An activity history based approach for recognizing the mode of transportation

By understanding a person's context, smart assistant systems aim to simplify and automate tasks. One type of useful context information is the mode of transportation. Commonly, the mode of transportation is recognized by analyzing raw sensor data of mobile devices, such as a smartphone. However, fine-grained sampling and collecting sensor data at high frequency to analyze raw sensor data is also conducted by activity recognition systems. These systems have been shown to accurately recognize activities such as sitting, standing, or walking. This means several systems analyze raw data for their needs. In our opinion, this doubled use of resources should be avoided. In this paper, we propose an approach, which avoids doubled raw sensor data analysis by taking advantage of the already available information of a person's activity. As a hypothesis, the approach assumes that observed activity sequences can lead to the mode of transportation. Thus, we aim to map activity sequences to modes of transportation. Applying the proposed approach, we show that the three different modes of transportation car, tram, and bus are recognized with an accuracy of up to 97 %.

[1]  Peter Widhalm,et al.  Transport mode detection with realistic Smartphone sensor data , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[2]  Luciano Bononi,et al.  By train or by car? Detecting the user's motion type through smartphone sensors data , 2012, 2012 IFIP Wireless Days.

[3]  Zhezhuang Xu,et al.  Learning Transportation Modes From Smartphone Sensors Based on Deep Neural Network , 2017, IEEE Sensors Journal.

[4]  Carlo Ratti,et al.  Transportation mode inference from anonymized and aggregated mobile phone call detail records , 2010, 13th International IEEE Conference on Intelligent Transportation Systems.

[5]  Klaus David,et al.  Implicit positioning using compass sensor data , 2016, UbiComp Adjunct.

[6]  Kenji Mase,et al.  Activity and Location Recognition Using Wearable Sensors , 2002, IEEE Pervasive Comput..

[7]  Muhammad Awais.  Use of acceleration data for transportation mode prediction , 2015 .

[8]  Xuegang Ban,et al.  Vehicle classification using GPS data , 2013 .

[9]  Deborah Estrin,et al.  Using mobile phones to determine transportation modes , 2010, TOSN.

[10]  Paul J. M. Havinga,et al.  Complex Human Activity Recognition Using Smartphone and Wrist-Worn Motion Sensors , 2016, Sensors.

[11]  Hjp Harry Timmermans,et al.  Transportation mode recognition using GPS and accelerometer data , 2013 .

[12]  Sasu Tarkoma,et al.  Accelerometer-based transportation mode detection on smartphones , 2013, SenSys '13.

[13]  Stephan Sigg,et al.  An Alignment Approach for Context Prediction Tasks in UbiComp Environments , 2010, IEEE Pervasive Computing.

[14]  Hirozumi Yamaguchi,et al.  Tracking motion context of railway passengers by fusion of low-power sensors in mobile devices , 2015, SEMWEB.

[15]  R. Waters,et al.  Energy‐speed relationship of walking: Standard tables , 1988, Journal of orthopaedic research : official publication of the Orthopaedic Research Society.

[16]  Nicholas Wickström,et al.  A Symbol-Based Approach to Gait Analysis From Acceleration Signals: Identification and Detection of Gait Events and a New Measure of Gait Symmetry , 2010, IEEE Transactions on Information Technology in Biomedicine.

[17]  Sungyong Lee,et al.  VehicleSense: A reliable sound-based transportation mode recognition system for smartphones , 2017, 2017 IEEE 18th International Symposium on A World of Wireless, Mobile and Multimedia Networks (WoWMoM).

[18]  Catherine T. Lawson,et al.  A GPS/GIS method for travel mode detection in New York City , 2012, Comput. Environ. Urban Syst..

[19]  Hesham A. Rakha,et al.  Applying Machine Learning Techniques to Transportation Mode Recognition Using Mobile Phone Sensor Data , 2015, IEEE Transactions on Intelligent Transportation Systems.