About | Contact Us | Register | Login
ProceedingsSeriesJournalsSearchEAI
Communications and Networking. 17th EAI International Conference, Chinacom 2022, Virtual Event, November 19-20, 2022, Proceedings

Research Article

Research on Crowd Movement Trajectory Prediction Method Based on Deep Learning

Cite
BibTeX Plain Text
  • @INPROCEEDINGS{10.1007/978-3-031-34790-0_30,
        author={Ruikun Wang and Xinyu Gu and Dongliang Li},
        title={Research on Crowd Movement Trajectory Prediction Method Based on Deep Learning},
        proceedings={Communications and Networking. 17th EAI International Conference, Chinacom 2022, Virtual Event, November 19-20, 2022, Proceedings},
        proceedings_a={CHINACOM},
        year={2023},
        month={6},
        keywords={Trajectory prediction Deep learning Transformer Neural networks},
        doi={10.1007/978-3-031-34790-0_30}
    }
    
  • Ruikun Wang
    Xinyu Gu
    Dongliang Li
    Year: 2023
    Research on Crowd Movement Trajectory Prediction Method Based on Deep Learning
    CHINACOM
    Springer
    DOI: 10.1007/978-3-031-34790-0_30
Ruikun Wang1, Xinyu Gu1,*, Dongliang Li1
  • 1: Beijing University of Post and Telecommunication, 10 Xitucheng Road
*Contact email: guxinyu@bupt.edu.cn

Abstract

Since the 21st century, the vigorous development of the Internet has brought about the rapid rise of social platforms. People’s desire to share has been satisfied, and the information such as time and location shared by users has become the trajectory data of users. The analysis of these trajectory data is helpful to the study of crowd behavior, among which the prediction of crowd movement trajectory is an important content of trajectory data analysis.

This paper investigates the Transformer model which has an excellent performance in the field of Natural Language Processing (NLP). According to the characteristics of the data adopted in this paper, a prediction method of crowd movement trajectory based on the Transformer model is proposed and the future trajectory prediction is realized. Markov model, Long Short Term Memory Network (LSTM), and Gated recurrent unit network (GRU) are selected as baseline methods to compare with the model in this paper. The final results show that the prediction method proposed in this paper performs well in the dataset of this paper. The result also shows that the prediction methods based on deep learning have higher accuracy in predicting the future movement trajectory of the crowd compared with the prediction scheme based on the traditional model, even other parameters.

Keywords
Trajectory prediction Deep learning Transformer Neural networks
Published
2023-06-10
Appears in
SpringerLink
http://dx.doi.org/10.1007/978-3-031-34790-0_30
Copyright © 2022–2025 ICST
EBSCOProQuestDBLPDOAJPortico
EAI Logo

About EAI

  • Who We Are
  • Leadership
  • Research Areas
  • Partners
  • Media Center

Community

  • Membership
  • Conference
  • Recognition
  • Sponsor Us

Publish with EAI

  • Publishing
  • Journals
  • Proceedings
  • Books
  • EUDL