Machine Learning and Intelligent Communications. Third International Conference, MLICOM 2018, Hangzhou, China, July 6-8, 2018, Proceedings

Research Article

Pedestrian Walking Model for Floor Plan Building Based on Crowdsourcing PDR Data

Download
135 downloads
  • @INPROCEEDINGS{10.1007/978-3-030-00557-3_31,
        author={Guangda Yang and Yongliang Zhang and Lin Ma and Leqi Tang},
        title={Pedestrian Walking Model for Floor Plan Building Based on Crowdsourcing PDR Data},
        proceedings={Machine Learning and Intelligent Communications. Third International Conference, MLICOM 2018, Hangzhou, China, July 6-8, 2018, Proceedings},
        proceedings_a={MLICOM},
        year={2018},
        month={10},
        keywords={Floor plan Mobile crowdsourcing IMU PDR},
        doi={10.1007/978-3-030-00557-3_31}
    }
    
  • Guangda Yang
    Yongliang Zhang
    Lin Ma
    Leqi Tang
    Year: 2018
    Pedestrian Walking Model for Floor Plan Building Based on Crowdsourcing PDR Data
    MLICOM
    Springer
    DOI: 10.1007/978-3-030-00557-3_31
Guangda Yang1, Yongliang Zhang2, Lin Ma2,*, Leqi Tang2
  • 1: Mobile Communications Group Heilongjiang Co., Ltd.
  • 2: Harbin Institute of Technology
*Contact email: malin@hit.edu.cn

Abstract

Indoor navigation has gained lots of interest in the last few years due to its broad application prospect. However, indoor floor plan for position display is not always available. In this paper, we utilize the crowdsourcing pedestrian dead reckoning (PDR) data got from the smart phone to build the indoor floor plan. According to the crowdsourcing PDR data, we propose new walking model that reflects the distribution of indoor pedestrian trajectory. This model is can well express the pedestrian walking pattern. In addition, the proposed model can also estimate the hallway width through the PDR data in hallway. According to the proposed model, we can draw the floor plan with the width of hallway. We have implemented the proposed algorithm in our lab and evaluated its performances. The simulation results showed that the proposed algorithm can efficiently generate the floor plan in the unknown environments with lower cost, which can contribute a lot for indoor navigation.