驶向智能未来:深度学习在轨道交通革新中的应用

doi:10.11896/jsjkx.240300099

摘要/Abstract

摘要： 目前,轨道交通凭借其便利性、高效性等特点,在城市交通中扮演着重要角色。然而,现有轨道交通系统的运行过程也存在着复杂的问题,客流预测、列车调度等环节仍采用人工方式,效率和准确率均较低,对系统性能造成了一定影响。近年来,深度学习蓬勃发展,其强大的特征提取与图像识别能力,也为轨道交通的自动化、智能化发展提供了更多的可能性。文中首先阐述了当前轨道交通在现实生活各种应用场景中面临的挑战;其次从轨道交通感知任务、预测任务、优化任务等方面分析了深度学习赋能轨道交通领域的主要应用;最后,从高精度和高鲁棒性的安全性检测,轻量级的轨道交通模型,全自动的轨道交通智能化运行,以及借助云计算、大数据的信息处理高效化4个方面展望了未来深度学习在轨道交通中的发展方向。

关键词: 深度学习, 轨道交通, 感知, 预测, 优化

Abstract: Nowadays,rail transit plays a crucial role in urban transportation due to its convenience and efficiency.However,the operation of existing rail transit systems faces complex challenges.Processes such as passenger flow prediction and train scheduling still rely on manual methods,leading to low efficiency and accuracy,which has a certain impact on the system’s performance.In recent years,with the flourishing development of deep learning,its powerful feature extraction and image recognition capabilities provide more possibilities for the automation and intelligence of rail transit.This paper first outlines the challenges faced by current rail transit in various real-life scenarios,and then analyzes the main applications of deep learning in the rail transit field,including perception tasks,prediction tasks,optimization tasks,etc.Finally,the future development direction of deep learning in rail transit is prospected from four aspects:high-precision and robust safety detection,lightweight rail transit models,fully automated intelligent operation of rail transit,and efficient information processing through cloud computing and big data.

Key words: Deep learning, Rail transit, Perception, Prediction, Optimization

中图分类号:

TP181

孙宇墨, 李昕航, 赵文杰, 朱力, 梁雅楠. 驶向智能未来:深度学习在轨道交通革新中的应用[J]. 计算机科学, 2024, 51(8): 1-10. https://doi.org/10.11896/jsjkx.240300099

SUN Yumo, LI Xinhang, ZHAO Wenjie, ZHU Li, LIANG Ya’nan. Driving Towards Intelligent Future:The Application of Deep Learning in Rail Transit Innovation[J]. Computer Science, 2024, 51(8): 1-10. https://doi.org/10.11896/jsjkx.240300099

参考文献

[1]LIN Y W,HSIEH C C,HUANG W H,et al.Railway track fasteners fault detection using deep learning[C]//2019 IEEE Eurasia Conference on IOT,Communication and Engineering(ECICE).IEEE,2019:187-190.
[2]ANWAR N,SHEN Z,WEI Q L,et al.YOLOv4 based deeplearning algorithm for defects detection and classification of rail surfaces[C]//2021 IEEE International Intelligent Transportation Systems Conference(ITSC).IEEE,2021:1616-1620.
[3]QIN Y L,HE D Q,JIN Z Z,et al.An improved deep learning algorithm for obstacle detection in complex rail transit environments[J].IEEE Sensors Journal,2024:24(3):4011-4022.
[4]RAKSHIT S,SANDEEP B S.Railway Track Fault Detectionusing Deep Neural Networks[C]//2022 IEEE 6th Conference on Information and Communication Technology(CICT).IEEE,2022:1-5.
[5]GUO W P,WANG H,MAO L N,et al.Research on DeepLearning-based Deraining Method of Catenary Images[C]//2021 IEEE 2nd China International Youth Conference on Electrical Engineering(CIYCEE).2021:1-5.
[6]DING X W,CAI X N,ZHANG Z Y,et al.Railway foreign object intrusion detection based on deep learning[C]//2022 International Conference on Computer Engineering and Artificial Intelligence(ICCEAI).IEEE,2022:735-739.
[7]ZHU K E,XUN P,LI W,et al.Prediction of passenger flow in urban rail transit based on big data analysis and deep learning[J].IEEE Access,2019,7:142272-142279.
[8]XIAO Z S,MAO B H,ZHANG T.Integrated predicting model for daily passenger volume of rail transit station based on neural network and Markov chain[C]//2018 IEEE 3rd International Conference on Cloud Computing and Big Data Analysis(ICCCBDA).IEEE,2018:578-583.
[9]WANG Z R,XIE X Z,CHEN L,et al.Intrusion detection and network information security based on deep learning algorithm in urban rail transit management system[J].IEEE Transactions on Intelligent Transportation Systems,2023,24(2):2135-2143.
[10]WANG Y J,DU X,LU Z H,et al.Improved LSTM-based time-series anomaly detection in rail transit operation environments[J].IEEE Transactions on Industrial Informatics,2022,18(12):9027-9036.
[11]ZHANG Y,HAN J Z,LIU J,et al.Safety prediction of rail tran-sit system based on deep learning[C]//2017 IEEE/ACIS 16th International Conference on Computer and Information Science(ICIS).IEEE,2017:851-856.
[12]ZHANG J L,CHEN F,SHEN Q.Cluster-based LSTM network for short-term passenger flow forecasting in urban rail transit[J].IEEE Access,2019,7:147653-147671.
[13]ZHANG Y,WANG Q R.Short term prediction of traffic flowbased on deep learning[C]//2020 International Conference on Virtual Reality and Intelligent Systems(ICVRIS).IEEE,2020:982-985.
[14]ZHU G Y,DING J C,WEI Y,et al.Two-stage OD flow prediction for emergency in urban rail transit[J].IEEE Transactions on Intelligent Transportation Systems,2023,25(1):920-928.
[15]ZHANG J L,CHEN F,CUI Z Y,et al.Deep learning architecture for short-term passenger flow forecasting in urban rail transit[J].IEEE Transactions on Intelligent Transportation Systems,2020,22(11):7004-7014.
[16]GAO H L,SHI J Q,BAO C Q,et al.Fault prediction of track circuit compensation capacitor based on MFO-LSTM[C]//2023 CAA Symposium on Fault Detection,Supervision and Safety for Technical Processes(SAFEPROCESS).IEEE,2023:1-5.
[17]LI C,CHEN J,XUE C J,et al.Simultaneous Multi-Spot Temperature Prediction of Traction Transformer in Urban Rail Transit Using Long Short-Term Memory Networks[J].IEEE Transactions on Transportation Electrification,2023,9(3):4552-4561.
[18]PENG C,SHANGGUAN W,CAI B G,et al.Fault predictionmethod for on-board equipment of CTCS based on Seq2Seq mo-del using LSTM[C]//2021 Global Reliability and Prognostics and Health Management(PHM-Nanjing).IEEE,2021:1-5.
[19]BAI Y H,LI H H,ZHAO S,et al.Remaining Useful Life Prediction of Wheel of Heavy-duty Railway Train based on Dual Channel Multi-scale Deep convolution Multi-scale Deep Long Short-Term Memory network[C]//2022 Global Reliability and Prognostics and Health Management(PHM-Yantai).IEEE,2022:1-8.
[20]TESAR M,MÜLLER-BROICH S,HILMER L,et al.Predicting traction energy demand of a light rail vehicle using deep learning methods[C]//2021 IEEE Vehicle Power and Propulsion Confe-rence(VPPC).IEEE,2021:1-5.
[21]HA J B,ZHOU W P.Research on Energy Consumption Prediction of Urban Rail Transit Based on Data Mining[C]//7th International Conference on Information Science,Computer Technology and Transportation(ISCTT 2022).VDE,2022:1-4.
[22]YUAN H,CHEN H,LIU S W,et al.A deep convolutional neural network for detection of rail surface defect[C]//2019 IEEE Vehicle Power and Propulsion Conference(VPPC).IEEE,2019:1-4.
[23]LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection[C]//Proceedings of the IEEE International Conference on Computer Vision.2017:2980-2988.
[24]ZHANG H,XU J,ZHANG X W.Pedestrian tracking andcounting based on YOLOv4 and DeepSORT in Subway Stations[C]//2023 35th Chinese Control and Decision Conference(CCDC).IEEE,2023:732-737.
[25]ZHANG Z H,MENG W W,LI Z,et al.A Small-Scale Pedes-trian Detection Method in Railway Scene[C]//2022 IEEE 5th International Conference on Electronics and Communication Engineering(ICECE).IEEE,2022:188-192.
[26]LI Q C,ZHAO J Y.An Intrusion Detection Method for CBTC Systems Using Blockchain and LSTM[C]//2023 3rd Asia-Paci-fic Conference on Communications Technology and Computer Science(ACCTCS).IEEE,2023:609-612.
[27]GAN Y F,LU N Y,ZHANG B L,et al.A CNN-BiGRU Based Life Prediction Method for Rolling Pins of Rail Vehicle Door System[C]//2023 6th International Symposium on Autonomous Systems(ISAS).IEEE,2023:1-5.
[28]CHEN S,WU X T,ZHOU M,et al.Train delay predictionbased on a multimodal deep-learning method[C]//2021 China Automation Congress(CAC).IEEE,2021:3241-3246.
[29]BOATENG V A,YANG B.A Global Modeling Pruning Ensemble Stacking With Deep Learning and Neural Network Meta-Learner for Passenger Train Delay Prediction[J].IEEE Access,2023,11:62605-62615.
[30]YANG G,ZHANG F,GONG C,et al.Application of a deep deterministic policy gradient algorithm for energy-aimed timetable rescheduling problem[J].Energies,2019,12(18):3461.
[31]LIAO J L,YANG G,ZHANG S W,et al.A deep reinforcement learning approach for the energy-aimed train timetable resche-duling problem under disturbances[J].IEEE Transactions on Transportation Electrification,2021,7(4):3096-3109.
[32]NING L B,LI Y D,ZHOU M,et al.A deep reinforcement lear-ning approach to high-speed train timetable rescheduling under disturbances[C]//2019 IEEE Intelligent Transportation Systems Conference(ITSC).IEEE,2019:3469-3474.
[33]OBARA M,KASHIYAMA T,SEKIMOTO Y.Deep reinforcement learning approach for train rescheduling utilizing graph theory[C]//2018 IEEE International Conference on Big Data(Big Data).IEEE,2018:4525-4533.
[34]WU W,YIN J T,PU F,et al.A deep reinforcement learning approach for the traffic management of high-speed railways[C]//2021 IEEE International Intelligent Transportation Systems Conference(ITSC).IEEE,2021:2368-2373.
[35]LI G N,OR S W,CHAN K W.Intelligent energy-efficient train trajectory optimization approach based on supervised reinforcement learning for urban rail transits[J].IEEE Access,2023,11:31508-31521.
[36]WEI W B,WANG X M.Trajectory Optimization for Urban RailTransit Considering Regenerative Energy Utilization:A Reinforcement Learning Approach[C]//2022 China Automation Congress(CAC).IEEE,2022:300-305.
[37]NING L B,ZHOU M,HOU Z P,et al.Deep deterministic policy gradient for high-speed train trajectory optimization[J].IEEE Transactions on Intelligent Transportation Systems,2021,23(8):11562-11574.
[38]LI Z H,ZHU L,LI Y,et al.A deep reinforcement learningbased resource allocation method for urban rail transit cloud systems[C]//2021 IEEE International Intelligent Transportation Systems Conference(ITSC).IEEE,2021:3922-3926.
[39]ZHAO J H,ZHANG Y,NIE Y W,et al.Intelligent resource allocation for train-to-train communication:A multi-agent deep reinforcement learning approach[J].IEEE Access,2020,8:8032-8040.
[40]YU J M,WU S C,LIANG L,et al.Resource Allocation in Vehicular Networks Based on Federated Multi-Agent Reinforcement Learning[C]//2023 IEEE 23rd International Conference on Communication Technology(ICCT).IEEE,2023:84-89.
[41]SUN Y G,XU J Q,CHEN C,et al.A deep reinforcement lear-ning based control approach for suspension systems of maglev trains[C]//2020 International Conference on Sensing,Measurement & Data Analytics in the era of Artificial Intelligence(ICSMD).IEEE,2020:496-501.
[42]WANG C,LIU W T,TIAN Q H,et al.An energy-efficient train control approach based on deep Q-network methodology[C]//2020 IEEE 23rd International Conference on Intelligent Transportation Systems(ITSC).IEEE,2020:1-6.
[43]LIU W T,SU S,TANG T.An intelligent control approach for heavy haul trains using deep reinforcement learning[C]//2021 IEEE 17th International Conference on Automation Science and Engineering(CASE).IEEE,2021:2262-2267.
[44]HOU Z W,BAO F,DU Z X,et al.Research on operation control model of FAO system under compound-fault scene in urban rail transit[J].Systems Science & Control Engineering,2018,6(2):32-36.
[45]HE C C,YAN F.Research on Fault Prediction of Fully Auto-matic Operation System Based on Ontology and Bayesian Network[C]//2022 34th Chinese Control and Decision Conference(CCDC).IEEE,2022:1434-1439.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed