨道车辆环境感知的基础,实现感知的实时性和高鲁棒性是车辆环境感知的追求目标,结合物联网、机器学习等新兴技术建立更加高效的多源算法融合处理方式或将成为环境感知技术研究热点;路权分配策略旨在有限道路资源下,优化提高虚拟轨道车辆和社会车辆的通行效率,当前主要基于车辆实时动态信息反馈至交通信号控制平台,实现虚拟轨道车辆的通行,道路信息、环境信息及出行规律信息的数据挖掘为路权分配策略提供进一步优化空间;由于虚拟轨道车辆结构的新颖性,基于胎地耦合、机电耦合、车间耦合等多体动力学模型被提出,车辆运行性能、控制策略匹配关系、结构参数优化得到检验,未来大系统的耦合关系和敏感因素需要重点考虑。
Recent years have seen the development of a new type of rail transportation system based on virtual rail-guided rubber wheels,which has become a research hotspot.As an emerging rail vehicle,virtual rail vehicles face numerous technological challenges.This paper analyzes the research status,technical principles,and future development trend of virtual rail vehicles from four aspects:electronic guidance and virtual rail technology,environment sensing technology,right-of-way allocation strategy,and dynamics-bearing characteristics.The research results show that electronic guidance and virtual rail technology mainly focuses on selecting the optimal control strategy for the steering system on the basis of reliable guidance information.The traditional control strategy is highly reliable,and for the multifactor and multivariable control system,neural network-based control algorithms have greater advantages.However,they still need to be improved in terms of general applicability;Multi-source sensing techniques,such as vision,laser,microwave,etc.,can provide vehicles with a large amount of context-aware information.Redundancy,complementarity and screening of key data are the basis for virtual rail vehicles to sense the context;real-time sensing and high robustness are its goals.Combining the Internet of Things,machine learning and other emerging technologies to establish a more efficient multi-source algorithmic fusion processing may become a research hotspot for context-aware technology;The right-of-way allocation strategy is aimed at improving the passage efficiency of virtual rail vehicles and other vehicles with limited road resources.At present,the real-time dynamic information of vehicles is mainly fed back to the traffic signal control platform in order to carry out data mining of the passage information of virtual rail vehicles,road information,environmental information and travel pattern information,and provide further optimization space for the right-of-way allocation strategy;Due to the novelty of the virtual rail vehicle structure,a multi-body dynamics model based on wheel-ground coupling,electromechanical coupling,and vehicle-vehicle coupling is proposed,and the vehicle operational performance,control strategy matching relationship,and structural parameter optimization are examined.The coupling relationship and sensitivities of the system require further consideration.
作者
张弛
朱涛
张卫华
范晓达
黄文杰
ZHANG Chi;ZHU Tao;ZHANG Weihua;FAN Xiaoda;HUANG Wenjie(State Key Laboratory of Rail Transit Vehicle System,Southwest Jiaotong University,Chengdu 610031,China;CRRC Nanjing Puzhen Rolling Stock Co.,Ltd.,Nanjing 210031,China)
出处
《铁道车辆》
2024年第5期1-9,共9页
Rolling Stock
基金
国家重点研发计划项目(2021YFB3400703)
四川省杰出青年基金项目(2022JDJQ0025)
四川省国际科技创新合作项目(2022YFH0075)。
关键词
车辆工程
虚拟轨道车辆
电子导向
虚拟轨道
环境感知
路权分配
动力学性能
vehicle engineering
virtual rail vehicle
electronic guidance
virtual rail
context awareness
right-of-way allocation
dynamics performance
