Proceedings of the 13th EAI International Conference on Mobile Multimedia Communications, Mobimedia 2020, 27-28 August 2020, Cyberspace

Research Article

An Uncertain Trajectory Modelling Method Based on Kernel Density Estimation

Download407 downloads
  • @INPROCEEDINGS{10.4108/eai.27-8-2020.2296731,
        author={Yuan  Cheng and Ronghua  Chi and Yahong  Wang},
        title={An Uncertain Trajectory Modelling Method Based on Kernel Density Estimation},
        proceedings={Proceedings of the 13th EAI International Conference on Mobile Multimedia Communications, Mobimedia 2020, 27-28 August 2020, Cyberspace},
        publisher={EAI},
        proceedings_a={MOBIMEDIA},
        year={2020},
        month={11},
        keywords={uncertainties kernel density estimation modelling method distribution characteristics},
        doi={10.4108/eai.27-8-2020.2296731}
    }
    
  • Yuan Cheng
    Ronghua Chi
    Yahong Wang
    Year: 2020
    An Uncertain Trajectory Modelling Method Based on Kernel Density Estimation
    MOBIMEDIA
    EAI
    DOI: 10.4108/eai.27-8-2020.2296731
Yuan Cheng1,*, Ronghua Chi2, Yahong Wang3
  • 1: Harbin University of Science and Technology
  • 2: Heilongjiang University of Science and Technology
  • 3: Beijing Institue of Technology, Zhuhai
*Contact email: changuang7@sina.com

Abstract

The accurate analysis of trajectories is of great significance for route selection, traffic status analysis, and urban traffic planning and so on. Existing researches lack effective methods for dealing with possible uncertainties in trajectories caused by objective enviroment and subjective intention etc. This work studies the method of constructing an uncertain model for the trajectories with the same starting point and end point based on kernel density estimation, to discover the distribution characteristics of the trajectories between two points in historical data, and to lay the foundation for trajectory prediction. Finally, the validity of the proposed method is verified on the real trajectory dataset.