11th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services

Research Article

ClimbTheWorld: Real-time stairstep counting to increase physical activity

Download747 downloads
  • @INPROCEEDINGS{10.4108/icst.mobiquitous.2014.258013,
        author={Fabio Aiolli and Matteo Ciman and Michele Donini and Ombretta Gaggi},
        title={ClimbTheWorld: Real-time stairstep counting to increase physical activity},
        proceedings={11th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services},
        publisher={ICST},
        proceedings_a={MOBIQUITOUS},
        year={2014},
        month={11},
        keywords={activity recognition energy consumption mobile computing ubiquitous applications},
        doi={10.4108/icst.mobiquitous.2014.258013}
    }
    
  • Fabio Aiolli
    Matteo Ciman
    Michele Donini
    Ombretta Gaggi
    Year: 2014
    ClimbTheWorld: Real-time stairstep counting to increase physical activity
    MOBIQUITOUS
    ICST
    DOI: 10.4108/icst.mobiquitous.2014.258013
Fabio Aiolli1, Matteo Ciman1,*, Michele Donini1, Ombretta Gaggi1
  • 1: Department of Mathematics, University of Padua, Italy
*Contact email: mciman@math.unipd.it

Abstract

The increasing number of people that are overweight due to a sedentary life requires persuasive strategies to convince people to change their behaviors. In this paper, we present a machine learning based technique to recognize and count stairsteps when a person climbs or descends stairs. This technique has been used as part of ClimbTheWorld, a real-time smartphone application that aims at persuading people to use stairs instead of elevators or escalators, since an engaging activity has more chance to change people's life habits. We perform a fine-grained analysis by exploiting smartphone sensors to recognize single stairsteps. Data-dependent sliding windows are used facilitating the learning process and reducing the computational cost. Finally, energy consumption is widely investigated to optimize the trade-off between classification precision and battery usage, to avoid exhausting smartphone battery.