Internet of Things Platform and Services for Connected Cars
Chungki Woo
1,2
, Jihyun Jung
2,3
, Jang Euitack
2
, Jongwoong Lee
2
, Jaewook Kwon
1
and Daeyoung Kim
2,4
1
Samsung Electronics, Suwon, Kyungki, Korea, Republic of
2
Graduate School of Software, KAIST, Seoul, Korea, Republic of
3
Mando, Pankyo, Kyungki, Korea, Republic of
4
Auto-ID Lab, KAIST, Daejeon, Chungbuk, Korea, Republic of
Keywords: Connected Car, IOT, GS1, EPCglobal, Oliot, EPCIS, ONS.
Abstract: In recent years, the connected car market has been expanding. Various car manufacturers are trying to
provide Internet of things (IoT) services by collecting and analysing sensing data from cars. However, there
is not a well-defined standardized IoT platform to handle the big data for the various car OEM companies or
service providers. To resolve this issue, we propose a globally standardized IoT platform for connected cars
based on Global Standard 1 (GS1). We extended and remodelled Electronic Product Code global
(EPCglobal), one of GS1 standards, and developed a new IoT platform framework called open-language for
IoT (Oliot). Then, based on the framework, we modelled car events and developed some hardware and
software modules to capture, store, and share the event data. We also implemented demonstration services
using the shared data for verification. This research can provide a new ecosystem to the connected car
industries and service providers to enable standardized handling and processing of big data. As a result, it
will be much easier to create and provide a greater variety of services and combinations of services.
1 INTRODUCTION
The global connected car market is constantly
expanding. One study has predicted (Allied Market
Research, 2014) that the market will grow to $141
billion by the year 2020.
A ‘connected car’ usually means a car that
supports connection and communication between the
car and the outside world (Internet, mobile devices,
other cars, drivers and other things) using wireless
network technologies.
With growth of the market, the number of related
IoT services is also increasing. Such as BMW and
Volkswagen, are trying to collect and analyse the
sensing data from cars and provide various services.
BMW provides a car-as-a-sensor (CARASSO)
service and dynamically updated map information to
drivers. (Investor's Business Daily, 2015)
Volkswagen informs drivers of real-time traffic
conditions and car status through Car-Net.
(Engadget, 2015) HP has also experimented with
such a service by sensing driver behaviour, road
quality, and social media in the World Record Race.
Through these efforts, manufacturers are
introducing new value into the car market.
In this situation, a problem is that a number of
service providers and car OEM companies are
building and using their own private data silos. In
other words, they are using heterogeneous platforms
and different protocols for collecting, processing,
storing, and sharing car data. Having such different
ways to handle data is a major obstacle to data
sharing and to making good services using the big
data from connected cars and various things from
other domains.
To address this issue, we developed new
common IoT platform for Connected Car based on
Global Standards 1 (GS1) (Global Standards 1).
For a long time, GS1 was a global common
business language that has been used for distribution
business. The GS1 provides a few standards having
three following abilities. Identify, Capture and
Share. These three abilities enable industries to
uniquely identify object and capture and share the
event in life-cycle of object. We applied this concept
into new IoT platform for connected car. In the
conclusion, we can expect that generated event data
in the life cycles of cars can be standardized,
collected, and shared easily.
Woo, C., Jung, J., Euitack, J., Lee, J., Kwon, J. and Kim, D.
Internet of Things Platform and Services for Connected Cars.
DOI: 10.5220/0005952904690478
In Proceedings of the International Conference on Internet of Things and Big Data (IoTBD 2016), pages 469-478
ISBN: 978-989-758-183-0
Copyright
c
2016 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
469
To make the platform, we modelled four car
events, namely, 'Selling', 'Driving', ‘Repairing’ and
'Scrapping', and we constructed a server following
concept of GS1 to store and share the event data.
After that, we developed a special hardware module
for capturing the events and implemented a
capturing application running on the hardware. In
addition, we developed the demonstration services
called the car data \management system (CDMS)
and car data monitoring in real-time mobile
application service (CDMR) for the purpose of
verification.
In this paper, we explain how to the connected
car platform was built and how the services were
designed in detail.
The rest of this paper is organized as follows. In
Section 2, Oliot IoT platform framework including a
detailed description of GS1. Section 3 describes the
design of the connected car platform in detail.
Section 4 describes the development of services.
Finally, Section 5 concludes this paper.
2 GS1 & OLIOT
2.1 GS1: Global Standards 1
Global Standards 1 (GS1) supports the identification
of objects, capturing and sharing of data for business
related to products, services, assets, shipments,
physical locations, and so on. GS1 standards have
three important abilities.
First, IDENTIFY is related to identification
using unique codes called GS1 identification keys.
Second, CAPTURE is related to data capture
which support bar code and radio frequency
identification (RFID) data carriers and specify
consistent interfaces. Software components that
connect the data carriers to business applications are
also included here.
Figure 1: GS1 EPCglobal Architecture (Traub, 2005).
Third, SHARE is related to information sharing
that supports standards for data, communication, and
discovery.
Figure 1 shows the architecture of GS1
EPCglobal and three steps (IDENTIFY, CAPTURE
and SHARE) from bottom to up.
GS1 EPCglobal (Traub, 2005) is one of the GS1
standards and for using RFID-based identification. It
also supports the sharing of information about the
status of products and global visibility of items
through EPCIS.
EPC Information Service (EPCIS) (GS1, 2007)
is for capturing and sharing events. This is enabled
by code systems, such as GS1 identification keys,
global product classification (GPC) and Electronics
Product Code(EPC). These universal identifiers
provide a unique identity for any physical object in
the world. The identification of objects is done by
reading RFID tags. The data from objects is filtered
and grouped by an event capturing application.
Finally, the EPCIS-level events are transferred to an
EPCIS server and stored and shared.
Object Name Server (ONS) (Mealling, 2004) is
for finding and sharing data which are related with
specific object.
GS1 Rail Vehicle example (GS1, 2014) shows
the importance and strength of GS1. The example
explains how to implement EPCIS for Rail Vehicle
visibility. The example shows that the event data
from a rail vehicle can be effectively captured and
stored into the EPCIS.
We thought that making a new standardized IoT
platform using GS1 would be very powerful, so we
extended and remodelled the GS1 standard as an IoT
platform. We finally developed an open-source IoT
platform framework called open-language for IoT
(OLIOT), and we constructed a connected car
platform based on it.
2.2 Oliot: Open Language for IoT
Figure 2: Oliot IOT platform architecuture (Oliot 2015).
RAIBS 2016 - Special Session on Recent Advancement in IoT, Big Data and Security
470
Oliot (Oliot, 2015) is an open-source project to
build an ID-based framework to identify, capture,
control, and share information about smart things.
The overall architecture of the Oliot IOT
platform is shown in Figure 2.
It was implemented by referencing the newest
standard of GS1. We tried to extend the GS1
concept to a standardized IoT platform framework.
As a result, Oliot extended and remodeled the basic
architecture of the GS1 EPCglobal standard.
The EPCIS of EPCglobal was extended and
handled as ‘Internet things global data repository’
called Oliot Object Name Service (Oliot-EPCIS)
(Byun and Kim 2015) complying with the GS1
standard EPCglobal and supporting various sensor
devices not only RFID.
Also, ONS of EPCglobal was extended as
‘Internet things service discovery’ called Oliot
Object Name Service (Oliot-ONS).
In Oliot, various sensors can be used, unlike GS1
EPCglobal, which can only use RFID tags. These
include passive tags, such as barcodes; active tags,
such as wireless IDs; as well as sensor and actuator
networks, such as Zigbee, LoWPAN, mobile phone,
BLE, and so on.
The event data from sensing devices can be
processed and delivered to the Oliot-EPCIS system
by sensor middleware and domain-specific capturing
applications which developed with domain
knowledge. A domain-specific accessing application
can access the information stored in the Oliot-EPCIS
and provide services to end-users. Finally, end-users
can find the supported services for specific things
using Oliot-ONS. The Oliot-ONS server has task of
providing a service list regarding the request of an
end-user with a thing identification key. Thus, the
end-user can receive a service list and get the
services.
Oliot makes it easy to construct IoT services.
Our connected car platform was also built based on
this platform framework. The collected event data
from a car using this platform can be shared and
used very easily for various service privoders.
3 CONSTRUCT PLATFORM
Figure 3 shows that simple overall platform
architecture based on Oliot framework and the data
stream from a car to Oliot-EPCIS. When event data
is generated from a car, the sensing hardware
module receives the event data. The data is
processed and sent to the Oliot-EPCIS server using
the HTTP POST protocol by the event capturing
application. Oliot-EPCIS stores the event data and
shares it. The following subsections explain each
component in detail.
Figure 3: Overall Architecture of the Proposed Platform.
3.1 Modelling Car Events
Modelling car events follows the format of Oliot-
EPCIS event type.
To describe an event formally, we describe it
with ‘WHEN’, ‘WHAT’, ‘WHERE’, ‘WHY’ of the
event. ‘WHEN’ means the timestamp at which the
event occurred. ‘WHAT’ means objects related to
the event. ‘WHERE’ means the location at which the
event happened. ‘WHY’ explains the business
context of the event.
What events can be generated for a car? We
imagined the following scenario.
“A car is sold in a retail shop and moved. The car is
driven by its owner and repaired when the car has a
physical problem. Finally, the car will be old and
scrapped after some time.”
Hence, we defined four events in the lifecycle of
a car, namely,
‘Selling’, ‘Driving’, ‘Repairing’, ‘Scrapping’
Table 1: Data for car events.
NO.
Event
Data
1
Selling
From location to location
2
Driving
Speed
3 Longitude
4 Latitude
5 Altitude
6 RPM
7 Brake ON/OFF
8 The degree of steering wheel
9 The degree of accelerator
10 Repairing IDs of repaired components
11
Scrapping
IDs of output components
Internet of Things Platform and Services for Connected Cars
471
We decided to sense all 11 types of data related
to these events. Table 1 shows the four events and
the data generated by each event.
Tables 2 to 5 in each show detailed descriptions
of the four events.
Table 2: TransactionEvent for car ‘Selling’ event.
EPCIS
event
Event type
TransactionEvent
Action
ADD
WHEN eventTime 2015-11-20 13:06:23.999 +09:00
WHAT epcList 4012345.077889.1111 (SGTIN)
WHERE
bizLocation 0614141.07346.1235 (SGLN)
geo
extension
37.4836,127.044 (GPS)
WHY
bizStep retail_selling (CBV)
dispostion sold (CBV)
Extension
VIN
number
KMHJF32J7TU123
456
sourceLis
t
urn:epc:id:sgln:4012
345.00001.0
destinatio
nList
urn:epc:id:sgln:4012
345.00002.0
Table 2 is the description of ‘Selling’ event for
the below example situation.
“A car (WHAT-epcList) was sold (WHY-disposition)
and moved from one location (WHY-Extenson-
sourceList) to another location (WHY-Extension-
destinationList) in the retail shop (WHERE-
bizLocation) at the moment (WHEN-eventTime).”
In the table, the type of event appears in the first
row in each table. Each event is classified as one of
the three event types following Oliot-EPCIS event
type.
‘ObjectEvent’,‘TransactionEvent’,‘TransformationE
vent’
In the case of a ‘Selling’ event, the event type is
‘ObjectEvent’.
The second row gives a description of the action.
Three actions were already defined in the EPC event
definition.
‘ADD’, ‘DELETE’, ‘OBSERVE’
The action of a ‘Selling’ event is ‘ADD’. We
also should note that SGLN (Serialized Global
Location Number) and SGTIN (Serialized Global
Trade Item Number) which are the ‘GS1 standard
identification keys (GTIN or GLN) + unique serial
number’ of the car are used to identify the car and its
location. This is charm point about using GS1
concept. We can use GS1 code system for
identification of things.
‘Extension’ is additionally defined here. This is
an extension of the basic EPCIS event and describes
extended data. The first vehicle information number
(VIN number) is always included to be used by
various applications. After that, in this event, the
location to and from which the sold car moved are
described in WHY-Extension-sourceList, WHY-
Extension-destList in the table.
One more important thing is that the ‘WHY-
bizStep’, ‘WHY-disposition’ are described by core
business vocabulary (CBV) (GS1, 2014).
The detailed meaning of event types, actions,
and how to describe events are officially given in the
GS1 EPCIS standard (GS1, 2007).
Table 3: ObjectEvent for car ‘Driving’ event.
EPCIS
event
Event type ObjectEvent
Action OBSERVE
WHEN eventTime 2015-11-20 13:06:23.999 +09:00
WHAT epcList 4012345.077889.1111 (SGTIN)
WHERE
bizLocation 0614141.07346.1235 (SGLN)
geo
extension
37.4836,127.044 (GPS)
WHY
bizStep
driving (User Defined
Vocabulary)
Extension
VIN number
KMHJF32J7TU
123456
Speed 2
Longitude
127.044
Latitude
37.4836
Altitude
45.3
RPM
785
Brake
ON/OFF
on
The degree of
steering
wheel
-540
The degree of
accelerator
0
Table 3 gives the description of an ObjectEvent
for ‘Driving’ and it explains the following example
event.
“A car (WHAT-epcList) is being driven (WHY-
bizStep) at the moment (WHEN-eventTime) and its
current speed, position, and car status are like this.
(WHY/Extension-Speed, Longitude, Latitude,
Altitude, RPM,Brake ON/OFF, The degree of
Steering Wheel, The degree of accelerator).”
We should note that the WHY-bizStep is not
described by CBV here. We defined a new word
(user-defined vocabulary)‘driving’ and used it.
RAIBS 2016 - Special Session on Recent Advancement in IoT, Big Data and Security
472
Table 4: TransactionEvent for car ‘Repairing’ event.
EPCIS
event
Event type TransactionEvent
Action ADD
WHEN eventTime 2015-11-20 13:06:23.999 +09:00
WHAT epcList 4012345.077889.1111 (SGTIN)
WHERE
bizLocation 0614141.07346.1235 (SGLN)
geo
extension
37.4836,127.044 (GPS)
WHY
bizStep repairing (CBV)
Extension
VIN
number
KMHJF32J7TU12
3456
repairedList
urn:epc:id:sgtin:4
012345.077889.21
urn:epc:id:sgtin:4
012345.077889.22
urn:epc:id:sgtin:4
012345.077889.23
Table 4 gives the description of a TransactionEvent
for ‘Repairing’, and it explains following example event.
“A car (WHAT-epcList) was repaired (WHY-bizStep)
at the moment WHEN-eventTime), and the repaired
components are these things (WHY/Extension-
repairedList). The location of the repair shop is here
(WHERE-bizlocation)”.
Table 5: Transformation for car ‘Scrapping’ event.
EPCIS
event
Event type TransformationEvent
Action DELETE
WHEN eventTime 2015-11-20 13:06:23.999 +09:00
WHAT
inputEPCLi
st
4012345.077889.1111 (SGTIN)
outputEPC
List
4012345.077889.18 (SGTIN)
4012345.077890.19 (SGTIN)
4012345.077891.20 (SGTIN)
WHERE
bizLocation 0614141.07346.1235 (SGLN)
geo
extension
37.4836,127.044 (GPS)
WHY
bizStep destroying (CBV)
Extension VIN number
KMHJF32J7TU
123456
Table 5 is the description of a
TransformationEvent for ‘Scrapping’, and it
explains following example event.
“In the junkyard (WHERE-bizLocation), a car
(WHAT-epcList) was scrapped and destroyed (WHY-
bizStep) at the 2moment (WHEN-eventTime), and
the output components are these things (Extention-
outputEPCList).”
We can model not only above four events but
any events using this description mechanism.
3.2 Sensing Car Events
After modelling and describing car events, we
considered how to obtain event data from a car. For
‘Driving’ events, we decided to sense the data using
a self-produced hardware module. It virtually
emulates a connected car on an ordinary car with
wireless communication. Here, we explain how the
hardware was designed to obtain event data.
Figure 4: OBD (II) Interface (Top Left), OBD Cable (Top
Right), Raspberrypi2 (Middle Left), NEO-6M GPS
Module (Middle Right), SKPANGS PICAN board (Down
Left), WIFI Module (Down Right).
Figure 5: Combined Hardware Module.
First of all, a car was needed that would support
an on-board diagnostics (OBD) interface. An OBD
interface is normally used to check the car status
which is data usually called car area network (CAN)
data from the electronic control unit (ECU) network
in the car. We could get four types of event data,
namely, RPM, brake on/off status, the degree of the
steering wheel, and the degree of the accelerator. To
get data from OBD interface, we needed a special
cable called an OBD cable. In addition, we used a
GPS sensor to get the speed, longitude, latitude, and
altitude of car. When a ‘Driving’ event happens
while the car is being driven, all eight events (No. 2~
No. 9 in Table 1) can be received from the OBD
interface and GPS sensor. All data is collected and
Internet of Things Platform and Services for Connected Cars
473
processed by an application running on a
Raspberrypi2. The Raspberrypi2 functions as the
central processing unit in the hardware composition.
Also, we used a PICAN board (PICAN CAN-Bus)
to receive data from the OBD interface. A PICAN
board is specialized developed add-on hardware for
the Raspberrypi2. Figure 5 shows the completed
hardware configuration for the project. Table 6
summarizes the arranged hardware and its purpose.
Table 6: Hardware list.
No. Hardware Purpose
1 Car Generate car events
2 Raspberry pi2
Run capturing application and
send formatted data to EPCIS
server using network
3 OBD2 Cable
Get car status data from OBD
interface of car
4
GPS sensor
module
Get GPS data
5 PICAN board
Get car status data through OBD
cable
6 WIFI module
Support wireless communication
of Rasberrypi2
7 WiFi Access point
Communication from the car to
the external Oliot-EPCIS server.
(Here, just mobile hotspot is
used.)
Figure 6 shows the hardware installed in the car.
Power for the hardware environment is supplied by
the car.
Figure 6: Combined hardware (TOP) , Regular Car - KIA
Sorrento (DOWN LEFT), Installed Hardware in the car
(DOWN RIGHT).
Figure 7 shows the overall data stream with
interfaces from the car to the Rasberrypi2.
Figure 7: Event data flow.
Finally, collected and processed data from the
Raspberry pi2 is sent to the Oliot-EPCIS server
presented in section 2.2. In this case, wireless
communication is used. A WiFi module and a WiFi
access point are necessary for that.
3.3 Building Oliot-EPCIS
The Oliot-EPCIS stores and shares the data. It
supports web service interfaces and RESTapi for the
capture and query services, so we can store the
sensed data in the EPCIS and share it using the
interfaces.
To store event data in the EPCIS server, the
schema of the database (mongoDB) should be
defined first. The basic schema for a basic event,
called an ‘EPCIS event’, was already pre-defined,
but if new extended data for a specific domain (such
as a connected car) should be stored, another
extended schema is necessary. The schema was
saved in the XML .XSD file. For example, the
schema for ‘Driving’ event is like below Figure 8.
All eight data and data types of ‘Driving’ event are
specified.
Figure 8: XML schema (.XSD) for car ‘Driving’ event
data.
RAIBS 2016 - Special Session on Recent Advancement in IoT, Big Data and Security
474
In case of ‘Selling’ and ‘Scrapping’ events, we
do not need an explicit XML schema because the
default schema of an EPCIS event already includes
the necessary schema.
3.4 Event Capturing Application
Modelling events, building the EPCIS server, and
defining the schema were necessary, but the most
important aspect of this work was to develop an event
capturing application to operate the hardware and
control the process of collecting and sharing the event
data from a car. The event capturing application
initializes the hardware module and collects the data
(RPM, break, the degree of accelerator, and the
degree of steering wheel), GPS data (latitude,
longitude, altitude, and speed). After that, it sends the
data to the EPCIS server after making an EPCIS
XML formatted document. In addition, it periodically
(once per second by default) generates virtual events,
such as ‘Selling’, ‘Repairing’, and ‘Scrapping’, and
also send this data to the EPCIS server.
Figure 9: Structure of capturing application.
Figure 9 shows the simple structure of the event
capturing application. To obtain GPS data, a modified
GPSD (GPS daemon) is used, while to obtain CAN data,
a modified CANDUMP utility is used. Both are open-
source, and they can be obtained easily from the website.
A ‘Sensing data parser’ is applied to the GPS and
CAN sensor data. An XML document sender sends
XML documents to the Oliot-EPCIS using the HTTP
Post protocol. The ‘libCurl’ is a popular open-source
library (Libcurl library, 2015) that provides the HTTP
protocol used to send XML documents to EPCIS.
Figure 10: Sequence of capturing application.
Figure 10 shows the processing sequence of the
capturing application.
The XML document is sent to the EPCIS server
using the HTTP/Post protocol by the event capturing
application.
It is important to note that the XML string must
be adapted to the XSD schema, which we defined in
section 3.3.
Figure 11: XML string to send car ‘Driving’ event data.
Figure 11 shows an example XML document for
a ‘Driving’ event. It means that the car is being
driven, and the car speed is 2 Km/hour, the car break
is on, the degree of accelerator is 0, and the degree
of the steering wheel is 30 to the left direction. The
current car position is latitude : 37.4836, longitude:
127.044, and altitude : 45.3 at the moment.
4 DEMONSTRATION SERVICES
In the prior sections, the design of the platform to
collect connected car event data and share it by
Oliot-EPCIS was explained. In this section, we
describe the implementation of demo applications
based on the platform for verification.
First, the web services for searching and
managing car data are explained. Secondly, we
present an Android app on a mobile device which
obtains car data in real-time.
Figure 12: Overall structure of services.
Internet of Things Platform and Services for Connected Cars
475
Figure 12 shows the overall structure of the
services. The services were implemented on a
mobile device and a Web service server.
Figure 13: Overall Service.
Figure 13 shows the demo application
implemented as a service layer, and others were
already mentioned above. A Web service and mobile
app query was sent to EPCIS using an Http Request
and an XML document was received as a response
as shown in Figure 14.
Figure 14: Service operation.
4.1 Car Data Management System
(CDMS) Web Services
Figure 15: Main intro of CDMS Web service.
CDMS (Figure 15) provides services to a user to
search and manage various car events (Table 1).
CDMS provides two searching services: the first is
searching ‘Driving’ history (Table 3); the second is
searching the history of a car (Table 2, 4, 5).
The ‘Driving’ history search service, provides
the driving history more specifically by simple web
service. The user inputs the related car ID
(SGTIN+VIN number) and period into CDMS, and
then a related query string is made. Using an Http
request, the query is executed, and its response is an
XML document as mentioned in section 3. The
XML document includes various data types: GPS
data, break sensor, and other data (Figure 8). The
XML document is parsed, and the result is displayed
on a webpage like Figure 16.
Figure 16: Speed variation (Up Left, Y-axis : speed
(Km/hour), X-axis : time), RPM variation (Up Right, Y-
axis : RPM, X-axis : time), Break sensor (Down Left, Y-
axis : break status - on : 1 otherwise 0, X-axis : time),
Steering wheel variation (Down Right, Y-axis : degree of
steering wheel, X-axis : time).
RPM, speed, break, and angle of steering wheel data
are shown as line graphs. Also, using the Google Maps
service, the car’s location is displayed like Figure 17.
Figure 17: Car location results (Google Maps).
The Web service is implemented using Active
Server Page Extension (ASPX).
Second, the car history search service provides
the specific history of the car after the user inputs
the related car ID. The user can find out when the
car was manufactured, which parts have been
replaced, and who has sold the car.
4.2 Car Data Monitoring in Real-time
(CDMR) Mobile Application
Service
CDMR is a mobile application (Android) that
RAIBS 2016 - Special Session on Recent Advancement in IoT, Big Data and Security
476
provides car data in real-time. CDMR queries the
Oliot-EPCIS server with a specific car ID every
second, and the latest car data is obtained and
displayed via a mobile application. The displayed
data is related to RPM, speed, location, break sensor,
and so on (Table 1). Thus, a user can monitor car
data and location in real-time. Also, a user can
monitor different cars by entering another car ID.
Figure 18 shows the display of real-time car status
on a mobile phone.
Figure 18: Real-time car position and status in the mobile.
4.3 Building Oliot -ONS
After designing these services, we constructed Oliot-
ONS for the end-user to find a list of services, such
as CDMS and CDMR of specific car, regardless of
time or place. Consumers or other service providers
can obtain the service list using only the GS1
identification key of a car. Service lists normally
include the address of the EPCIS server, the Web
service server, and so on. Figure 19 shows the
overall demonstration service architecture for
connected cars including Oliot-ONS.
Figure 19: Services with Oliot-ONS.
If someone wants to develop a service based on
car event data, they can access the event data shared
by the Oliot-EPCIS Server and make and deploy the
service, or a previously developed service based on
data shared by EPCIS can be used. Oliot-ONS
supports Web-based ONS record management.
Using ONS API, Web-enabled devices can register,
delete, and edit information of main databases. The
EPCIS server and a separate service server would be
registered in the ONS server in this case. In this
paper, this function was not implemented, but it
could be easily implemented and used.
5 CONCLUSIONS
In this paper, we proposed a new IoT platform for
connected cars based on GS1. We modelled car
events and produced a hardware sensing module. In
addition, we developed a capturing application to
capture and send event data to the Oliot-EPCIS.
Finally, the car event data is shared and used with
various end-users and service providers. For
verification of the feasibility of the system, we
implemented CDMS and CDMR services for
demonstration.
Any car manufacturers can use this concept and
build their platforms to process and share their data
through this system, which is based on a globally
standardized GS1 mechanism. Through this, all data
from connected cars can be shared easily, and
service providers can design and deploy IoT services
easily. This work represents an innovative new
means of processing big data from connected cars.
We can easily eliminate the silos and share the data.
Finally, this will be a major breakthrough in big data
processing for connected cars.
REFERENCES
Allied Market Research, 2014. "Connected Car Market -
Opportunities and Forecasts, 2013 – 2020".
Byun, J., & Kim, D., 2015. Oliot EPCIS: New EPC
information service and challenges towards the
Internet of Things. In RFID (RFID), 2015 IEEE
International Conference on (pp. 70-77).
Engadget, 2015. “Keep tabs on your Volkswagen with the
Apple Watch".
Global Standards 1, Available from : http://www.gs1.org.
GS1, 2014 Core Business Vocabulary (CBV) version 1.1.
GS1, 2015. “EPCIS for Rail Vehicle Visibility
Application Standard Release 1.0.1, Ratified"
available from : http://www.gs1.org/epcis/rail/latest.
Internet of Things Platform and Services for Connected Cars
477
GS1., How GS1 standards work from
http://www.gs1.org/how-gs1-standards-work.
Investor's Business Daily, 2015. “Amazon: Cloud Will
Make Internet of Things BetterWeb Service Guides IoT
Data New 'AWS IoT' can power apps for online
devices".
Libcurl library, 2015. Available from http://curl.haxx.se/li
bcurl/
Oliot, 2015. Open Language for Internet of Things.
Available from : http://oliot.org.
PICAN CAN-Bus Board for Raspberry Pi. Available from
http://skpang.co.uk/catalog/pican-canbus-board-for-ras
pberry-pi-p-1196.html.
GS1, 2007. EPC Information Services (EPCIS) Version
1.1 Specification.
Mealling, M., 2004. EPCglobal Object Name Service
(ONS) 1.0. EPCglobal Working Draft.
Traub, K., Allgair, G., Barthel, H., Burstein, L., Garrett, J.,
Hogan, B., ... & Stewart, R. (2005). The EPCglobal
architecture framework. EPCglobal Ratified
specification.
RAIBS 2016 - Special Session on Recent Advancement in IoT, Big Data and Security
478