1662 IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 18, NO. 7, JULY 2017
A Survey on Approaches of Motion Mode
Recognition Using Sensors
Mostafa Elhoushi, Member, IEEE, Jacques Georgy, Member, IEEE,
Aboelmagd Noureldin, Senior Member, IEEE , and Michael J. Korenberg
Abstract—Recognition of the mode of motion or mode of
transit of the user or platform carrying a device is needed
in portable navigation, as well as other technological domains.
An extensive survey on motion mode recognition approaches is
provided in this survey paper. The survey compares and describes
motion mode recognition approaches from different viewpoints:
usability and convenience, types of devices in terms of setup
mounting and data acquisition, various types of sensors used,
signal processing methods employed, features extracted, and
classification techniques. This paper ends with a quantitative
comparison of the performance of motion mode recognition
modules developed by researchers in different domains.
Index Terms— Activity recognition, motion detection, machine
learning, navigation, sensor fusion.
I. I NTRODUCTION
T
HE need to detect the mode of motion of a person, i.e.,
knowing whether he is walking, running or in an auto-
mobile, bus, train, etc., is required in several fields including
navigation. For a portable navigation device to be robust and
flexible in any mode of transit or mode of motion, in any
environment, and with low-cost sensors, motion mode recog-
nition is a necessity. This is because different motion modes
require different navigation algorithms and constraints in order
to obtain an accurate positioning solution [1]. For example, if
walking is detected, then pedestrian dead reckoning (PDR) is
used [2], and if cycling is detected then cycling dead reckoning
is used [3]. Another field where motion mode recognition is
required is remote activity monitoring of medical patients [4].
Manuscript received February 28, 2015; revised August 18, 2015,
December 24, 2015, and September 8, 2016; accepted October 8, 2016.
Date of publication October 31, 2016; date of current version June 26, 2017.
This work was supported by the Natural Sciences and Engineering Research
Council of Canada, Collaborative Research and Development Grant through
Trusted Positioning Inc., now a fully owned subsidiary of Invensense Inc. The
Associate Editor for this paper was A. Amditis.
M. Elhoushi is with InvenSense Inc., Calgary, AB T3B 4M1, Canada, and
also with the Department of Electrical and Computer Engineering, Queen’s
University, Kingston, ON K7L 3N6, Canada (e-mail: m.elhoushi@ieee.org).
J. Georgy is with InvenSense Inc., Calgary, AB T3B 4M1, Canada (e-mail:
jgeorgy@invensense.com).
A. Noureldin is with the Department of Electrical and Computer
Engineering, Royal Military College of Canada, Kingston, ON K7K 7B4,
Canada, and also with the Department of Electrical and Computer
Engineering, Queen’s University, Kingston, ON K7L 3N6, Canada (e-mail:
aboelmagd.noureldin@rmc.ca).
M. J. Korenberg is with the Department of Electrical and Computer
Engineering, Queen’s University, Kingston, ON K7L 3N6, Canada (e-mail:
korenber@queensu.ca).
Color versions of one or more of the figures in this paper are available
online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TITS.2016.2617200
Fig. 1. Overview of motion mode recognition process, showing each of its
steps and the section or sub-section covering each step in the paper.
In literature, motion mode recognition can be vision-based
or sensor-based [5]. Vision-based motion mode recognition
has been an active research topic for a long time in different
fields, such as robot learning and surveillance [6]. Interest in
sensor-based motion mode recognition has increased recently,
possibly due to the emergence of low-cost and miniature
motion sensors, as well as the increased popularity of pervasive
and ubiquitous computing. This survey shall concentrate solely
on sensor-based motion mode recognition.
Furthermore, sensor-based motion mode recognition may be
categorized into recognition using on-body sensors or using
ambient sensors. On-body sensors can be a portable device
carried by the user or custom sensors or devices attached
to one or more parts on the body of the user. Alternatively,
ambient sensors, can be placed at certain parts in a room or
building to detect the presence or motion of a user [7], [8], or
on objects to detect the interaction of a user with them [9]).
The focus of this survey is on-body sensors.
Fig. 1 shows an overview of the steps of motion mode
recognition that are followed by the majority of researchers
in literature [10]. The first step is data acquisition where
1524-9050 © 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.