A multi-agent system for distributed smartphone sensing cycling in smart cities
| Pages | 119-134 |
| Date | 06 April 2020 |
| Published date | 06 April 2020 |
| DOI | https://doi.org/10.1108/JSIT-12-2018-0158 |
| Author | Theodoros Anagnostopoulos,Chu Luo,Jino Ramson,Klimis Ntalianis,Vassilis Kostakos,Christos Skourlas |
| Subject Matter | Information & knowledge management,Information systems,Information & communications technology |
A multi-agent system for
distributed smartphone sensing
cycling in smart cities
Theodoros Anagnostopoulos
Department of Business Administration, University of West Attica, Athens, Greece
Chu Luo
School of Computing and Information Systems, University of Melbourne,
Melbourne, Australia
Jino Ramson
School of Engineering Technology, Purdue University, West Lafayette,
Indiana, USA
Klimis Ntalianis
Department of Business Administration, University of West Attica, Athens, Greece
Vassilis Kostakos
School of Computing and Information Systems, University of Melbourne,
Melbourne, Australia, and
Christos Skourlas
Department of Informatics and Computer Engineering, University of West Attica,
Athens, Greece
Abstract
Purpose –The purpose of thispaper is to propose a distributed smartphone sensing-enabledsystem, which
assumes an intelligenttransport signaling (ITS) infrastructure thatoperates traffic lights in a smart city (SC).
The system is able to handle priorities between groups of cyclists (crowd-cycling) and traffic when
approachingtraffic lights at road junctions.
Design/methodology/approach –The system takes into consideration normal probability density
function (PDF) and analyticscomputed for a certain group of cyclists (i.e. crowd-cycling).An inference model
is built based on real-time spatiotemporal data of the cyclists. As the system is highly distributed –both
physically (i.e. location of the cyclists) and logically (i.e. different threads), the problem is treatedunder the
umbrella of multi-agent systems (MAS) modeling. The proposed model is experimentally evaluated by
incorporatinga real GPS trace data set from the SC of Melbourne, Australia. The MAS model is applied to the
data set accordingto the quantitative and qualitative criteria adopted.Cyclists’satisfaction (CS) is defined as
a function, which measures the satisfactionof the cyclists. This is the case where the cyclists wait the least
amount of time at traffic lightsand move as fast as they can toward their destination. ITS systemsatisfaction
(SS) is defined as a function that measures the satisfaction of the ITS system. This is the case where the
system serves the maximum number of cyclists with the fewest transitions between the lights. Smart city
satisfaction (SCS) is defined as a function that measures the overall satisfaction of the cyclists and the ITS
system in the SC based on CS and SS. SCS defines threeSC policies (SCP), namely, CS is maximum and SS is
minimum then the SC is cyclist-friendly(SCP1), CS is average and SS is average then the SC is equally cyclist
and ITS system friendly (SCP2) and CS is minimum and SS is maximum then the SC is ITS system friendly
(SCP3).
Distributed
smartphone
sensing
cycling
119
Received3 December 2018
Revised22 February 2019
4 October2019
Accepted24 February 2020
Journalof Systems and
InformationTechnology
Vol.22 No. 1, 2020
pp. 119-134
© Emerald Publishing Limited
1328-7265
DOI 10.1108/JSIT-12-2018-0158
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/1328-7265.htm
Findings –Results are promisingtoward the integration of the proposed system with contemporarySCs, as
the stakeholders are able to choose between the proposed SCPs according to the SC infrastructure. More
specifically, cyclist-friendlySCs can adopt SCP1, SCs that treat cyclistsand ITS equally can adopt SCP2 and
ITS friendlySCs can adopt SCP3.
Originality/value –The proposed approach uses internetconnectivity available in modern smartphones,
which provide users controlover the data they provide to us, to obviate the installation of additional sensing
infrastructure. It extends related study by assuming an ITS system, which turns traffic lights green by
consideringthe normal PDF and the analytics computed for a certain group of cyclists. The inferencemodel is
built based on the real-time spatiotemporal data of the cyclists. As the system is highly distributed –both
physically (i.e. location of the cyclists) and logically (i.e. different threads), the system is treated under the
umbrella of MAS.MAS has been used in the literature to model complex systemsby incorporating intelligent
agents. In this study, the authors treat agents as proxy threads running in the cloud, as they require high
computationpower not available to smartphones.
Keywords Smart city, Crowd-cycling, Distributed smartphone sensing,
Intelligent Transport signaling system, Multi-agent system
Paper type Research paper
1. Introduction
In our previous work (Anagnostopoulos et al., 2016), we described a system that senses
cyclists and prioritizes them. Such a system, which favors cyclists, can potentially
improve enjoyment and reduce accidents because of crossing a red light (B-icycle,
2015), thereby encouraging people to cycle more (Miller, 2013). Several technologies for
sensing cyclists exist such as lane counters (Pai and Jou, 2014), cameras (Tan et al.,
2008) and road radars (Cyclemeter, 2018), but these alternatives incur costs. A more
economical approach is to re-program all traffic lights to prioritize cyclists by default
(Cycle-tracker, 2013). Such an approach, however, is time-consuming and works well
only with sporadic vehicular traffic. We proposed sensing cyclists using their
smartphones, instead of specialized hardware currently used onboard safety vehicles.
Evidently, such an approach introduces a number of challenges related to power
efficiency, modeling and privacy in estimating the position, speed and direction of
cyclists. In another previous study (Anagnostopoulos et al., 2017), we enhanced
smartphone-sensing to enable efficient time-of-arrival (ToA) estimation for cyclists
moving toward traffic lights in a smart city (SC) –by using GPS sensors to locate the
actual position of cyclists on their way to the traffic lights. The proposed approach
tackles inefficient GPS energy consumption by using velocity to estimate ToA and aims
to enable efficientcyclinginSCsbyturningtraffic lights green proactively.
In this paper, we extend prior work by assuming an intelligent transportsignaling (ITS)
system that prioritizescyclists by considering the normal probability density function(PDF)
and the analytics computed for certain groups of cyclists (crowd-cycling). The inference
model is built based on real-timespatiotemporal data of the cyclists. As the system is highly
distributed –both physically(i.e. location of the cyclists) and logically (i.e. different threads),
we propose to treat the system under the umbrella of multi-agent systems (MAS) modeling
(Laghari and Niazi, 2016). The MAS approach has been used in the literature to model
complex systems by incorporatingintelligent agents. In this study, we treat agents as proxy
threads running in the cloud, as they consume high computational power not available to
smartphones. The systemis comprised of the following agents:
traffic light agent (TLA);
single cycling agent (SCA); and
group cycling agent (GCA).
JSIT
22,1
120
To continue reading
Request your trial